EgE=Tg: EEei!:r :::€=€EYtr: tE gt9 zz it Etilz g1!i;1EE:1+ ;;lii; 1E5E!T=z g€iiiti1tlil*=E iisi:;;sE; ii i?i i' li E;z gillg;iiigEsl i;g;EjEi,"?'' IZ #i 7t it=z.itt';it i;EiE:eiiii;g ft.,i EE€EEiEiE ?rsE;, EEi!il: .iEEi ;ifei:i::t3;g i{:sE:: 3l;LEit i;;€;iiE;z iiilltz=.iiilll ?E *zE=EE.i ii+ E; 'EEl g : 'i=E t+ 'ulEl?ti -v fr F q) L I EZi ;'l Hi t€l: iE i; E;; g:EEE>: 6 ll|=li\llizz€zeilZtl a L F €- 'J=<Z'z=- -==-=== j t:== i=:::: Zz := 7,J., := Ez+ ,E F - ,E i= EEi: i=1E2 :. zE?E=, E=E?i,E E Z Y*TEit!= E:*E'ii)tt E+ :;iFs $ -!E i?. E : Z ; =7=:*= sAFE€ t,;=r'+=E,i: iEzl11ii=t, ;ii; Z i1=i =sz'=i:i, z=in'iEi +, ;; O. li! ?fiE z=|i iaiqf=riZZZEiZii :iiatlllAE !,a1 *3g Elltiel_5ig* rh.5 \-/+J -' 5 ;: E ! - =iJ- -=-^li -- vuev+ !- =i:i-=*= ?z+ *Ebt'EE:jHsa F ; '?EiiiiViE'*= EHE:EFE i d ,ltEEzzE €#:Z*ce -n>bzjrig.I5 ; F:il=1;zzi3i* :aE;EFEE E E ra3F=':\iZ:gI ?i_==ft?.. z si i :E gE:lE:;;qs*E i;?z$zE? :, !{:1EEFgZ*Ee lEEEizi€ E = 7 IE=g=.*E^EtE;i i{-€dbE: ? g i :gsys5 *=r?tl:i;ii; *Eri:s?itir{ ;E*;l:; = ; t tr F q) L r- - , a G L I .-a v It F ii;iE?=, EV;:yryiEiEEi IEEE;;7e y_zve1=ttsi}i :Et,E**i,iEsi E|E!g:;Ei a c rrl ii != = , ;' j =* : i 1|I=IE1;;,EEE =H Eai'a= =tt t E:l,EEil ;:s:'Fb==EsE i;l;E+; iEi 3,:+a!; EEEr!;;;-;1:,;E,==E=E*i.=i g;:iaiEiE:iii E;iii3si6!r Eitit=4 s i.=i=-i=ZEV:Z E*ii'yEigiE '#gtri;FEi"; il a! ' :€;e E??i:::?,;Igr+ ;iiEEEigE,ec; ?E E;i#li,Ei=Ei=i+ EEE;:EiEIi:iF; *i =i E:i fii tq€:i li?:*;€Ig5*:ia i:izEEE:,=zz='e =€a:i:rtEi;UA; €;a'E6i*+u?s!jE a! r iE:i riE! toaT= ii;'riii Efig!;: ; ;3!E;=r:EiEiiE =i!x siizEifE $E Ea E E ! ;: E 8E€;€u gEifir:-i 3sa; i:?',!?-=;:-'= , =.: H€€ q€*e ;; iEl i+;g 1* i;EEiJigggI I €;* :i[ iEEF; !;li Eii E;ltt=t $EEE;EHlsFgi gg .z IIrEI E iii!€i? 1z z1z +E s! i g IEi i:+ IiisE i;;rir+ii1 FI ss ; Er ; aE i gF*!;s SElBie i12 tl!?gi ri,; EE E:;gs$i*i .E!i;;i stiEEi,z:z 'c a LL. 2 ': -c , ri H FE€iF1 E I !?i,i ;i:; S-SggEi s-il?:itii.tz i?'i i :;€ : ! i: iii! zi;ielilEZ :EElze3;; ii,t='ri:;:i t;=?.=:= i g: i;=1;= : +iiiltljilI i:E=i+i,Ei1l;zia 5liEi::;:;i:!!::i=1 !ryE$;'.IliiillIuE,EE ; 3E;SEPi .a = EslE:;:'ii!!gi iii;: rilE;t!; iii!lig|iEigil|s 1= gs iiE Ii*Esii E:iiiiiFiliisiig: iigigl ii; X o boo tr E ? ' ! cJ! e'e !6!r o F: E i ts= € i €:H :E + i'! 3 ; 5;! *e- F ;Z'=E;-:ii ;i c; ve:5 E:-T;iEFEEEi ;*nFg Ia rE iE',Eg? e:irF Efi'gi: nEE'oz Et ji::E E= 'E2+7Ez=E E€5E.:+ - c ' = c_ =Z;;:: if,,=c/;--t- :EiaE ltl= i=1i7,=., 2Zr,r=?iz= g! ^ziii= a -> g.q -E :i= 1;: X*EJ'! EE|3:; E=: =:;:!€i-t:::;:l c O a !. o .: ! fr. L Irl Ha:A:?E!;;:;13: ;E: z7i 1eiE+I; E ?eA 0 r.Z - ;.= ; ,: = ; i ;.2 y .,,r gF E:g;EEi-laEEriI:a -i:r ii"? E 7Z=e!:E:E;::f ==El EErt I i2 ii-aE ;rA*9.q€9i.Ei 2:j a'E-2=:-=t cca;; ;ee; : g$gi$ il:I ;i;t iisiEiigiiiElgFilii,r::: Elrai Tllttii!{i$ is;sie :lft;;:g:g i!1 iiiiiiziv{z il=u ietl::ie5ll: l u'il; !: F I =;E Eii E lr iisiig g gi lii 3 EFt ;'€i$$ EIliIgliliig iis E[i:flE €il;:Ii 1; ig 'iZE, i?iEEii Eiiz !;;!;1; iE3gigllin!Ei:: . X tliiii$=;3zi;lFsrr i:EB{E:zft ilZ ggiIietg ,,rr :lEitltlEiE 1i::9l -o l!.1 a E a o a U .- Ea+i;:E| ?ii za+E*zE fiiil;iEil:zt s':i;zEr EEi?giesir ,elgg: E:Ei ?; g;i1: iii;EEi iiEiEgii, iltis€Bigg itill g llt iE ;illLltil l ei iZilE e i ?EE3 t= ! rzzite.:? I €E*a9r! o! 6 boQ= qil g .i i r ?= ?3. i t?Eise i;;:ii;i Eg:f *:;:Er"FirFi t .- +a E. l'vl €-q) !a *a a q) -A *a *) FI v L F q) L -t iEei;i!iElFg!i FIi !$E i!:ligi€ ::-:'!g' =iE:5i:: ;ecFeEEiiE= ;9Es; ii;i;ii;glFgieE i E*gliE E#lilE : *;i1lr! F;;ruig;;+5;gi;l : i€iEie *r;!*'E;;ssff *i*iiEiEig; li*gE ;;:i;;i: ilisiEE;;$i,i; rE?:=E i: a:E3.4-;'3;iF:€g;;i:sI ir .;Eg ;;i;tE g gi !;ilil E ;Fg; e ;: I i5 i€lFggE ! iiiFi i g; gs5 g€€ i sss;#:,i:i EEi[:5': II:*si *E I!' gE:! eEr Eg ii i:l! EI iIIIE ;:: igg€E E9gsilli,gF ffE! Eg;: z: txii E;*ttE iiil iiliEi€ii iggi rlE z= lgi lii:iigEEiEg€ is;gli=iiEEIEilE !Iilil;ffi !ggll'a,sgEigliiE #1-'i-iss grtiE 1 s€r:;€;I isEF: giE?li Eg tsa/ :!v-ts J?(JiJ>c: -'_E 'a'7 ? V - a v d - ! ! s i- 'F: -t se ! o t E € a"ln Y"= EO '"-6; >. €ha 3: ) c ( ) , = - ! Oaltso d x.Y ' - u : ! e >i i ^ v F F !/ c =;:.= Airvi 'U - . ,-O; E > ,€ a : V ) = u--vk oo @ >,H C0; I 'l':.^;q)- C qr - v ) a r - ^ F F ! ' - v C).= cr i =nOrd .:i:ltr € . - g H E iL.l a P 1+F: =vrv9v dg^i? ! - d k ^ - ^ : - v r v v . r J ! H - J=5 o;c,.! >.= ^i E u'-!rw ti;(!> a,)==pC) - : H ' F _ c O ': El E a ?, I rY)rv l--r9O! iligiiiigi$ii:giiiig:*ii,E atrilg:ig;g;lili iliiiggffgitll i;i X.eE 5 EE r i X €i 5i rs r€ E+,a?.'=Li!'l;i! gFrsEc€i iEz i€ 3:Eiisr; !;lll:igg yF,.; Etbi=E .eF9. 199 E4?!;; FEg€E;TE: ;'* #= E -e^- :; ilE i s ;e; iEE 3E! F7,7:€36EtFq;;X; 6' f!!i*€ ! ?VEEtr?iEE!;ig$E E ;2t;EZlt,?=7 EEE}3: .EIiE*.oe.66i-a l=E.!3. ;i=E:i €EE.q+=!=t?^ =itEE= -=i7Ev,=-iE?! # Er€+;l !g*;;iv::AE -a :='il;E= aEi?€HlsEgs i = zt E= E= 1i ; - a J J ! - 4 t t a \ '=?49'h ! trl o E s;;Er5=;I= 8_ yl 0'5+a-- 9-+-EE ;; +E+-i; :i 7,.= zE r$$i l: E!E l€€,5! ?E *=+? ! i tI i3 = pjiE=_i e€ i,=:c ; EE= gE u#i;;iiEE; *E i 1 ; Eei iij i t ^i i 4 I : $ ry € f Ft 5 $ i ! s I S: F;+ g: = ==e?,;E EEy,€?p i= t E:=?,7,8 a A€i $iE;ii; ;; E;=;;;r5; aaz! yZ:=i= =oe*F' Fc= itiSiTt=Z '' *t E=82! :,=ZtE=?=!zii=' -c ! u E rrl E o ! = 6 O ; o a tL ?i-- q) XJX ; 9'--i r-tr --9 E: E ?U 5a= () 50! or ar ii X :- B - r./ P> € :n E9 I >E B € fi.== cde cH X F c 9!lo - qr9r9 - o F -CE o g9J r-O Ou A a O !qJtrtr:O. y * = o-^ iFi6 E' ; - - a;: ^, q'.Y A o ! tr ",! jzi. a'= s =uil'l'== i -s .c n i I C- I =:3E?-,2 3Uc -..:t="'- -: -l 2 2 xi :1 .= ^,=- = \ S v =.=i -.,= =!r< '-:^ t- N .')c- .J '= - i- t = : - 7= = =- - . , . / = . : . ; - - -'i l : , E + =i =-3z= : :- -i - t -: .-=l-'t-'.'t - H= .- J=,t e 3 > :: N>: J<= : q d 9 . o = B-:: -17 .,?! r w - L -yPUk ail d , Y( S .= "=" E3-P !, O HH:- c! >.o: EH A} - 9 J P : 5.y es ;:g€s*;u-t; -c 0;5 aF c a= e !*- 3t E"Z=+fi !:E=;iui+ a7 egg iE E - -= a.:J B ?a 2 2r? a) =iE , , -. Z i=titii ,*iltrili:lt;T {lfl zla1ll? .H E{; ; E; E giicTalliEli {11i ii1 sii;i-E iz* g ;tE :;i !i;; #l1ili tE1Elltlv :;lltlli,iIE 3:g iig $ i; iillttll; llii?efi tEill*l iiE;ri ;:;aiiii;ii;sgr*Eggg'* ,llt!iillitz iii*:EeaE il Fgis:E tEssiii;EEEiIi: grE;€ E-s= O al O a] *.a -l -l \O OO C * c] ol .7) o t o i o N ca .7) ca r) ca =:i n ca 6 m.a Ff €,4 E tr e c 8 E c € -a- a id ata tr oV 2.?9 y.=.€=; -= - ^ i 2 = .l Z: ^. = - =i . E3 Sx .; = d. = : ;-.YEi.ia h'EEEiEdieY Eo;o?xZi :i5Z- YE4i | _ 'T Eo i = 6 ' J l , E {c;r V -.: 6Shoo-o .E ? -V < tc' E t - P- t.= ' J_ h a ? ^ i 2== i / ;, 1 7i -- 7 7u ?.= pETa =+E4t = =V:ti c.=-?z !;^-,?> - * uE = t =+! -.: E== f;tE;g -U€!F zZ='= i ,=-.2 : , . : 7 .? 9 E ; ; : =3: . i 2+'i 2 ? =1'== a- _Et >a = t a = 2 - = =u ==D=- C=- !. =. ). q) +) ; =L-. )- : - :- l-. :l ^-1r. i-. l. = ' l E F r) a ! o tri -.1 =- +EdE ! 7!! = = ei-n E y y y i= i=<,6a'JT Eo+.,--tnd-cr € F6aFtrFo "'l P?-rn:r.v:'.2: a +l t4 - rrl z ei 7 d.r4- -- g = 's 1EbSg qp'E .l ; *r .;L= X;g i -t a-E? E?Z !v. :v! +- ^!-^t a- e. p"""""""". :- H a?b- ;g b . c r-:g ^6 ta!i !ti;!E€a;€EiF ;,E: ;;rtIE irs::;r =iE; ;E?ei;jliiriE= ;E=EIeaEgg ;iig+:ci 1;' i=:;i!i3 E!{7 :s€ iii;*gsi!!;-;gc€g$;a giEgg X IE!EE1; 1;g!E!;: FIi;i;iligiggi'' iEilig !aE ;s; sis;iisii!!:i;iE:;*:ee - =? e= oC C' C' * c! al cl =J- \O cl cJ \C al * ca C ci tl ca -l ca C\ ci C. o n f-- rC \c Cl F- ca F- '.- '_ a.l CC aa.t).a t-. F- f- C\ cc =t o\ \O c\ \C o, t-c. ca 5 3\' c N V) o cl n o N \O o N ol ca + N ca \O N et cO O o] N N ca.a n Fc! -l a! c] Cl C.] C\t a] F- aa aa = v. c] cr cr -. -C! ct .l -r.l .! .= ? o .F ! a ?, J -=; a = : =+ = '- i- 1, - f . J - E dX ! - - - lu t '=i-=-==-- u=6 t; ! =;{. =.=r. ;;=7'=?. =*.=!=! - - _- - .. ') : c F, - C-u .3L\ - ' - ' -==. '-! ' = o! t:4:..t2.",7'i-= a =; 7, u,P +pL - v al ca t r - . o I ivv', v', l ! n r =i Ltl r. - t: = --=-- t , ,- ., - vt \c F- cc 3\ c\ 3 t =J .t =J { t i- r, v' ol.a v, v' \c) \c v' v' ) '!=-i.t:r, - :: .i' z> ,J- 1 E : , - -' =' ^=- - I -- -tj tE=3zi. +ia i:aE --.= : - 4 = = = !=xrzt --.-- 4 = a v--aa.L i'-.=g.,/:'i-.|-.t7L E',4 i'l -==a= Ec,-"==jj['O=A:,.? =+, -.7 = : = = J '7.{J=,.a= .= == == =, -a'r= air= -t?z Ei_2 c d eE &6'".: ==! laJ t ,.: a= = ._-> ! =-.t-. t- :'l e.e.. ^- - i.a-.-i;;; 7--'-. r]cgs n i ;,c :: u; . 7 ;, . - -a _ - > 1 u Zl=2-= E= i 9!Y - 4-"'=i,q--:+=a== t * c.: E 9 ! =.= a = ! .='-:2v<-iv!=:C-N-,-rr,a:rx-:i i'=7 : r, t ; a,'aE -7^r,--zEs^,-'- C i. 7 t =J.-a 9- = : t= =t= -.t ' !..= cPF U .a- = - : -t ; du, = = . = tI t l = = a = = , i : . ! . ; 1 ?,72 i='l -+i1**-t,=!?--'.== _ _ , _. _ a C >:=.= = oc D, =i =; =, OL - € F ,F F A € € € E e R F ? 8 ! s ? s 3 3 D A a 8 5 S S 3 = = = = : i € U U f: 3E - ! ' = L E o'-065 3 -o- o'.=c-c o!=3,:E a L is F o bE i !- =. E i .O E = r -. = ,E 9 {=t a-:-a c -E cd tts =' z= EZ --_ ii= E lt t z i+ i = . . 2tz - := -: cc t= =t= . 5 r= 7'7 z = i :cj. ;2# eZ!. i i lE EE a- E ia 1' .ai E^= : t3-=Z p + A :g e t e a 8.. ',- ;t - ! = 7+98 : =Zi./rE E 2;iE ;; {; a- 7 i - = "Z Z , t 2*a =.;,? -*E 2=aE=.tti =777 .i=== tt Z==-e.7'.T=.2 - = = u ,. z- , ; = ,ilE7:. Z7 y2 = . =' ==2=-i 7 z a . = t-i= = 1!;zJ--=.c.=i Ei;ti-iv=' =-a'i;'==='i--_ ,z!= Z='r 2E! ! 1n ait7==z z== E ==tt,?=i =i E:: E t9 zzE?*t'yry, + == t,,?u1= = i t 7 , i = = t = E = t = : ' . = | ; , . ' : , 2 . = = = ' = "z E=: ?a: j t ti,-.1!=;?i:;?*z-tzttEaEZii::iz =v-=E2 = .n2- \i4 / 1iZ lEE i: - *-.:ca:1:E; = i: 72 .!-z==aZ iEET EE2== ::: ?-=ii 2tr7=_zy:;: -.-tv. '.iai.i,q=i--.r ! =f =tv1E-:n ?€-lc1 c-.r Ea-j--j-j-j€ri :i+++++(r++ cl i-: o.d X + Vaz ==z.t E 1 4 i, i = = = = .=.!a-;a=i=!::-V *i;: r F ",,-tr i i 2i==:Z2E v y=== E; a:zhh='!i77'= clEiZr^iz=t*^2.:=:?=a' - N - . * ii=;i;i -, l-^-.; ---io-'=t =\n-7:=v a it}-cin€-:.'llt":l-: .1 nnfn cl;..ri Egiorci sl.';'v"i'ncin E;4;'virl .* -, .5-.".r Q"i; ,r' ;, cl n v) - =-',a n tr) l7J: -:! -atl= E:l-Uf ,\ ? ) vn' :an' :aF' \^ , i \ct-r') ';' v'vt x x b3a c€ clcn\o<-n <-d-dnr) -6ocac.) 6 .a 6 ..) c7) 6 o E oo€ A:F O O- o.! :"??- c! a) : ^ ^ O ! L ! d D ! ^ - - : a>><,jj = a9 Vo n-lalf,+ ea!clcial : tr a.l r o *.o c c Sii-.if i-l al C.l (.l ,: .-: cs x tr trFF E-^, t t:a - >rrs> a 1! >\v at 9': E. i.9 x = =1t gI- \o o, o'\ c\ v)\o\o coc ol $<-$<-t/]r)v)r)V)\O\O\C\O\C (\ al cl N cJ c] cl cl a.l cl ^l !j- <-\o o.l N al -j-'f N * d c\ o\-i <- t-- Ftr-cOcOcOcOCO@O\C\O\O\O\O\ a.l N a.l (.l o1 o] tl al (\ 61 6l oI oo: = 6.1 ca N c.rca * OC O = ca.a cr) ca v) r) o'\ cl v) ooc) N +iia.tGlCicaf,f,f,-,-t ci ca o o o o o - c\ O ca ca r, r, F- al (,€.-, (, E !2 6 ,- ^ c-) nE du C qc r-i $ -S ! E Ez Fe E E EE El ! E= \1= :5 1- " iz 6 E ir *Y : + a ,^ V ir - E. =*si; Ee :E E E Ea ;-;i=z=Z_ii;_.:2i_= € a+;+;+ e" E ;i a E €F !!?-pt ! -; o-f ! E 9 t; H"-=e =i Ez E Zl I :5 !u u *.i -F 25, :: E .c ; {;oz= .=F, ;} 4 E !;;:*iF flP-E== d: 'E E b .r ! =EZa|;EEEZ i77=-. ?": E. Lr il F E i |rip i,,i!== I rE E'0 a!!= iEgii?{EEi;ii;+;i,E=-ii E iE!2:?_-rE .. 4zv.=rztE+zvE?Eizzz*,Ei=i=Ez;z =+=E; E it -.a:,- - - ;iriii?,=:I:ii; F i iaEaE|EE E ; |-=i-yL -=e==Z z=: . . 5; v =;;E =:EEE 2 = =E7aii; 2t a iEP+ =",;:'!!7:2 ,:ii ..yPEi,E ?. 3? a =: i! e! z " +? : : 1 - = t 4; = E i *; . = . ++++eT!= i= l = ? =. ? !+2 E" ai -Fii ;i=? ;-E E = i i zEi i .:E ! :v1+=F iiiiizig;gi:s: iltETli€: ia;\?Ei; t3a&dta5ic zar.s"4,iaaeaad 4€oo6;EE=tdJdi ;9! t4i#35idI O r n v) .o cl.o ?.: s O * 1 v) .o ! cr c. + O * n v, .o N c 7 )+ O * 9 n 6 c.rco <-,..o s O = c . rc . o \ r) .o gE 7c. O.d r.r.n O *...r 3(, E ; EAA N .= -.+c!..$r) 1 c. ; 3.3--j-:i .o |.l..1 iAsF.-Fs \o fr- .o .o $a AF- i; * ". - * - F \c -: F- Introduction .l The scopeof design ,l.l Tasksandactivities : engineer's main task is to apply his scientific knowledge to the solution of nical problems and then to optimise that solution within the given material, nological and economic constraints.To this task the designermakes a highly rrtant contribution. His ideas, knowledge and abilities have a fundamental t on the nature of manufactured products, their customer appeal and their rall profitability. Dcsigning is the intellectual attempt to meet certain demands in the best blc way. It is an engineering activitythat impingeson nearlyeverysphereof irn life, relies on the discoveries and laws of science, and creates the itions for applyingtheselaws to the manufactureof usefulproducts[1.2, . t . 11 . . 2 67, . 5 2\,. 5 3 1 . l ) i x o n [ 1 . 1 1 ] a n d l a t e r P e n n y [ 1 . 3 4 ]h a v e p l a c e dt h e w o r k o f t h e e n g i n e e r i n g igrrcr at the centre of two intersectingcultural and technical streams(Figure l), ()ther modelsare also possible: respects,designingis a creative activity that calls for a sound Iu 1t.t.vt'hologlcal rtling in mathematics, physics, chemistry, mechanics, thermodynamics, rrtlynamics,electrical engineering, production engineering, materials tech- Politics I Sociology, psycn0r0gy I Economics Sclence Englneering E n g i n e e r i n g ili;l,:d:desrgn science Industrial deslgn I I riesr0n Arlislic of l l , l . T h e c e n t r a activity de*i3n,from n I l..ial I Arl Production I Introduction nology and design theory, together with practical knowledge and experience in specialist fields. Initiative, resolution, economic insight, tenacity, optimism, sociability and teamwork are qualities that will stand all designersin good stead and are indispensable to those in responsible positions 1I.32]. ln systematicrespects, designing is the optimisation of given objectives within partly conflicting constraints.Requirements changewith time, so that a particular solution can only be optimised in a particular set of circumstances. ln organisational respects,designing plays an essentialpart in the manufacture and processingof raw materialsand products. It callsfor closecollaborationwith workers in many other spheres.Thus, to collect all the information he needs,the designer must establish close links with salesmen, buyers, cost accountants, estimators,planners, production engineers,materialsspecialists,researchworkers, test engineers and standards engineers. A good flow of information and regular exchangesof experienceare essentialand must be encouragedby proper organisation and personal example. Because of the different demands made upon him and the many possible procedures,the designer can play a variety of roles in the productive process. With original designs not commissioned by an outside client, the conceptual and embodiment design phasesare often organisedseparatelyfrom the execution of the order (see 3.2) as shown in Figure 1.2. This organisationalmodel Conceptual design Embodiment design Design activity E- Product creation ---- Inlormation I 'fhe scope of design thc senior staff providing backup with technical advice and cost calculations llrigure 1.3). Such work must be scheduledvery carefully-innovations can only hc introduced step by step and within limits; otherwise the risks may become too Areat, especially in heavy engineeringprojects. lu1111ll, \t- r.iI Conct -__jlEmbo( Figure 1.3. Organisationof the design activity: the conceptual design,the embodiment design, the detail designand the cxecution of the order are organisedjointly Product creation ----- Inl0rmatl0n When it comes to mass production, it is often useful to involve the experirnental development department in the design activity (Figure 1.4) because ht'rc the fusion of theory and experimental development greatly facilitates the dcsigner's task. Many problems are solved more simply and cheaply by l)r('lirninary and relatively cheap experimentsthan by calculation and work on tlrt' rlrawing board alone. Some designs require further development work ht'lorc production; in other casesthe job can be started straight away. f-l |l -|l Mana0ementl L Figure 1.2. Organisationof the design activity: the conceptual design and the embodiment design are separated from the d e t a i l d e s i g na n d t h e e x e c u t i o n of the order offersthe designermuch greaterfreedomand scopethan conventionaldevelopment design.The drawbackis a possiblesplit betweenthe conceptualand detail design departmentsand also the introduction of a distinction between 'high class'and 'run-of-the-mill' designers.This trend must be deliberatelycounteractedby regular staff exchanges.Thus, when a new designis sentto the detail designdepartment it is advisableto allow a proportion of the staff to move acrosswith it. This will alsoensurean optimumflow of information.Conversely, staff from the detail designdepartmentprovide a useful sourceof experience duringthe conceptualdesignphaseand canhelp to keep productioncostsin line with market realities. With large, one-off products, an order often calls for further developmentin the form of adaptivedesigns.Inthis case,the conceptualdesign,the embodiment design,the detail designand the executionof the order go hanclin hand, r-) design Detail Execution of order r' I ..1.Organisationof the I rptrrr r h ' r r i r ri r c t i v i t y :t h e c o n c e p t u a l ( h ' r r r , r rt .h c c a l c u l a t i o n t, h e f r r r l r o t l i r nnet d e s i g na n d t h e t re tt||r' I rllcntal developmena | c f r , r r ; r l c tl lr o m t h e d e t a i ld e s i g n l n r l t l r c e x c c u t i o no f t h e o r d e r activity t ,-J Hr00ucl crealr0n ---- lntormation Irt lroth psychologicaland organisationalrespects,designwork on a particular :rrr rrnclin a particular place is performed by people with extremely varied l l i o t t a n c lc x n c r i c n c c .F o r t h c m o s t c f f c c t i v ed i v i s i o n o f l a b o u r , t h e w o r k b c s l t i r r c do u t l r c t w c c t ti t c i t d c t t t i c i t l ltyr a i n c c lc n g i n c c r s ,e n g i n e e r sw i t h a p r a c t i c i r lb i r c k g r o u r r dt,c c h n i c i t n s t t t d t l r i r u g l t l s t r t c na, l l w o r k i n g i n c l t t s c )ntIt()n, I Introduction 1.1.2 Typesof design Design terminology has changed considerably within the past few years. Thus' Wogerbauer [1.60] speaks of original development, further development, and adaptive design; Opitz [1.31] of original, adaptive,variant and principle design. In this book, we distinguish between three types of design whose boundaries, however, are not precisely fixed: - Original design which involves elaborating an original solution principle for a system (plant' machine or assembly)with the same, a similar, or a new task. - Adaptive design which involv es adapting a known system (the solution principle remaining the same) to a changed task. Here original designs of parts or assembliesare - often called for. Variant design which involves varying the size and/or arrangement of certain aspectsof the chosen system, the function and solution principle remaining unchanged.No new problems arise aS a result of, say, changesin materials, constraints or technological factors. The last heading also covers commissioned work in which the solution principle and the finished design remain the same and only the dimensions of individual parts are changed on previously prepared drawings. Some authors' 'fixed principle design'. -[1.31] and U.501, call this An enquiry [1.5] held among members of VDMA (German Association of Mechanical Engineering Companies) in 1973 showed that, in the mechanical engineering industry, some 55 per cent of products were based on adaptive designs, 25 per cent on original designs and 20 per cent on variant designs. Although the imprecision of the boundaries of each type makes classification difficuli, the large proportion of original and adaptive designsshows that a good designerhas to be both highly creative and flexible. 1.1.3 The nature of, and needfor, systematicdesign The designer has to be a man of many parts. If we consider the vast range of products he helps to make and the specialisedknowledge or experience that goes into them, it is clear that his work does not fit into a rigid mould. Because design has a crucial effect on the technical and economic value of the productproduction methods can only be optimised within the framework he has bstablished-the designer must aim at a dependable approach. To that end, he must be taught, or be expected to learn, all the special skills underlying systematic thought and procedure. A design method, therefore, must: -"n.ouiuge a problem-directed approach; that is, it must be applicable to every type of design activity, no matter in what specialistfield; I I fhe scope of design foster inventiveness and understanding; that is, facilitate the search for optimum solutions; ,be compatible with the concepts, methods and findings of other disciplines; not rely on chance: 'facilitate the application of known solutionsto related tasks; - - be compatible with electronic data processing; -- be easily taught and learned; and --- reflect modern management-sciencethinking; that is, reduce workload, save time, prevent human error, and help to maintain active interest' Such an approach will lead the designerto possiblesolutionsmore quickly and drrectly than any other. As other disciplinesbecome more scientific,and as the usc of computers calls increasingly for logical data preparation, so designing, loo, must become more logical, more sequential,more transparent, and more ()pcn to correction U.14]. An enhanced appreciationof the designer'scontribulion and status is only possible when his methods and style of work are in line with current developments in scientific and industrial practice. 'fhis is not meant to detract from the importance of intuition or experience. qtritethe contrary-the additional use of systematicprocedurescan only serveto increasethe output and inventivenessof talented designers.Any logical and rvstcmatic approach, however exacting, involves a measureof intuition; that is, tn inkling of the overall solution. No real successis likely without intuition. ln teaching design methods, it is therefore important to foster and guide the rtrrdent's own abilities, to encourage creativity, and at the same time to drive Irorne the need for objective evaluation of the results. Only in this way is it prrssibleto raise the designer'sgeneral standingand the regard in which his work rr hcld. Systematicprocedureshelp to render designingcomprehensibleand also r.;rscthe teacher'sload. However, the student should be warned againsttreating lris teacher'sopinions as so many dogmas.The best teacher merely tries to steer tht.student'sefforts from unconsciousinto consciousand more fruitful paths. As rr result, when he collaborateswith other engineers,the designerwill not merely l',' lrcrldinghis own, but will be able to take the lead [1.32]. Systematic design alone can produce a truly rational approach and hence gt'rrcrallyvalid solutions-that is, solutions that can be used time and again. It rrlsohelps to establisha workable schedulebased on rational project planning, rtrth as Critical Path Analysis, and hence enables the designer to predict how rrrrrchtime he will have to spend on a feasibility study, how much on the search for a solution and how much on the evaluation of the result. Increasedreliance orr similarity laws, so useful in model testing, along with consistent use of ltlrnclard specifications, size ranges and modular methods, facilitates further tirtionalisation, not only in the design activity, but throughout the entire p r ( ) d u c t i o np r o c e s s . 'I'imc i s r n o n c y . L c s s d e m a n d i n g t a s k s c a n b e d e l e g a t e dt o s u b o r d i n a t e s . N f t r r c o v c r ,i t i s r c l a t i v e l yc a s yt ( ) d c t c r n r i n ch o w m u c h o f t h e w o r k i s u n n e c e s s a r y o r b c s t l e f t t o c o m p u t c r s a n d c o m p u t c r - a i d c dd r a u g h t i n g [ 1 . 8 , 1 . 2 6 1 .T h c s c lcmarks also apply to such indircct dcrign activitic$as collccting intirrmation on I Introduction standards,components,matcrialsclc. systcnrttictlcsigngreatlyfacilitatesthe rationaldeploymentof computcrand dutu $yritcnls. A rationalapproachmust als'cover thc designcr'scostcomputations. More accurateand speedypreliminarycalculutions with thc help of better data will becomea necessityin the dcsignfiekJ.lt is csscntialtcl devisemethods with which it is possibleto estimatefinal costs.at lcast approximateiy, even ar an early stagein the designprocess.This. too. callslirr a systematic and rational approach. 1.2 The development of systematic design 1.2.1 Historical backgroun-d and important contributionsby Kesselring,Leyer,Niemannandothers All developmentshave antecedents.They mature when there is a need for them, when the right technology is available, and when they are economicallyfeasible. This also applies to the development we have called 'systematicdesign'. It is difficult to determine its real origins. Can we trace it back to Leonardo da vinci? Anyone looking at the sketchesof this early master must be surprised to see-and the modern systematistdelights in discovering-the great extent to which Leonardo used systematicvariation of possiblesolutions [1.30]. Right up to the industrial era, designingwas closely associatedwith arts and crafts. with the rise of mechanisation,as Redtenbacher[1 .35] pointed out early on in his Prinzipien der Mechanik und des Maschinenbaas (principles of Mechanics and of Machine construction), attention became increasingly focused on a number of characteristicsand principles that continue to be of great importance, namely: sufficient strength, sufficient stiffness,low wear, low friction, minimum use of materials, easy handling, easy assemblyand maximum rationalisation. Redtenbacher'spupil Reuleaux [1.36] developed these ideas but, in view of their often conflicting requirements, suggested that the assessmentof their relative importance must be left to the individual designer's intellieence and discretion. Important contributions to the development of engineering design were also m a d e b y B a c h [ 1 . 1 ] a n d R i e d l e r [ 1 . 3 9 ] , w h o r e a l i s e d t h a t t h e s e l e c t i o no f materials, the choice of production methods and the provision of adequate strength are of equal importance and that they influence one another. Rotscher [1.42] mentions the following essential characteristics of design: specified purpose, effective load paths, and efficient manufacture and assembly. Loads should be conducted along the shortest paths, and if possible by axial forces rather than by bending moments. Longer load paths not only waste materials and increase costs but also require considerablechanges in form. Calculation and laying out must go hand in hand. The designerstarts with what he is sivcn 1 . 2 T h e d e v e l o p m e n t o f s y s t e m a t i cd e s i g n rrnd with ready-made assemblies.As soon as possible, he should make scale tlrawings to ensure the correct spatial layout. Calculation can be used to obtain cither rough estimatesfor the preliminary layout, or precisevaluesfor checking the detail design. Laudien II.24], examining the load paths in machine parts, givesthe following rrclvice:for a rigid connection, join the parts in the direction of the load; if llexibility is required, join the parts along indirect load paths; do not make unnecessaryprovisions;do not over-specify;do not fulfil more demandsthan are rcquired; save by simplification and economical construction. Modern systematicideas were pioneered by Erkens [1.12] in the 1920s.He insistson a step-by-stepapproach basedon constant testing and evaluation, and rrlsoon the balancing of conflicting demands, a processthat must be continued turrtila network of ideas-the design-emerges. A more comprehensive account of the 'technique of design' has been prcsented by Wogerbauer [1.60] who divides the designer's overall task into rubsidiary tasks, and these into operational and implementationaltasks.He also cxrrmines (but fails to present in systematic form) the numerous interrelalionships between the identifiable constraints the designer must take into lc('ount. Wogerbauer himself does not proceed to a systematicelaboration of lrrlrrtions. His systematicsearch starts with a solution discoveredmore or less Inlrritively and varied as comprehensively as possible in respect of the basic for rrr, materials and method of manufacture. The resulting profusion of possible trrlrrtionsis then reduced by tests and evaluations,cost being a crucial criterion. W,'qcrbauer's very comprehensivelist of characteristicshelps in the search for In ()l)timum solution and also in testing and evaluating the results. llrough some need for improving and rationalising the design process was fclt evcn before World War II, progress was impeded by the following tttls' tlrc lrbsenceof a reliable means of representingabstract ideas; and tlrc widespread view that designing is a form of art, not a technical activity l r k c i r n yo t h e r . lht' risc of systematicdesign had therefore to wait until these obstacleshad rr t lcared and for the wider adoption of systematictechniques,not least in tct'hnical areas, and the emergenceof modern data-processingmethods. A rotl ol staff shortages[1.54] provided a further impetus. lvftrtlt'r'lricleasof systematic design were given a greal boost by Kesselring, rt'hncr, Niemann, Matousek and Leyer. These men were not merelv rttrrnl 1'rioneers; their work continues to provide most useful suggestionsfor ittg thc individual phasesand steps of systematicdesign. :lring Il.l9l first explainedthe basisof his method of successive t r r l i o r r si n 1 9 4 2( f o r a s u m m a r y s e e[ 1 . 2 1 ] a n d [ 1 . 5 6 ] ) .I t s s a l i e n tf e a t u r e cvirluirtion ol fornr variants accordingto technical and economic criteria. l h c o r y o t t c c h n i c i r cl o r n p o s i t i o nI I . 2 0 1 ,K c s s e l r i n gp r e s e n t s - i n a d d i t i o n t o r t l b r r s i ci d c a s o n t h c t c c h n i c a lc o n t r i b u t i o n o f t h c d c s i g n e ra n d o n h i s . u t t i t u d c i u r d r c s p o n s i l l i l i t y - t n a c c ( ) u n lo l ' t h c u n d c r l y i n g s c i c n t i f i c I Introcluction principles (the mathematical and physical relationships) and the economic constraints (the manufacturing costs). In the theory of form design which he derives from the above, he mentions five overlying principles: -the principle of minimum manufacturing costs; -the principle of minimum space requirement; -the principle of minimum weight; -the principle of minimum losses;and -the principle of optimum handling. The design and optimisation of individual parts and simple technical artefacts is the object of the theory of form design.It is characterisedby the simultaneous application of physical and economic laws, and leads to a determination of the shape and dimensions of components and an appropriate choice of materials, manufacturing methods etc. If selected optimisation characteristicsare taken into account, the best solution can be found with the help of mathematical methods. Tschochner [1.53] mentions four fundamental design factors, namely working principle, material, form and size. They are interconnected and dependent on the requirements, the number of units, costs etc. The designer starts from the working principle, determines the other fundamental factors-material and form-and matches them with the help of the chosen dimensions. Niemann [1.29] starts out with a scalelayout of the overall designshowing the main dimensions and the general arrangement. Next he divides the overall design into parts that can be developed in parallel. He proceeds from a definition of the task to a systematicvariation of possible solutions and finally to a formal selection of the optimum solution. These steps are in general agreement with those used in more recent methods. Niemann also draws attention to the then lack of methods for arriving at new solutions. He must be considereda pioneer of systematicdesign inasmuch as he consistentlydemanded and encouragedits development. Matousek [1.27] lists four essential factors: working principle, material, manufacture and form design, and then, following Wogerbauer [1.60] , elaborates an overall working plan based on these four factors. He adds that, if the cost aspect is unsatisfactory, all four factors have to be re-examined in an iterative manner. Leyer 11.25] is mainly concerned with form design. He distinguishes three main design phases. In the first, the working principle is laid down with the help of an idea, an invention, or establishedfacts; the secondphase is that of actual design;the third phaseis that of implementation. His secondphaseis essentially that of embodiment, that is, layout and form design supported by calculations. During this phase, principles or rules have to be taken into account-for instance the principle of constant wall thickness, the principle of lightweight construction, the principle of shortest load paths and the principle of homogeneity. Leyer's rules of form design are so valuable because.in practice, failure is still far lessfrequently the result of bad working principlcs than of poor detail desisn. I h c d e v e l o p m e n t o f s y s t e m a t i cd e s i g n 1.2.2 Modern designmethods 'l hese preliminary attempts made way for intensive developments,mainly by runivcrsityprofessorswho had learned the art of design in practicalcontact with prorlucts of increasing complexity. They realised that greater reliance on phvsics,mathematics and information theory, and the use of systematictechnics. were not only possible but, with the growing division of labour, quite lispensable.Needless to say, these developmentswere strongly affected by .' requirements of the particular industries in which they originated. Most rc from precision, power transmissionand electromechanicalengineering,in ich systematicrelationshipsare more obvious than in heavy engineering. Svstematicdesign according to Hansen rrscnand other membersof the IlmenauSchool(Bischoff,Bock) first put rrrdtheir systematic designproposalsin the early 1950sII.6, I.1,I.16]. nscn presented a more comprehensivedesignsystem in the secondedition of standard work published in 1965 [1.17]. He himself has summed up his rt'eclure as follows: l)etcrmine the crux of your task, becauseit is common to all solutions. ('ombine the possible elements purposefully, for all solutions develop from r r r c hc o m b i n a t i o n s . l)e tcrmine the shortcomingsof every solution and try to reduce them or their t ' l Ic c t s . St'lect the solution with the fewest shortcomings. l'rovide documentation to permit practical evaluation. I lresc procedural rules are the basisof a systemcomprising four steps.Figure 1 rlrows the steps for the conceptual phase. The same four-step system is also rcrl cluring the subsequentdesign phases.Hansen begins with the analysis, r(lu('. and specificationof the task, which leads him to the basic principle. rnrrst be formulated abstractlv enough to comprise everv conceivable Iro11.111.',1 must embrace the overall function derived from the task. 'l ht' sccond step is the systematic search for solution elements and their l r r r u r l i o ni n t o w o r k i n g m e a n s . lrttachesgreat importance to the third step, in which any shortcomings l.rrrs.'n I rt'r'icwcclin order to develop improved working means. In thc lirurth and last step, these improved working means are evaluated to r r r i r r cl h c o p t i m u m w o r k i n g m e a n sf o r t h e t a s k . l()7-l lf irrrscnpublished another work, entitled Konstruktionswissenschafl s c i c n c co l c l c s i g n )[ l . l t t l . I n i t , h c u s e ss y s t e m sa n a l y s i sa n d i n f o r m a t i o n t o t l c f i n c t h c r l c s i g n p r o c c s si r n c lt h c n i t t t t r c o f t e c h n i c a la r t e f a c t s .H e o r r l h c v l r i o u s t y p c so l ' s l r u c t u r ci r n r ll u t t c l i o t ri r n c lt h c i r i n t e r r e l a t i o n s h i p s , 'l'hc lrook is tn<lrcconccrncclwith alur discusscsthc prtlblcln of dnlu ril()ritgc. thun with ruler of pructicll tlcsign. icrl l'undanrcnttls 1t) I Introductio Task 11 t l t v c l o p m e n t o f s y s t e m a t i cd e s i g n Input (Concept, layout) signals material, Energy, Prel iminary considerations principle Basic Search lorandcombinati0n 0f eiementary soluti0n elements Working means (Workrng principle, principle) lormdesign Review ol shortcomings lmproved working means (Working principle, principle) lormdesign - Rationai evaluation 0ptimum working means (Working principie, principle) lormdesign Function of a Machine fulflledbya Physical Process realised bythe FormDesign Features of theoverall design Concept (Layout, Production documents) Figure1.5. Designstepsaccordingto Hansen[1.16,1.17] 1. From the taskvia workingmeansto the conceot 2. Fromthe conceptviawor-king principlesto the layout 3. From the layoutvia form designprinciplesto the productiondocuments Similarly, Miiller [1.28] in his Grundlagen der systematischenHeuristik (The fundamentalsof systematicheuristics)presentsa theoretical and abstractpicture of the design process. 2 Systematic design according to Rodenacker After Hansen, it is Rodenacker [1.40] above all whose original design method has made the greatest impact. He starts out from the fact that every machine must fulfil certain purposesor functions. Rodenacker looks upon designingas a transformation of information, leading from the abstract to the concrete. Designing is a reversal of physical experiments.Figure 1.6 showsthe main steps of Rodenacker's design method. He starts out by defining and abstracting the requirements and by establishing a function structure. Next he looks foi the appropriate physical process and finally for the required form design features. Rodenacker develops his systematic principles chiefly by means of examples taken from process engineering, but his ideas apply quite generally to the development of technical systems.He proposesthe folrowing rules [1.a1]: Rule 1. Clarify the task (the required relationships). Rule 2. Establish the function structure (the logical relationships). Rule 3. Choose the physical process (the physical relationships). Rule 4. Determine the embodiment (the constructionalrelationships). Rule 5. check the logical, physical and constructionalrelationshipsby appropriate calculation. Rule 6. Eliminate disturbine factors and errors. 0utput 1.6. Design steps according to Rodenacker [1.401 7 . l i i n a l i s et h e o v e r a l l d e s i g n . fi. I{cview the chosen design. this rrrcthod, the function structures (Rule 2) are based solely on functions 'rl lrtrrn two-valued logic. These functions are separation,connection and tlrc llow of energy, material or signalsin technicalsystems)channelling.We I'r' r'rrrnriningthe problem of logical functions at greater length in 5.3.3. ' losiL',,1rcquirements having been satisfied,the next step is to choose the Srlrrsit'irlprocess(Rule 3). Rodenacker usesphysical effectsand equations, p;rrlicular attention to the time factor. Experimentation is said to be the \{ ru| ( (' of information. tlrc rrcxt concrete step, Rodenacker determines the embodiment fixed by lnr,'rrl lrrrclthe form desiqn features. The latter result from the variation of t's. rrrrlcrials and motions to achievethe required characteristics(Rule 4). ' \ \ ( ) l k c k r r . r cs o f a r m u s t b e c h e c k e d b y c a r r y i n g o u t t h e a p p r o p r i a t e t r o r r s .s r r c ha s c o m p o n e n t s t r e s s e s( R u l e 5 ) . nirr'kcr is particularly concerned with the identification and elimination rrrlrillti lirctors causing quantitative and qualitative fluctuations (Rule 6). tnirv surrr it irll up by saying that the main factor of systematicdesign, trg to l{otlcnackcr, is the determination of the underlying physical :r. ln his irl.rproachR . o c l c n a c k c rc o n s i c l c r sn o t o n l y t h e s y s t e m a t i c n l o l c o n c r c t c r l c s i s nt i r s k si r s o u t l i n c r ll b o v c . b r . r ta l s o t h e m e t h o d o l o g v i t t g ' n c w r l c v i c c si r r t r rl t t i r c l t i t t c s . ' lt' lrtl i r lc t r t .l h c l ) r o p o s c ist s c i t r c hf o r s l k n o w n g t l t y s i c i rcll l c c t s i r s i r n l c i u r so l ' i t r r i v i n g l r t n c w i t l ) l ) l i c i t l i o l ro xrlttlions. 72 I Introduetion 3 Algorithmic selectionprocedure for design basedon design catalogues according to Roth Roth divides the design process into several phases,each with specific steps that, depending on the results, must be repeated several times [1.44, 1.48] (Figure 1.7). 0. Task-formulation requtrements echnical cosls andspecitied I h c d e v c l o p m e n t o l s y s t e m a t i cd e s i g n 1a IJ I lrc first phase is the analysisof the product environment leading to a precise dr'lirrition of the problem. This definition includes the specified function, the 'hrrical requirements and the specified costs, three selection criteria with t'sc help it is possible to choose from design cataloguesin due course. Next 'lunctional relationship has to be elaborated in two steps. To discover the t'rirll function structure, every statement in the problem definition must be rciated with a system of general functions so selectedthat different alterna's result from changesin their arrangement.Roth's 'general functions' refer gcneral characteristicsdetermining various technical artefacts-that is, which cct, change, store and channel material, energy or information [1 .45-I.47, ll. Once the general functional relationship has been determined, the ible combinations of sub-functions are split up into recurring elementary l r l c m sw h i c h c a n b e c a t a l o g u e d . 'l hc next step, the determination of the special function structure, is the rpt to solve these elementary problems with the help of basic physical irtions-that is, of physical effects expressed,if possible, in mathematical trrlite. of overall 1.1Determination lunction structure ol special 1.2Determinatton lunction structure and 2.1Formdesign selection material forproduction 2.2Design tunctronAclualcosts Actual lunction ,"^. '""" actual-specified costs, actual-specilied drawings Production 'he actual product is elaborated in the next phase. First comes the embodirt oi the solutions of the elementarv Droblems. A seneral solution is then rrirted by combining these embodiments in accordance with the overall ti()n structure. Variation will now produce a host of general solutions from 'lr tlrc best can be chosen and modified in the light of production requirerrts.'fhe resulting variants are evaluated in turn so that the most promising tlr('l)roblem in hand can be determined before oroduction drawinss and i l ) r r ' l r t a t i o na r e c o m p l e t e d . [,'t lr rcfers to the whole process as an'algorithmic selection procedure for rr lxrsed on design catalogues'. He suggeststhat the information needed tlr, inclividualstepsis best chosenfrom catalogueswith the help of selection r, lt'r.istics.He accordingly attachesgreat importance to the compilation of r : r r r r l < r g u[e1s. 4 3 , I . 4 9 1( s e e5 . 4 . 3 ) . :rlgorithmic-physical design method according to Koller (",\('nliirlfeaturesof Koller'smethod1I.22,I.231are the breakdownof the r l)r'occssinto a larger number of steps and the emphasis placed on 'rrlrrrv lrhvsical connections. The aim is the algorithmicisationand hence ir\rng corrlputerisationof design. Figure 1.8 shows the various phases of r'r schcrne and the elementarv desisn activities associatedwith them. He trishcs bctwccn function svnthesis.oualitative svnthesisand cuantitative s, l'or iur cxact dcscription of the individual steps the reader is referred l i l c r r r t t r Ir lc. : . ] 1 . r r s c o l c ( ) r r r p u t c r as l s o c a l l s l i r r t h c f i l r r n u l a t i o no f c l e a r r u l e s g o v e r n i n g l r o c c s s e tso c l c t t t c n l i r r ys t c p . ' l i r t h i r t c r t d K o l l c r r c t l u c c sc o n t p l c xt c c h n i c a p nunll)cr ol physicll lunctiotts lnd lltclt slipulatcs rulcs lirr thcir F i g u r c I . 7 . P h r r s c rsr n t ls t c P so l t l t c r l c s i g t lt l I ( ) c c s ist c c t r r t l i t ttgo l { o t l r I I ' l t i l cirn bc cxprcsscdby nt. llis basicprcmisc is thrl thcrc l'unctiort$ | ' I h e d e v e l o p m e n t o f s y s t e r n a t i cd e s i g n hrrownmachine elementsand also by elementsthat have still to be developed. ln tlt'riving his functions, Koller?s starting point is the fact that, in technical rvslems, only the properties and states of energy, material, signals and their llows can be changed in magnitude and direction. Together with the physical ittputs and outputs, this gives 12 functions and correspondinginversefunctions, which Koller calls basic operations, amongst them channelling and isolating, :reasingand decreasing,coupling and interrupting (see Figure 2.5). Ilccause technical systems involve logical as well as physical relationships, ion synthesis will yield a function structure consisting of logically and ysically interrelated basic operations. Koller contends that, in the qualitative design phase, the basic operations of Irnical systems depend exclusively on physical, chemical and/or biological :ts. The choice of the appropriate effect producessolution principlesfor the rlormance of the basic operations which, with more complex requirementsor nctlon structures, may be combined into assembliesand overall systems.A rirlitativedesign', finally, demands the determination of shapes,to which end rller has devised several rules of form desiqn. Variations involvins several tcria are stipulated for every step, for example variations of the material or of shape,thus paving the way for severalsolution variants. Koller's quantitative ltthesiscomprises the classicaldesign activitiesof calculation and layout. lrr accordancewith his aim of algorithmicisingand computerisingthe design rcess,Koller lays particular emphasison small (elementary) developmentsor ps and on clear mathematical rules and definitions. .9 C -F = c L 15 struclure lunclion Elem. I '6 C E E theconcept Fixing = L .2.3 Other proposals out(arranglng) Laying 'I .6 .z = <J torproduction Preparing I c= iIo Figure1.8. Stepsand elementarydesignactivitiesaccordingto Koller [1.22. he systems approach ternstheory as an inter-disciplinaryscienceusesspecialmethods, procedures I rridsfor the analysis,planning, selection and optimum design of complex t r ' r n s[ 1 . 3 , 1 . 9 , 1 . 1 0 ,1 . 6 1 ] . I r'ehnical artefacts, including the products of light and heavy engineering hrstry,are artificial, concrete and mostly dynamic systemsconsistinsof setsof 'rccl elements, interrelated by virtue of their properties. A system is also rrcterisedby the fact that it has a boundary which cuts acrossits links with t'rrvironment (Figure 1.9). These links determine the external behaviour of \vstem, so that it is possible to define a function expressingthe relationship s'ccn inputs and outputs, and hence changesin the magnitudesof the system l r l c s( s e e2 . 1. 3 ) . l;ronr the idea that technical artefactscan be representedas systems,it was a t s t c p t o t h c a p p l i c a t i o no f s y s t c m st h c r l r yt o t h e d e s i g np r o c e s s t, h e m o r e s o tltc objcctivcsof systcms thcorycorrcspund vcry largclyto thc cxpectations Itavcof a good dcsignmethodus specificdat thc bcginningof this chaptcr l, 'l'hc sy$tcmsupproachrellcctf tho gonoralupprcciationrhar complcx d e v c k r p m e n t o f s y s t e m a t i cd e s i g n l6 1'7 System studies problem analysis, definition State Goalprogramme Goalsetting, criterla lisl System synthesis Development ol solution variants ors2; orerements ors;s2,- s,a:subsvstems Slti::"t*,tJll.,Jlff,}i131ft',T.r,,",'., 1, - 13:inPuts; O1-O2: outputs System analysis Properties andbehaviour of variants eaih involving analysisand syn problems are best tackled in fixed steps' ,r";;';?"t].10 The first of theseis t showsthe stepsof rhe systemsapproach. * ;---^-:^ -^^,,i-amcnrq (rr irr'ur'rou"l' nto,-uiio'i9" g or gatherin gatnerlrlg 111"i-:: Tlo: -T;::i. the clea :.""*- th"'v*1."T., The :?*"l".tJ' he is aim here trend studiesor known requirements' System evaluation goalpr0gramme Evaluation ol varlants against market analyses, to U" t91".:.1.,tl formulationof ttreprobieir(o' 'ut'-ptot'lem) up the purpose: is drawn :T:::":i.:t:3 programme perhapseven during the first step' a goalsof th""tft:l thl torrnui'"^f*ion'to which is to give System decision 0ptimum system selected ::::i:i]:51'J andhenc variants of solution eviruation ffi:}.';ij,',[r,;^i;;;",.Jrlqr"r, for the discovery.f th;';;ti"tu"i'otutio'.t Soluti.onvzrriantsare tneXs;'r1tfe;s"1 Beforether u.quir",rduringthefirsttwosteps. :,T;: ilH.?ir;'t;;;#;*. 6J:u:1,*"-t':: lilt^t*::: the performunt" canbe variants l "uuruut"i, tt" evaluationthar follows,the performance ;rd ;#;;r.'rn ;?:'r*"". each variant is comparedwith the original Coul:'ing,,:",:l:-:i:i:-tl.*1: System lmplementation Plans phase Planning thenexlsystems , I . 10.Stepsof the systems approach decisionismadeandtheoptimumsystemselected.Finally,informationlSglve As outintheformof-y';;;;l;meniation plans' ltsl: 1 l9::^"^Y,l:lniii'l goil' so that iterativeproceduresmay do not alwayst"uOstruigntto tn" final this optimisationprocess,which con needed.Built_in o".irir", ,t"ps facilitatJ , rscn [1.15] are currently developing this method, paying particular attent,r thc optimisation of dynamic systems. tutes a transformation of information' An importantrprt"rl^ti "ppiit"io,n of -th"-:tt:::::p^p::ffi1"f # concept basisof a known or a developedsolution "r;;,;;;;"r;;;;r;{."o"the ;';";;;i;' to p.oou." ispossible ""d"1.,!fil::'-'T,-',::"'::'i'JJ1:li:L,'5t with their tinks,canbe subjected iogether hil;*;; ;*:I'ffi# ttt" d:.*-1111:tjf,lt:||:,T; opti-it"d to satisfv variationinJ mathematical about1 statements mathematical such of ir, ,r," use ffi::H',i';il;;; t:r"f: n* uehaviour ( ; r r a v r v uoi j: h e:L?,':9t a n l l t a l l v sb quuantitative "r^ . so ' : tutio e l e m e n3:' t s f. : ; ^ - ; ' :and11:l]: , . 1 - . - ; .li"^ h e v i o r r of or I the t h e :il*:il behaviour dynamic relations expresslng tne statistical and the formulation of a goal functttl instance in the form of transfer functions, 1 . . hhavc ,'rr" h btr ' of systems-toq:l* Such mathematicaltu*' fo' the description RichtcrIl '37' I 3t'tlt equipment' aboveall, for signal-processing cstablished, th.signas a learning process rrrplcmcntto the methods we have been describingis basedon the view that rrrrt'-sidccl cmohasison discursivemethods is inadequate and cannot, in any , srrtisl'ythc clesigncr.For that reason various attempts have been made to lop tlcsign mcthoclswith the help of automatic control techniquesinvolving i r r r tl c c d b i r c k .S r " r cm h c t h o d s n o t o n l v h c l p t o e l u c i c l a t et h e i n t e r r e l a t i o n s h i p ' e r t t l c s i g r t c t ' l r t ccl r t v i r o n n r c n l . l ' r uitt l s t lt h r o w l i g h t o n t h c g e n e r a ls c o p eo f tt tltouglttl)roccsscs W l i c l r t l c r 1 1 , 5 7 . l . 5 f { l l r i r s t t r g t t c t l .h y i t t t i r k r g vw i t h s t t c l tc y b c r t l et i c n s c o n l r o l i t n d l c i t r n i n g .t h u t c f c l t i v c d c s i g n i s l l t c t t t o s lc o t t t p l c xl o r t n I lntroduction 18 that of the ,learningprocess'.Learning representsa higher form of control, one also but (rules), quality at constant involves not onty quantitative "nung"t as quantities technical changesin the qualiiy itself. Similarly, designingchanges well as working PrinciPles. In structuralterms,iearning and control can, despitequalitativedifferences, Thus Figure1.11representsa be consideredas comparablecircularprocesses. designcycle involvinga learning system,an environmentand the relevantflow lCommunicationDecision Storage Action 'l'he | ' d e v e l o p m e n t o f s y s t e m a t i cd e s i g n 19 'l hc learning processthus keepsincreasingthe level of information and hence flt'ilitatesthe searchfor a solution. General comparison and statementof the authors' own aims closerexaminationthe methodswe havebeendescribingprove to havebeen rnglyinfluencedby their authors'specialist fields.They nevertheless resemone another far more closelythan the variousconceptsand terms would n to suggest. ln particular, several of these methods have been strongly influenced by the rtluctsof precision engineering and power transmissiontechnologywhich, in r, bear a strong structural resemblance to electronic systems. This has sted a breakdown of functions and associatedsub-solutionsinto elements mbling the functional building blocks of electronics.Moreover, the compilaof classificationschemesand design catalogues,as well as the combination solution elements, is easier in these fields than it is in general mechanical rrcenng Itr ull the methods examined, the requirementsare abstractedfor the purpose trriving at generally valid functions. The degree to which the various authors irk down their functions. however. differs from case to case. All stressthe lr'Iirnceof physicalprocessesduring the first phase.They alsosharethe idea of :p-lry-step advancefrom a qualitativeto a quantitativephase.Furthermore, ol t ltem stipulate a deliberate variation and combination of solution elements tlillcrent complexity. All try to algorithmicise the design process and to css it by simple rules or laws. This common approach has been embodied in Environment system Learning F i g u r e1 . 1 1 D e s i g n c y c l e w i t h l e a r n i n g S y s t e m a n d e n v i r o n m e n t , a f t e r W i i c h t l e r [ 1 . 5 8 1 'learning system:designer'takes a problem from' of information. The pro returnsa solution to, the enviionment. Discursiveand intuitive actions learning system' solutions (ideas) that are held in the short-term store of the demands environmental the with solution comparison of the proposed lt,"^Tt hiscrepunciescallingfor new decisionsand henceleadir ;h;";;; ;;';"J;;; are reducedto a minimum, the optimu discrepancies the to new actions.once of optimisationis called a learni process in this cycle solution is to hand. A from 1 element. The learning iystem musi not be consideredin isolation the requir environment.In otheiwords, the environmentnot merely imposes in findi part ments and receivesthe solutions,but frequentlyplays a crucial which presel the latter. we distinguish between a passiveenvironment, to informati information on demuni, and an activeenvironment,which reacts in involved returned by the learning system,or in other words is directly optimisati of discoveryof the solutioti.Wnut mattersis that, for the purpose as a con the designprocessshouldbe treated, not statically,but dynamically process-inwhich the information feedbackmust be repeatedY"iit llt llllll can bc ft solution iion content has reached the level at which the optimum ' l i n cV D I 2 2 2 2 |.551. rtlr thc help of the methods examined above together with our own work we , irr what follows, endeavour to present a comprehensivetheory of general 'r'r'ingdesign. Most of the argumentsare elaborationsof a seriesof papers frtrlrlishedin 1912-1914[1.33] and which we have since discussedat some lr rvith a number of practising designers and research engineers, tested ltt'tllv in practice, and changed and amplified accordingly.Our own theory rrrlrrckresnot claim to be the final word on the subject-it simply tries to rre various methods in a coherent and oracticable wav. We hooe that it \('r'\'c as an introduction and springboard for the learner; as a help and lrtiorr lilr the teacher; and as a source of information, and perhaps of r l c l r r r r i r r gl i.r r t h e p r a c t i t i o n e r . 2T I F u n d a m e n t a l so f e n g i n e c r i n g s y s t e m s Fundamentals A concrete example is the combined coupling shown in Figure 2.1. This can be treated as two sub-systems-a flexible coupling and a clutch. The sub-system 't ltrtch' can, in turn, be subdivided into system elements, in this case ( ()llrponents. Designingisamany-sidedandwide-rangingactivity.Itisbasednotonlyon etc-but math"emaiics,physics and their branches-mechanics, thermodynamics industrial elements, machine also on proOuition technology, materials science, and cost u..orrrting, which are not discussedin this book' -unug"..nt i;?;;";p a theory of designthut .un serveas a strategyfor the development -__^L^*^ ^^A systems of solutions, we must first examine the fundamentalsof engineering make detail p.oc"du."s.' only when that has been done is it possible to recommendationsfor design work. I 2.1 Fundamentalsof engineeringsystems 2.1.1 System' plant' equipment, machine' assemblyand componen as plant Technical tasks are performed with the help of such technical artefacts approxi in the here listed components, and equipment, machines, assemblies uses idenli-cal have not may terms These mate order of their complexity. is sometim evaporator) (reactor, equipment of piece a different fields. Thus, as 'plant' i consideredto be more complex thin plant, and artefacts described 'machines' in others' certain fields may be described as is u A machine consistsof assembliesand components.control equipment compon( in olant and machinesalike and may be made up of assembliesand re terms of these uses various The machines. small of ani perhaps even historical develoPments. Hubka [2.10] has drawn up a comprehensivelist of possibleclassifications technicalirteficts basedon such criteria as function, solution principle, on plexity, manufacture,product etc' It is., however' impossibleto agree fort and tasks,applications i"n"ruity acceptable,yrt"- of classification-the much to be said for Hubk is there Hence complex. and varied too ir" *u.-h to t suggestionthat technicalartefactsshouldbe treated assystemsconnected i divided be by means of inputs and outputs' A systemcan sys the "niironr',.nt what belongsto a particular systemis determined-by sub-systems. t boun'dary.The inputs anJ outputs crossthe systemboundary(l'2'3)' With l: . -SvslelHoulOarv l . S y s t e m :' C o u p l i n g ' / r s r s t e r ne l e m e n t s r; . . / c o n n e c t i n ge l e m e n t s S ; o v e r a l ls y s t e m S ; , subsystem r l ' l , r ' o u p l i n g ' ;. t 2 s u b s y s t e m' c l u t c h ' l 1 i n p u t s ; O o u t p u t s 'llr,'s',51cm d e p i c t e di n F i g u r e 2 . 1 i s b a s e do n i t s m e c h a n i c a cl o n s t r u c t i o n .I t ftrrrrs'c r'. equally possibleto considerit in terms of its functions (see2.1.3). In r;rrt'. the total system'coupling'can be split up into the sub-systems 'rrrg' lrnd 'clutching'; the second sub-systeminto the further sub-systems 'transferring 1,rrrqclutch operating force into normal force' and torque'. llrc svstcrnelement g could equally well be treated as a sub-systemwhose r{rn rt i\ to convert the actuating force into a larger normal force actins on llrttion strrfaces. r('n(lir)gon their use, any number of such subdivisionsmay be made. The 't lrirs to cstablish particular systemsfor particular purposes, and must lltt'ir virrious inputs and outputs and fix their boundaries.In doing this he w h r r t t er n r i n o l o g yh c l i k e s o r i s c u s t o m a r yi n h i s p a r t i c u l a rf i e l d . upprourhitispossibletodefineappropriatesystemsateverystageofabstr parts ilf la tion, analysis or classification. As a rule such systems are superiorsystems. ___J 5 (lottversion of energy,materialand signals ) t t t t l c r st t t i r l t c ri t t r t t i r t t ys l t t t p c su n d l i t r r t t s .l t s r r i r l t r r i rtli l r r n . < l rt h c l i l r n t lmprcsscdu;roltit. ;rrovidcrhim with infitrmutiorrirlroutits possiblcuscs. I r r r r d a m e n t a l so f e n g i n e e r i n gs y s t e m s 2 Funclamentals Z3 22 is a primary source of information Matter without form is inconceivable-form concept of force the development ;f physics'-the about the state of matte'' Wlttt the means by being Force was ton'eiued as became increasingly important' was explained process this ;;;",;; ;;, .t ung"a. Ultimately which the motion of energy equivalence "f of relaiivity postulated the. in terms of energy. f'tt" in"o'y offi: [2'30] lists."nergi.'*"ll:i, *,'::"^T1':"':t" andmatter.Weizslcker asa run,n,,rtbeintroduced l|l.S]ll"il [llrrl'"r'ni* rl ir""r""o,"ii*" t::i:1si31 g quantitv.o;iv ;v referencet-ini"'pruv damental ;:: Xt"T informaand1;'ffu1 matter of energv, il l*:, 9::', i:["J:ffi1'J;t#;i:,"jn;;;;' be adequately described' tion ,tTtiJi::;lfil w -^:c:^A x., mqnif'eqt form. form. We itsmanifest by ito is ofrenspecified iln"r" , enersy mechanical,"'i""i'"i"'"t:'.:1 of*"tl"l]iil: oI speak sp:a\ etc' Thr ::".:g,,;l'":''i,fif,l,il"',i *Zigrtt, colour,conditionn:{t; as substitutematerialwith such propertles givJn more concreteexpressionb is'o"n"'ally generalconcept i'i;';;;;ti t::,:if";T:t-::J "f is'itreptrvsic?t of the,"r^,'inoi'hat means f::TtL *l:: betweenpeopleis often calleda exchanged ff#":;[i#"tt";; 12.\rl. svstems-qrunt of technicar I "qlllT::i :1r.1*l5;i.l''T$il il! un",rrls lrrirrirlswithout any flow of material. In many casesenergy has to be specially prol'ided for this purpose; in other caseslatent energy can be drawn upon rrcctly. Every conversion of signalsis associatedwith a conversionof energy, rrrghnot necessarilywith a conversion of material lrr what follows, we shall be dealing with: l rrcrgy: mechanical, thermal, electrical, chemical, optical, nuclear etc lrlso force, current, heat . . . lrlirterial:gas, liquid, solid, dust etc also raw material, test sample, r r o r k p i e c ee t c . . e n d p r o d u c t , c o m p o n e n te t c . Si{niils: magnitude, disptay. control impulse, diSa, information etc . taken into lnr cvery type of proposed conversion, quantity and quality must be taket task, for the choice of sideration if rigorous criteria for the definition of the is fullv No statement rrtions and for an evaluation. are to be established. aspects have been taken quantitative as well as its qualitative lirrcd unless its r l c c o u n t . T h u s . t h e s t a t e m e n t : ' 1 0 0k g / so f s t e a ma t 8 0 b a r a n d 5 0 0 ' C ' i s n o t rrrllicient definition of the input of a steam turbine unlessthere is the further tilication that these figures refer to a nominal quantity of steam and not, for trrnce,to the maximum flow capacityof the turbine, and unlessthe admissible (turrtions of the state of the steam are fixed at, say, 80 bar + 5 bar and t (' + 10oC,that is, extended by a qualitative aspect. lrr very many applications,it is also essentialto stipulate the cost or value of irrputsand/or the maximum permissiblecosts of the outputs (see [2.23]). l o sgtn up: all technical systemsinvolve the conversion of energy, material rl or signalswhich must be defined in quantitative, qualitative and economic r r r r ' ( F i g u r e2 . 2 ) . ilff;#:il#'ffi ."ffi"Jff l';r;T:ilil::::::::::'l'.,'.T'.'"','[:l o]::,:: *n"1' u'"'r' ;;;i' ll:lfi:H;# ,iliJ-.'i#.:;;ffi ".1 :: "'" ;;u:"'i1"rv;:'"iJ"*:'::TI":J3# J;#llilft in"a,varietv ::'*:t :;:T^:il;i;:'; :?"Jf T"Zir)-'r;"T;;il''"danctthermalenergy, convet "-r:iI':a combustion ::;::1ff engine n:H:::il:l :Ii7i;"i;,J"*"".n"r,.a1 ,--- ^l^.-- ^^ir/ar c f q f i n n cconve onvel statlon thermal energy a nuclear power chemical into mechanical and ';;:;:;; so.on' nuclearinto thermalenergy,and ff'ffi il.:k;i ; ; varietY "l Y:I^lh;1;T.?:,'T ,"{':n::',';;:o''il{;;;;'t"J'"-'n"""4"::'nff li"}";"l,"Ti:"'; iil'lll'lI 1." ;; ;'**; o?J I nI't'"dp' l"::': T::?lll"',1i", iJ,li for test ;:ffiilT':,i#r""r; o"fffi ,"."iu"i, "t*are destroved "l surfacefinishesand some Signals a informationin the form of signals. ot"nt must process with others' transmitted, disPla prepared, comparetl or combined '" eot,.conver1i1J-"*l3s;;,1"::1111 ::'X".1''?J"1 ilJ.l;,",,onetvp #'iIn I j"p"nding.on piobrem th.e ::.]:,:**ton' .i;;".:l':lit.l"l",Tl";;r,",r, il;;i; ; il 1;""tJa.1 tl", T,il: r*,"iji'.I;' nversron convelsrurr heco case, case, tthe three cc frequently alllil:":'J type of converston, and quite accompanied by a second into play. rhus there-cu'"''u3no ::^I:T,:::::i?1""" small' ."t""^;tn or signalswithout of.energy' howe^ver I t:t:, rsionof'n"igvisoftJnT'o:i?.t:9 ll rheconve wit l]t,1 :::l:""':#i.trf compared nuclear, a (asin Hi:i-t.;;vl" "ir'rr" o -.ont,or l"i''*" ""H:r'r,l-n;'-ilil *"li;i^ il" power cnt coal-rirecr, thc:"; or1.;3 ::1::"1:l:j il flli::':i ;::#;iltttg flilt',Tf; 3,'iliJ,iii'i;"";i;";; otiltli"t"u"r' i," andregu'ation rcccivc' trltttsttlrltr numerousmeasuringinstrumcnts I rr,rrr2 - e. 2 . T h e c o n v e r s i o no f , r r ' rq v . m a t e r i a la n d s i g n a l s . \ , , l u t i o n n o t y e t k n o w n ; t a s ko r l r r r rtLi o n d e s c r i b e do n t h e b a s i so f r r r l ) l l 1 \ . n od u t p u t s Energy+ Material s ---Signa + Energy' ----c- lVlaterial --.- Signa s' The functional interrelationship ,'rrlcr to solve a technical problem we need a svstem with a clear and easily r.t1119".1 relationship between inputs and outputs. In the case of material \ ( rsions, for instance,we require identical outputs for identical inputs. Also, lrr r'r'n thc bcginning and the end of a process,for instancefilling a tank, there lrc ir clcrrr anclrcprocluciblerelationship.Such relationshipsmust alwaysbe n r r c r -l t h u t i s . c l c s i g n c ct ol n l c c t z rs p c c i f i c a t i o nF. o r t h e p u r p o s eo f d e s c r i b i n g s t r l v i r r gc l c s i g np r o t r l c n r s i.t i s u s c l u l l < li r p p l yt h c t c r m . f i l r < ' t i o nI < 'tth e g e n e r a l r l t o s cl ) u r l ) ( ) s ict i s t o p c r f o r m a t a s k . l l ( ) u t l ) u tr c l i r t i t l n s h i op l i r s v s t c r rw l o r s t l r t i c[ ) t ( ) c c s s ci sl i s c r r o t r g l lt o t l c l c t r t t i r t ct l t c i n l t t r t si r t t r lo u t l t u t s ;l o r l ( ' ( . s s clsl r i r t c l r i r r r g ew i t l r l i r t t c ( t l y l t l n t i c p f r l c c s s c s )t.h c t l r s k t t t t t s tb c t l c l i n c t l t l r e l b v i r d e s c r i p t i o rorl ' t l r c i r r i t i n l u n df i n u l m u g n i t u t l c sA, t t h i ss t i r g ct l t c r ei s a1 2 Fundamentals no need to stipulate what solution will satisfythis kind of function. The function thus becomesan abstractformulation of the task, independent of any particular solution. If the overall task has been adequately defined-that is, if the inputs and outputs of all the quantities involved and their actual or required properties are known-then it is possible to specify the overall function. An overall function can often be divided directly into identifiable subfunctions corresponding to sub-tasks. The relationship between sub-functions and overall function is very often governed by certain constraints,inasmuch as some sub-functionshave to be satisfied before others. On the other hand it is usually possible to link sub-functionsin various ways and hence to create variants. In all such cases,the links must be compatible. The meaningful and compatible combination of sub-functionsinto an overall function produces a so-calledfunction structure,which may be varied to satisfy the overall function. To that end it is useful to make a block diagram in which the processesand sub-systemsinside a given block (black box) are at first ignored (Figure 2.2). Functions are usually defined by statementsconsistingof a verb and a noun, f o r e x a m p l e ' i n c r e a s ep r e s s u r e ' , ' t r a n s f e rt o r q u e ' o r ' r e d u c e s p e e d ' .T h e y a r e derived from the conversionsof energy, material and signalsdiscussedin2.1.2. So far as is possible, all these data should be accompaniedwith specificationsof the physical quantities. In most engineering applications, a combination of all three types of convers i o n i s u s u a l l y i n v o l v e d , w i t h t h e c o n v e r s i o ne i t h e r o f m a t e r i a l o r o f e n e r g y influencing the function structure decisively. It is useful to distinguish between main and auxiliary functions. Whtle main functions are those sub-functions that serve the overall function directly, auxiliary functions are those that contribute to it indirectly. They have a supportive or complementary character and are often determined by the nature of the sofution. These definitions are derived from value analysis [2.4, 2.28, 2.29]1andare not identical for all levels of approach. While it may not always be possibleto make a clear distinction between main and auxiliary functions, the tcrms are. nevertheless.useful. It is also important to examine the relationship between the various subfunctions, and to pay particular attention to their logical sequenceor necessary interconnection. As an example, consider the packing of carpet squares, stamped out of a length of carpet. The first task is to introduce a method of control so that the perfect squarescan be selected,counted and packed in specifiedlots. The main flow here is that of material shown in the form of a block diagram in Figure 2.3. On closer examination we discover that this chain of sub-functionsrequires the introduction of auxiliary functions because: -the stamping-out processcreates offcuts that have to be removecl; -rejects must be removed separately and reprocessecl;ancl -packing material must be brought in. 25 , ' i I u n c l a m e n t a l so f e n g i n e e r i n gs y s t e m s ,' *ifffiTfl* 3i;ffi1'0"' ili1ii.ii,'o,o L00secarpel len0in + lTF-ffi l::':'-'r'-'-'^'-"' rJiJ n l0ls l\,'laterial f 0w l---l lVainlunctron --- System boundary 1.3. Function structurefor the packing ofcarpet squares 'l lrc result is the function structure shown in Fieure 2.4.It will be seenthat the lrrnction'count squares'can also give the signalto pack the squaresinto lots ruspccified size. -1 irernp lr0m L'1r(lln I .==+ lvaterlal i---l Marn{unctlon flow --- Signal flow function Auxiliary i_-_-_i - - - S y s t e mb o u n d a r y ' l. l:unction structurefor the packing of carpet squaresas in Figure 2.3 with r l r r r r c t i o nas d d e d llr,,, l'hirus [2.19] has defined functions in general as activities,effects, goals r orrrtrrintq. In mathematics, a function is the associationof a magnitude y ,r rrrrunituclcx so that a unique value (single-valuedfunction) or more than r . r l t r c( r n u l t i - v a l u e df u n c t i o n ) o f y i s a s s i g n e d f o r e v e r y v a l u eo f x . A c c o r d i n g flrc rlclinilion givcn in DIN69910 l2.al, all functions are determined by r v c s( t i r s k s ,u c t i v i t i e sc, h a r a c t e r i s t i c s ) . r t i u u s r v r i t c r s o n c l c s i g nr n c t h o c l s( s c c 1 . 2 . 2 ) h a v e put forward wider or t t l c l i r r i t i r l n os l g c r t c r l r l l yi r p p l i c i r b l lcu n c l i o n s( s c c -s3) l l t e o r y . i t i s l l o s s i t r l ct o c l i t s s i l ' yI t t t t e l i o t t ss o t l t l t l t h c l o w c s t l c v e l o f t h e s l r t t c l t t r c c o t t s i s t sc x c l t t s i v c l Yo l ' I t t l t c l i o t t sl l t t r t c i u n r o t b c s u b - d i v i c l c c l gcncrlllyupplicrblc, r wlrilc rcrrrairring 2 Fundamentals 2l | , r r . l . r n r e n l a ol sl c n g i n c e r i n S gy\tems Rodenacker [2.23] has defined functions in terms of two-valued logic, Roth 12.251interms of their general applicability, and Koller 12.12,2.131in terms of the required physical effects. Krumhauer [2.14] has examined generalfunctions in the light of possible computer applications during the conceptual design phase, paying special attention to the relationship between inputs and outputs after changes in type, magnitude, number, place and time. By and large, he 'change' he refers arrives at the same functions as Roth, except that by 'increase or exclusively to changes in the type of input and output, while by decrease'he refers exclusivelyto changesin magnitude. As Figure 2.5 shows, all these definitions are compatible if it is remembered 'connect' and'separate'to refer to the logical connections that Rodenacker uses only. cl .i ffi l- a lol lF r;l t"it 2.1.4 The physical interrelationship # c.) cl N + ii I a.l Establishinga function structurefacilitatesthe discoveryof solutionsbecausei simplifies the general search for them and also because solutions to functions can be elaborated separatelv. ci Individualsub-functions, by blackboxes,mustnow originallyrepresented replacedwith more concretestatements.Sub-functions are usuallyfulfilled physical processes-nearly all engineering solutions are based on physi 6 phenomena. In addition, of course, chemical and biological phenomena m also be involved, but they are relatively few and far between in our chosenfield. If, in what follows, we refer to physical processes,we tacitly include the effects possiblechemical or biological processes. Physical processes are based on physical effects. A physical effect can described quantitatively by means of the physical laws governing the physica quantities involved. Thus the friction effect is described by Coulomb's law Fs 'b; and the expansi F p : 1 z F N tl h e l e v e r e f f e c t b y t h e l e v e r l a w F a ' o : R o d e nacker [2.23] and Kolle effect by the expansionlaw ll: ct'l'A0. particular, have collated such effects. [2.I2], in Several phvsical effects mav have to be combined in order to fulfil sub-function. Thus the operation of a bimetal strip is the result of a combina of two effects, namely thermal expansion and elasticity. If, in concrete cases these effects are assigned to a sub-function, we obtain the physical principle E z O cd I-:TJ frilE O ,- that sub-function. Figure2.6 illustratesthe stages-sub-function,physicaleffect.physicalpri ple and solutionprinciple(see2.1.5)-of three sub-functions(only the inp and outputs of the main flow are shown): -transfer torque by the friction effect in accordance with Coulomb's law -amplify force by the lever effect in accordance with the lcvcr law; a - make electrical contact by bridging the gap by means of thc cxpansion cf i n a c c o r d a n c ew i t h t h e l a w o f l i n c a r c x p a n s i o no f s o l i c l so r l i c l u i c l s . A s u b - f u n c t i o nc a n o f t c n b c l t r l l ' i l l c t lt r y v a r i o u sp h y s i c i r cl l ' l ' c c t s , ' l ' h uisr l i can bc arrrplil'icd by thc levcr cl'l'cct, ihc wedgc cfl'cct. thc clcclro-ntlgnc c '(' ,', iu 9i i'; r{ i,i ,rS "gt ,,, (" 'i ii s,b ,"r' ,! h i'' j! i :' ?, i.i O al nr i\ 0 .!b ry) !;; 28 2 Fundamentals Sub-f unction principle principle Solution Physical Physical eflect (independent ol solution) (Subfunction and physical etlect) principle (Physical and leatures) formdesign Friction ft:pFn LeVCI FA ,F^ b FA Amplily muscular lorceby lever rs: rei Fa'a: Fs'b Expansr on q Zl=u I ZS by contact Close nsol expanding ry mercu ZI=a l-ZS Figure2.6.Fulfillingsub-functions by solutionprinciplesbuilt up of physicalprinciples andform designfeatures :ffect, the hydraulic effect etc. The physical principle found to satisfy a efie part rarticular sub-function must, however, be compatible with the physical principles rles of other, associatedsub-functions. A hydraulic amplifier, for instance, :annot be directly powered by an electric battery. Moreover, a given physical canr prin rrinciple will fully satisfy a given sub-function under certain conditions only. Thu fhus a pneumatic control system will be superior to a mechanical or electrical :ontrol system only in particular circumstances. conl Compatibility and optimum fulfilment cannot generally be assessedexcept in hc context of the overall function, and then only during its concrete embodi'l'o ttcrtt. that cnd, the required layout and final forms have to be specified. 21.1.5 . 1 The form interrelationship The fhe function is satisfied by the application of the solution principle, which ealised by the arrangement of surfaces (or spaces) and the choice of motic real [ 2.23). 2.2 The surfacesare varied in respect of, and determined by: - Type - Shape * Position - Size *Number [2.24). 29 I I u l t d a m e n t a l s o f e n g i n e e r i n gs y s t e m s S i r rilarly r i l the requisite motions (kinematics) are determined by: )e I r'1.1, translation-rotation N l r ture t regular-irregular l ) iircction rc in x, y, z-directions and/or about x, y, z-axes rrgnitude velocity etc N{rtg , N trrmnb e r o n e , s e v e r a le t c it< we need a general idea of the type of material with which the l n ircldition, ccs are to be produced, for example, whether it is solid, liquid or gaseous; fircc ftl or flexible; elasticor plastic; stiff, hard or tough; or corrosion-resistant.A rirl idea of the final form is often insufficient; the properties of the materials !criri tr specified before an adequate formulation of the requisite form design be fttt rc undertaken. trc I p r r lIythe combination of the physical principle with the main form design !urc:s (surfaces,motions and materials) allows the principle of the solution to . This combination is called the solution orinciple. and it is the first fcrctte step in the implementation of the solution. In l';igure 2.6 the examples discussedin 2.I.4 have been converted into ,n principles by the addition of certain form design features. futi.,r p'rirr rrsferring the torque by friction against a cylindrical surface in accordance l v i r l r Coulomb's law will, depending on the way in which the normal force is lead to the selection of a shrink fit or a clamp connection as the I l t P licd, l p rinciple. r t i o n &rl r' plifying muscular force with the help of a lever in accordancewith the Arrtl 'r r law .' after determining the pivot and force application points (working It' t t r rl ;rces) ; and consideringthe necessarymotions will lead to a description of : principle (lever solution, eccentricsolution etc). l l t t ' solution N ; r liing li electric contact by bridging a gap using the expansioneffect, applied l l l r (t'cordancewith the linear expansionlaw, only leads to an overall solution ( after determination of the size and position of Ihe surfaces needed l r i l lrt'iple fot Itltc motion of the expanding medium , a material (mercury) expanding by a t irmount and serving as a switch. I t r . ' tl .rrtisfy the overall function, the solution principles of the various subli' ., h ' l t t trrrshave to be combined. There are obviously severalways in which this , Guideline VDI 222212.211calls each combination acombination of I lrr' rltrrrc. tlcs. r1, )n, ln ",rilny cases, a combination of solution principles must be given more [ ' r c t c cxpressionbefore it can be evaluated.This involves more definite ideas hc nratcrialsto be used, a preliminary dimensionedlayout and a technical ity stucly.As a rule it is not until then that one obtains a solution concept :irn lrc cvaluated in the light of the objectives and the actual constraints 'l'lrc solution concept is thc funclamentalproposal of a solution satisfying .6) o v c r i r l l l u n c t i o n a n c lh o l c l i n g< l u t t h c p r o r n i s ct h a t t h e t a s k m a y b e r e a l i s e d . l.''r,. I Irxl. scvcrll conccptvariantsarc possiblc. *.giiEii- 30 2 Fundamen 2.1.6 General objectivesand constraints The solution of technical tasks is determined bv the seneral obiectives a constraints. The fulfilment of the technical function, the attainment of economic feasibili and the observance of safety requirements may be considered as genera objectives. The fulfilment of the technicalfunction alone does not complete t designer's task; it would simply be an end in itself. Economic feasibility i another essential requirement, and concern with human and environment safety must impose itself for ethical reasons.Every one of these objectives h direct repercussionson the rest. In addition, the solution of technical problems imposes certain constra or requirements resulting from ergonomics, production methods, tra facilities, intended operation etc, no matter whether these constraintsare t result of the particular task or the general state of technology. In the fi case we speak of task-specific constraints, in the second of general straints that, though not specified explicitly, must neverthelessbe taken i account. Hubka [2.10] separatesthe properties affectedby the constraintsinto catego ies based variously on industrial, ergonomic, aesthetic, distribution, deliver planning, design, production and economic factors. Besides satisfying the functional, physical and form interrelationships, solution must also satisfy certain general or task-specificconstraints.These be classifiedunder the following headings: - Safety alsoin the wider senseof reliability - E,rgonomics the man-machine context - Production type of manufactureand facilitiesfor the production parts - Quality control at any point during the manufacturing process - Assembly during and after manufacture 'f ransport inside and outside the factory Opcration intended use, handling Mirintcnuncc u p k e e p , i n s p e c t i o na n d r e p a i r lrxpcncliturc c o s t sa n d s c h e d u l e s 'l'hc corrstraints thnt can be derived from these characteristicsaffect t I u n c t i o r . rt,h c w o r k i n g p r i n c i p l e a n d t h e f o r m d e s i g n , a n d a l s o i n f l u e n c e anothcr. I lcncc they should be treated as checkpoints throughout the desi process.and adapted to each level of embodiment. It is advisable to consider them even during the conceptualphase, at least essence.During the embodiment phase, when the layout and form design of t more or less qualitatively elaborated concept is first quantified, both t objectives of the task and also the general and task-specificconstraintsmust considered in concrete detail. This involves several steos-the collcction further information, layout and form design, and the elimination of wcirk lin t o g e t h e r w i t h a f r e s h , i f l i m i t e d , s e a r c ho f s o l u t i o n sf o r a v i r r i c t y o l ' s r r b - l i r 31 I : u n d a m e n t a l so f t h e s y s t e m a t i ca p p r o a c h rrrlr-functions), until, finally, in the detail phase the elaboration of detail ;nvingsand production documents brings the design processto a conclusion. 2 Fundamentals of the systematicapproach 2.1 General working method 'lirre we deal with the specific steps and rules of systematic design, we must cliscussa number of general principles. These come from a host of different 'iplines, including non-technical ones, and are usually built on inter-disciplinv lundamentals. Management science,psychotogyand philosophy have been ong the main inspirations, which is not suprising when we consider that thods designed to improve working procedures impinge on the qualities, rrrcitiesand limitations of human thought. I lre following conditions must be satisfied by anyone using a systematic rr.oach: l tr.sr,tre the requisite motivation for the solution of the task, for instance by tliscussionof the objectivesand of the significanceof the entire project and by rlcneral intellectual stimulation. {'larfy the boundary conditions, that is, define the initial and marginal t onstraints. I )ispelprejudice to ensure the most wide-rangingpossiblesearchfor solutions rrrrclto avoid logical errors. I ttok for variants,that is, find a number of possiblesolutionsfrom which the lrt'st can be selected. \ltke decisions.This is facilitated by objective evaluations.without decisions t l r c r ec a n b e n o p r o g r e s s . I lrc following proceduresare based not only on our own professional rl', ricnce,but alsoand aboveall, on the work of Holliger12.8,2.91, Nadler ' 1t.2.113]and Mtiller when used as intellectual tools in the systematic [2.16]. ',rri n for iolutions ;; ;;;"i;;; ;;.ly and effectivethoughrthey are also "r.l ( t i heuristic ril\\ n r,,\\ l l as lIgUrlSllu principles. They underlymostsystematic procedures and are PIII Lv r) rLLrrrorr! , l , l i er r b l ei n a l l f i e l d s . I rrltritiveand discursivethought Itrrrritivc thought involves sudden ideas (flashes of inspiration) and cannot ' r r n r r l l yb c p r o c l u c c ctlo o r c l e r . A s a r u l e . i n t u i t i v e t h o u g h t p r o c e s s e si n v o l v e t rl v c o n t p l c x i t s s o c i i t t i o n so f i c l c i r s ,c l a b < l r i r t c ci ln t h e s u b c o n s c i o u sm i n d . I t t r t r g litn l t r i t i o r rh l r sl c c ll o i r l i r r g cr t r r r n b cor l g o o r l l r r r ccl v c n c x c c l l c n ts o l u t i c l n s p t r r c l vi r r t r r i l i v ci r p p r o l r c lhr t r st l r c l i r l k r u , i r rtgl i s l r t l v i r n t i r g c s : l l t c r i g l t l i t l c t rr i t r c l vc ( ) n l e sa t t h c r i g h l n l ( ) n r c nsl i r r c r ' i lc i u l t . t (b) tc c l i c i t c c il r l will: 32 2 Fundamentals - the resultdependsstronglyon individualtalent and experience;and - thereis a dangerthat solutionswill be circumscribed by one's specialtraini and experience. It is therefore advisable to use more deliberate procedures that tackl problems step by step, and such procedures are called discursive. Here the ster are chosen intentionally; they can be influenced and communicated. It is lmportant aspect of this procedure that a problem is rarely tackled as a whole but is first divided into manageableparts and then analysed. It must, however, be stressedthat the intuitive and discursivemethods are opposites.Experiencehas shown that intuition is stimulatedby discursi, thought.Thus while complexassignments must alwaysbe tackledone srepar time, the subsidiary problems involved may, and often should, be solved intuitive ways. In systematic work it is helpful to exploit certain general characteristics human thought. Holliger [2.9] distinguishesbetween unconscious,preconsci< and consciousthought and prescribesthe transformation of aimlessand uncon scious procedures and of disorderly and fantasy-charged preconscious proce dures into a conscious or deliberate approach. This can-be dbne with the help o methodical rules, clear task formulation and a structured procedure. A furtheiair to conscious thought is the assoclation of ideas. one should, however, avoid se complexes of ideas becausethese may turn out to be too inflexible, and complexes shouldbe deliberatelydissolved. It is obviousthat systematic thou is neededmore for originaldesignthanfor routinetasks,whichcanqenerallv performed successfullyeven if the underlying thought processesremain uncon scious. Another important property of human thought is the inevitability o errors, for which allowancesshould, if possible,be made from the start. In thi connection,Holliger speaksof 'catastrophe analysis'.one should,however, carefulto minimiseerrorsor the weak links resultingfrom them. This can doneby: -clearly definingthe requirementsand constraints of a particulartask; -not forcingintuitivesolutionsbut usinga discursive approach; -avoiding fixed ideas;and -adapting methods,proceduresand technicalaids to the task in hand. 2 Analysis Analysisis the resolutionof anythingcomplexinto its elementsand the study theseelementsand of their interrelationships.It calls for identification,deii tion. structuringand arrangement. If errorsare to be minimised,then problemsmust be formulatedclearlya unambiguously.To that end, they have to be analysed.problem analysismez separating the essential from the inessential and, in the .ur" of .o. problems,preparinga discursivesolutionby resolutioninto individual. transparent, subsidiaryproblems.If the searchfor the solutignprovcs6ifficult, reformulationof the problem may provc hclpful. Expericncehas shown thi u n d a m e n t a l so f t h e s y s t e m a t i ca p p r o a c h JJ clul analysis and formulation of problems are among the most important rs of the systematicapproach. 'llrc solution of a problem can also be brought nearer by structureanalysis, is, the searchfor hierarchicalstructuresor logicalconnections.In general, type of analysiscan be said to aim at the demonstrationof similaritiesor :titive features in different systems(see 5.4). rnther helpful approach is weak link analysis. It is based on the fact that causedby ignorance,mistakenideas,external ry systemhas weaknesses ances, physical limitations and manufacturingerrors. During the delment of a system it is therefore important to analyse the design concept or embodimentfor the expresspurposeof discoveringpossibleweak links prescribing the remedies. To that end special evaluation procedures (see and weak tink identification methods (see 6.6) have been developed. rience has shown that this type of analysis may not only lead to specific ovementsof the chosensolution principle, but also may trigger off new ion principles. nthesis r'.llsis the putting together of parts or elements to produce new effects and dcrnonstrate that these effects create an overall order. It involves search and rvcry, and also composition and combination. An essentialfeature of all rt work is the combination of individual findinss or sub-solutionsinto an rrllworkingsystem-that is, the association of components to form a whole. irrs the processof synthesisthe information discoveredby analysesis 'csscdas well. In general,it is advisableto basesynthesison global a or rtt.sapproach; in other words to bear in mind the general task or course of rtswhile workingon sub-tasks or individualsteps.Unlessthisis done,there lht' grave risk that, despite the optimisation of individual assembliesor steps, rtritirbleoverall solution will be reached.Appreciation of this fact is the basis f lrc inter-disciplinary development known as value analysis which proceeds thc analysis of the problem and function structure to a global approach ving the early collaboration of all departments concerned. A global orrch is also needed in large-scaleprojects, and especially in preparing tltrlcs by such techniques as Critical Path Analysis. The entire systems otl is strongly basedon the global approach,which is particularly important lltt' cvaluation of solution proposalsbecausethe value of a particular solution ortly be gauged after overall assessmentof all the requirements and r i r i n t s( s e e5 . t t ) . rn of labour and collaboration csscntial finding of management scicnccis that the implementation of large cttnrplcx tasks calls for the division of labour, the more so as specialisation 'I'his of modern is alsodcmandcdby tltc incrcaringlytight schcdulcs . Now, divisionof labou |nrg!{iFdlrciplinary f- which, collaboration 34 2 Fundament 35 I I r r r r c l a m e n t a losf t h e s y s t e m a t i ca p p r o a c h in its turn, involves special organisational and staff arrangements and attitude including individual receptiveness to the ideas of others. It must, however, t stressed that inter-disciplinary collaboration and teamwork also demand rigorous allocation of responsibility. Thus the product manager should be in sol charge of the development of a particular product, regardless of department boundaries. shattto hub Task: Connect andhub Shaft lnitral situation. 5 Generally applicable methods The following general methods provide further support for systematicwork, are widely used [2.9]. The method of persistentquestions (The 'Why?' technique) When using systematicprocedures it is often a good idea to keep ask questionsas a stimulus to fresh thought and intuition. A standard list questionsalsofostersthe discursive method.In short,askingquestions is one o the most important methodological tools. This explains why many authors ha drawn up special questionnaires(checklists). rc 2.7. Develooment of shaft-hub connectionsin accordancewith the method of rvlrrdsteps The method of negation bc made to turn this purely theoretical and ideal systeminto a technologilcasible one, and finally into one that meets all the concreterequirements. rrtunately, it is rarely possible to specify in advance which particular ideal rrr will satisfy all functions, especiallythose linked together in a complex The method of deliberate negation starts from a known solution, splits it in individual parts or describes it by individual statements, and negates the statements one by one or in groups.This deliberateinversionoftencreatesne solutionpossibilities. Thus,when considering a 'rotating'machineelement might also examine the 'static' case. Moreover, the mere omissionof an elemen can be tantamount to a negation. This procedure is also known as 'systemati doubting'[2.9]. Themethodof forward steps Starting from a first solution attempt, one follows as many paths as yieldingfurther solutions.This method is also called the method of diverse thought.It is not necessarily svstematic. but frequentlystartswith an unsystem tic divergence of ideas. The method is illustratedin Figure2.7. Themethodof backwardsteps Starting from the objectivesof the development,one retracesall the poss paths that may have led up to it. This method is also called the melhod convergent thought, because only such ideas are developed as converge on ultimate goal. The method is particularly useful for drawing up production plans an developingsystemsfor the manufactureof designedcomponents. It is similarto the methodof Nadler 12.17l,whohasproposedthe constructi of an ideal systemthat will satisfyall demands.This systemis not developed practicebut formulatedin the mind. It demandsoptimum conditionssuchas ideal environment which causes no external disturbances.Having formul sucha system,there follows a step-by-step investigationof whal concc rtrttltod of systematicvariation of the solutionare known, it is possible,by requiredcharacteristics variation,to developa more or lesscompletesolutionfield. This rlvcs the construction of a generalisedclassification,that is, a schematic 'rcntation of the various characteristicsand possible solutions (see 5.4.3). rr lhc viewpoint of managementscience,too, it is obvious that the discovery lrrlrrtions is assistedby the construction and use of classificationschemes. Ir rrll authors consider systematic variation one of the most important r t l i c a lp r o c e d u r e s . 2 I'roblem solving as information conversion ation conversion we cliscussed the basic ideas of the systemsapproach (1 .2.3) we found that a constant flow of information. Information is solvinq demands rrr (see Figure 2.8). and transmitted proccssed , analyses,trend studies,patents,technical markct frctm is rcceivetl nrirtion I , t { . ' l ' l t c c o t t v er s i r l t tr l l ' rn ,Eft- 2 Fundamen journals, research,licenses,inquiriesfrom customers,concreteassignment design catalogues, analyses of natural and artificial systems, calculations experiments, analogies,general and in-house standards and regulations, stoc sheets,deliveryinstructions,computerdata,test reports,accidentreports,a also through 'asking questions'.Data collectionis an essentialelement problemsolving[2.1]. Information is processedby analysisand synthesis,the development solutionconcepts,calculation,experiment,the elaborationof layout drawin and also the evaluation of solutions. Information is transmittedby means of drawings, reports, production do ments etc. Quite often provision must also be made for the information to stored. Information conversion is usually a very complicated process. Thus, solution of various problems requires information of different type, content a range. Beyond that, in order to raise the level of information and improve it, mav be necessarvto reiterate certain steps. To meet the growing demand for an optimal and rational flow of informati inside an enterprise, and also about its dealingswith the market, severalspeci procedures have been developed in recent years. Zimmermann [2.32] h published a comprehensiveanalysisof these procedures,basedon 74 thesesa 206 bibliographicalentries. What matters above all is, by organisation measuresand the appropriate techniques,to establisha quick and adequate of information between the various departments working on a particular To that end various models have been developedfor processingwritten or informationto satisfya variety of needs12.221. It is understandable that t researchwork should have been done initially for management systems. recently this type of researchhas spread to engineeringsystems[2.15], since it now recognised that technical developments in a particular industry depen largely on the efficiency and range of its information system. Useful criteria for evaluating the quality of information will be found in They include: -Reliability, that is, the probabilityof the informationbeingtrustworthyan correct. -Sharpness, that is, the precision and clarity of the information content. -Volume and density, that is, the indication of the number of words a pictures needed for the description of a system or process. -Value, that is, the importance of the information to the recipient. - Actuality, that is, an indication of the point in time when the information be used. -Form, that is, the distinction between graphic and alpha-numericdata. - Originality, that is, an indication of whether or not the original character the infbrmation must be preserved. -Complexity, that is, the structure of, or sirnilaritybctwccn, inlirrntirt s y m b o l sa n d i n f o r m a t i o n c l c r n c r t l s .u r r i t sr l r c o r n l l l c x c s , - l ) c g r c c o f r c f ' i n c r n c r ttl l.t i r t i s . l l t c t l r r i r r r l i tovl d c l i r i l i r r t l r c i n l i r n r r i r t i o n . I ' rrtllutrcntals 31 lrrlilrmation systems lrrrild up an information system, one may have to take into account, aparl rr the above criteria, the position of the user (for instance section leader, l n c r , o r d r a u g h t s m a n ) t; h e d e s i g np h a s e( c o n c e p t u a l ,e m b o d i m e n t ,d e t a i l ) ; t r p c o f d e s i g n ( o r i g i n a l d e s i g n , a d a p t i v ed e s i g n , v a r i a n t d e s i g n ) ; a n d t h e plcxity of the system to be developed (for instance plant, machine, :rnbly, component). The following stepsin the constructionof an information 'nr can be distinguished: :tcrmination of the requirements; cntification of the sources; rllcction: rrssificationand processing; )l llge; : t ri c v a l ; a n d rrnputerisation,if necessary. rcral information systems have already proved their practical usefulness. : rvc shall merely list a few important bibliographicalsourceson information d a t as y s t e m[s2 . 5 ,2 . 1 5 , 2 . 2 0 , i c a t i o n[ 2 . 6 , 2 . 2 I , 2 . 3 4 ]a n do n c o m p l e t e , .1.33].The most important conceptsof information theory are covered in .l.l 300 and DIN 44 301 12.2,2.3). t'onglusien,it must be stressedthat the ready availability of a wide range of llrellcnsive and problem-oriented information is of the utmost importance in tflrisn process12.261.The optimum and rational processingof information is tlr lrrcilitated b y t h e s y s t e m a t i ca p p r o a c h . 39 I r i er r c r a l p r o b l e m - s o l v i n g 3 The designprocess __-: lo to t€ t4 IE In the previous two chapterswe examinedthe fundamentalson which design work should be to best advantage.They form the basisof a systemaic .built approachwhich the practisingdesignercan foilow, regardlessof specialty, and involve a variety of methods,someof which havestilf to be described. r= l= ta - --- t O t d I I I ____J 3.1 General problem solving l . l . G e n e r a lp r o b l e m - s o l v i n gp r o c e d u r e meth,od invorves stepbv step ):,::::*Tt "l:::-?!"",:',:y.: it, weli:?1'T-:.rving proceed from tie quatitati;,;;;;,";;",',l,oi,'i!, :::!::.::i^yi,!,,:::r: each new step being more concrete than the last. lj,r:q:"*-* ::I 1."-:r"ydered as theconversion of information(see2.2.2). ru.fi" e.iai,r'ffi ;; J;., : n"r".nroi"'utiJ,il;;;;;;Ji;";;;;,";,'i Thigher lrt::: ::* .1^y,', 11 "I it.9".o "p " is,"1, to repear at a !v rvrrur4re"lfti; I :"r^,1"t"r,'that the necessary improvement virriant is selected. Because each step of the design process must be tcd, evaluation servesas a check on progresstowards the objective. :isionsinvolve the following considerations(see Figure 3.2): lhc results of the previous step meet the objective, the next step can be are incompatible with the objective the next step may not be has beei made. conversionof informationprovidesdata not only for the next step, however small, but also throws fresh iight "^lt"i:"r*l:"".1y on tr," fr"uious one. The splitting of the designprocessinto stepsensures . that the essentiallinks be^tween objectives,planniig, implementation andchecking are maintained [ 3 . 1 , Repeat thestep 0 na n r g n e r information level 3.e]. With theselinks we can,following Krick or::j:T:: [3.2] andpenny [3.5], generar sotutiori orprourem'sirir"r" llr). f:1.1. Every task invotves-,first.of atl, a confro";";;;;;f il;;;;;iem construct a with what is already known. The intensity of this confrontation depends on the designer,s knowledge, ability and expeiience, and on the particular field inr r which vt rrrvrr rtu he tis J however, more-detailediniormarion abourihetaskitself, lliii"i;^t"^1t]:T:r; about possibtesolutionprinciplesand about known 3];.1,,^ll.^.::,lljraints, olems .rrv :?,lJ:::'Tl-il 0." 140\ rtJgtl, Afolheresu ts In salrslactory l(rrnrs ol lhcobiccl ve? isextre mely "r"rrf "i ll.r;tffi i'i ;#; #.l:; of the requirements. Nextcomes the definition of the essential problems (the crux of the task) on a mo^re.abstract plane, to.ut fix the objectives and chief constraints. Such 1o definitions, far from prejudging the issue, open the way to the untrammclrcd search for solutions, including unconventional ones. lT,::llrl:f is creTtjon,when solutions aredeveloped by variousmc.ns.nd 'y'.1",T,ticauy rr,n."",Ir,;'"';iil;;i'il'1.;;: ji:l"jl1.i,,1.:.i::Tlif".l thcre mustalsr hc evuruution, foll'wcd riy a deci;;itt,, .n thc trrisisof wtrich Ooncrrldesirionprrrcctr ls a repetition ol the steplinancially viable andpromising? 40 3 T h e d e s i g np r o c e s s Flow of work during the designprocess - If repetition of the previous step (or if necessaryof severalprecedingsteps)is financially viable and promises good results, the step must be repeated on a higher information level. - If the answer to the previous question is no, the development must be stopped. Even if the results of a particular step do not meet the objective, they might neverthelessprove useful if the objective or the task were wholly or partly changed. This whole process, leading from confrontation through creation to decision, must be repeated in each successive,increasinglyconcrete, phase of the design process. 3.2 Flow of work during the designprocess II -t E =: Clarify thetask Elaborate thespecilication -9o T I problems ldentily essential Establish function structures principles Search lorsolution Combine andlirmupintoconcept varlants Evaluate against technical andeconomic criteria ll o '-- € =O o The main phasesinvolved are: - Clarification of the task - Conceptual design - Embodiment design - Detail design. Figure 3.3 shows this processstep by step. At every step, a decisionhas to be made as to whether the next step can be taken or whether previous steps have first to be repeated. Continuing right to the end only to discover that a serious mistake has been made at an earlier stage is somethingthat must be avoided at all costs. The obvious decision to stop a development that may not prove cost-effective (3.1) has not been included in the flow diagram. Clarification of the task This phase involves the collection of information about the requirements to be embodied in the solution and also about the constraints. It is followed by the drawing up and elaboration of the detailed specification .2 E O preliminary Develop layouts andl0rmdesrgns Select bestpreliminary layouts Reline andevaluate against technical andeconomic criteria E E E o C . oq E E E o - variants. The conceptual designphase consistsof severalsteps (see 5.1) none which may be skipped if the most promising solution concept is to be reached. I E E .E E 0plmiseandcomplete formdesigns (lhcck f0rerr0rs andcosteffectiveness parts l'rr:pare thepreliminary listandproducti0n documents O I rrr;t t:;t;dt;lails (,ompktlu rJclai drawngs andproduction documents {ilrtr;k ;rlldocLtmenls .6 -6 the subsequentembodimentand detail designphasesit is extremelydifficult impossible to correct fundamental shortcomings of the concept. A successf solution is more likely to spring from the choice of the most appropria principles than from exaggerated concentration on the finer points. This clai does not conflict with the fact that even the best principles may be frustratccl by l a c k o f a t t e n t i o nt o d e t a i l . € E Conceptualdesign The conceptual design phase involves the establishment of function structures g E (requirements list) (see4.2). the searchfor suitable solution principles and their combinationinto co ; = t I Steprof thedcrignprrrcs '...'*l&* 42 3 The design process The concept variants that have been eraborated must now be evaluated. _ variants that do not satisfy the demands of the specification have to be eliminated; the rest must be judged by the methodical application of specific criteria. During this phase, the chief ciiteria are of a technical nature, though rough economic criteria have also begun to play part a (see 5.g). on the basisof the evaluation the best solution concept can now be selected. It mav be that severalconceptvarianis look equally promising, and that a final decisioncan only be reached on a more concrete level. Moreover, various form designsmay satisfy one and the same solution conceot. Embodiment design D u r i n g t h i s p h a s e .t h e d e signer, i starting from the concept, determinesthe lavout and forms, and develops a technicai product o. ,yrt.- i. technical and economic considerations(iee """;;l;;;'#,h [3.7] and[3.g]). It is often necessaryto produce several layouts to ,.ui" simultaneouslyor successivelyin order to obtain more information about the advantagesand disadvantagesof the different variants and thus pave the way for a technicaland economic evaluation. Frequently, the evaluation of individual variants may lead to the selectionof one that looks particularly promising but which may neverthelessbenefit from, and be further improved by, incorporating ideas and solutions from the others. By appropriate combination and tire elimlnation of weak links, the best layout can then be obtained. definitive layout provides a check of function, strength, spatial compati, ..Thut bility etc, and it is also at this stage, at the very latest, that t*hefinancialviability of the project must be assessed. Detail tlesign This is the phase of the design process in which the arrangement, fbrm, dimensions and surface prop".ties of all the individual parts are finally laid ooyn:.r1" m-aterialsspecified,the technical and economic ieasibility re-checked and all the drawings and other production documents p.oau."Jl:.7, 3.g1. It is important that the designershould not relax his vigilance at this stage,lest his ideasand plans be changedout of recognition. It is a mistake to think that the detail design poses subordinate problemJacking in importance or interest. As we said earlier, the difficulties frequently arise from lack of attention to 6etail. Quite often correctionsmustbe mioe ouringthisphaseand the prece<ling steps repeated'not so much with regardto the overallsolution, as for the im-prove- ment of assembliesand components. In the flow diagram (Figuie 3.3), the crucial activities are: - Optimisation of the principle; and - Optimisation of the layout and forms. Theyinfluence each other and, as the figure shows, overlap to a consicrcratrll extent. It is obvious that important production criteria (such as maximunr sizc production mcthods etc) can play a crucial rore c v c n c r u r i n gt h c c ' n c c p t A' I Flow of work during the designprocess +J phase, much as the range of possible materials and the spatial requirements irrl'luence the choice of a particular solution during the embodiment design plurse.In general, however, the optimisation of the form designsand hence of lhc manufacturing processes will begin to assume growing importance as e r r r b o d i m e npt r o c e e d s . 'fhe main phases of the design process cannot always be clearly delimited. 'l hus even a conceptual decision may require a scale drawing for the purpose of :ciding on possiblelayouts. Conversely,the preliminary layout selectedduring hc embodiment design phase may involve nothing more than rough sketches 3.31.Moreover, certain optimisations may be postponed until the detail design itse. Such variations of the design process in no way detract from the value of : general scheme. l;igure 3.3 does not include models and prototypes becausethe information :y supply may be needed at any point in the design process and cannot :refore be fitted into any particular slot. In many cases,models and prototypes vc to be developed even during the conceptualphase, particularly when they : intended to clarify fundamental questionsin, say, the precisionengineering, ctronics and mass production industries. In heavy engineering,on the other rrd, if prototypes are needed at all, they must often be precededby a complete rr through the detail design phase. It rnust also be stressedthat the execution oforders (1.1) need not be part of ' clesign process, especially not in the case of size ranges and modular rrrlucts,where electronic data processingcan help to reduce the work to a tnple consultation of data banks. Apart from general layout drawings and 'rrrbly plans, no further design or drawing work is needed. lrr (iuideline VDI 2222 [3.6) the designprocessis representedby a flow rrrm (Figure 3.4) which we have included here becauseof its visual impact I rrlsofor purposesof comparison. On it, the embodiment designphase is not tlrvicledto the same extent as it is in Figure 3.3 or as we shall be dividing it up ( ' l r i r p t e6r . ( )rr looking at Figure 3.3, and after reading about the methods described in lolkrwing chapters,the practising designermay well object that he lacks the ' to go through every one of the many steps. He should bear in mind that: rrrostof the stepshave to be taken in any case,albeit unconsciously,in which t ' ; r s cr r n f o r s e e nc o n s e q u e n c e m s ay arise: lltc rlcliberate step-by-step procedure, on the other hand, ensures that notltirtg cssentialhas been overlooked or ignored, and is therefore indispenstltlc in thc case of original designs; In tlrc casc of adaptive designs, it is possible to resort to time-tested rachcsancl to reservethe step-by-stepprocedure for speciallypromising $cs I thc dcsigncr is cxpcctccl to pnrducc bctlcr rcsults. then he must be given the rt| timc tlrc systcmirtic approuchdcmands;and hcconrcsmorc occuratcif thc ttcp-by-stcpnrcthodis followccl --*fr*-- 3 Thc dcsignProcess 44 trl Tasks EE a m o -E s4 4 Product planning and clarificationof the task task Selected lunction Overall t0 (Function structure Subjunctions lunction) theoverall meet principles and/orbuilding Solutron forthesub-lunctions blocks prrnciples and/or s0luti0n @ Selected blocks building a E - 6 to ofptinciples Qsrn[tnstions tunction fulliltheoverall - -o C O V - (r0ugh dimensioned variants Concept orlaYouts) sketches @ - € concePt Solution lay0ut !irns65i6ned @ a 0tprinciples c0mbinati0n Selected - lnprouedla1oul aSSemDlles Selected Assignments are set not only by clients but increasingly, and especiallyin the cuse of original designs, they originate in the special planning dipartments of eompanies.In that case,the designersare bound by the planning ideasof others. lrven then, however, the designer'sspecial skills wiil prove most useful in the ltredium and long-term planning of products. The senior staff of the design dcpartment should therefore maintain close contacts not only with the production department, but also with the product planning department. l'lanning can also be done by outside bodies, for instance by the authorities, bv planning committees etc. llcfore he takes the first step, by proceedingto the clarification of the task in hlrlcl' the designer should familiarise himself *itn tne principles and procedures ol grroductplanning. E E = D ol assemDlles variants Formdesrgn E U A NI assemottes 0ptimum FinallaYout - ol components deslgn Detail rrrm * documents Plgiu6lion lists. instructl0ns) oarts ldrawings. Figure3.4.Flow diagramofthe designprocess'from [3'6] Experiencehasshownthat'byandlarge,theresultsof.thestep-byconventi procedure compare favourably with those involved in the approach. .l Product planning l.l Taskandprocedures e l.rg 11commercial product can be designedthere has tobe aprocluctidea;thar , rnt' lhat promisesto lead to technicallyand economicallyviable applications. z\t ctrrcfing to Brankamp [4.2] and [4.77], product planning is the systematic ttt h lirr. and selectionand development of, promising product ideas.In many rIrrrrics,accordingly, the product planning department is expectedto follow rlt'r'ckrpment of the product idea in the design and manufacturing departtr irncl to watch over its market behaviour. In this book we shall only be rrrr witlr product planning in the narrower sense.while it is often left to the irrg clircctor or other responsible individuals to develop and market the grrotlrrctat the right time, it is now increasinglyacceptedthat innovations tl bc systcmatically planned. A very important aspect of the systematic It is that it provides a better prediction of the timing and costs of a i r t 'l ) r ( ) . l c c t . r l i r t t u l u s l i r r i r p r o c l u c tp l a n c a n c o m e from outside or from within the y , W c i r c c o r c l i n g l cy l i s t i n g u i s hb c t w c c n c x t c r n a l a n d i n t e r n a ls t i m u l i . t t l l n d c c o l l o r n i co b s o l c s c c n c co l ' t h c c ( ) n l P a n y ' sp r o d u c t s ,i d c n t i l ' i c c l by n drop in lurnoveri 46 4 P r o d u c t p l a n n i n g a n d c l a r i f i c a t i o n o f t h e task discovery of new researchdata, proceduresor technologies; market requirements; - economic and political changes; and -technical and economic superiority of competing products' -the Internal stimuli include: capacitY; - drop in profitabilitY; - n"* discoveries by the company research department; and - introduction of new production methods' Therehavebeennu-",ou,proposalsforasystematicandorganisedapproach o involve the following D ,o pioOu", planning [4.4, 4.6, 4.i-, 4.I4,4.15] all of which steps: -situation analysis and definition of company objectives; -discovery of Product ideas; -product selection; and - product definition. (I'2'3)' These steps are compatible with the general systems approach 4.1.2 Situationanalysisand definitionof companyobjectives are am Market analysisand clarificationof companypotentialand.objectives the most important first stepsof successfulproduct planning' ^. Market analysisis first of all appliedto the turnover and profit situation. gre timely detection of profit shortfills and their correction are clearly of aspects: impoitance.Next the analysisis extendedto the following external - socio-political and environmental requirements (including laws regulations); - limits of growth; - overall market develoPments; - economic conditions; and - technological develoPments. The accuracy of the analysis is impeded by: - market fluctuations; - decreasing life cycles of products; and -uncertain forecasts. of An important aspect of product planning over and above the analysis externaliactors isitre anitysis of internal data, representedby the companv potential and its actual situation. The company potential charactertses Kraa tverall capacityof an enterpriseto meet a demand.Kehrmann[4.6] and a compa of areas and types the surveyof [a.9]have-ud" u comp.eh"nsive potential (Figure 4.1) pr Apart from data obtained from outside and inside the company' objectives. company's the of planning also calls for a clear definition objectives may include: - i i g t t m a r k e t g r o w t h a n d a g o o d s h a r eo f t h c m a r k c t : I I Product planning 4j high flexibility in case of market fluctuations; and high rates of profit and good liquidity. Market conditions, company objectives and company potential define the rrrca in which the search for a new product can be usefully pursued. That area is Areaol potentialDevelopmenl lypeot notential Information t acilities l'rtrson nel I lltance Procurement Production Distribution Fvnpripnro Experience Experience Experience - Development - Procedure of Negotiation ol Publicity functions and - Customer delivery terms Preparation service - 0rganisational properties - 0rganisational Materials - Working principles methods Dimensions melhods 0rgan isational Purchasing organiPrecision Sales organisation - 0rganisational Customer meth0ds sation relations Trademark rights withsuppliers methods Contacts Sale negotiators - Patents - Materials, - Final bought- 0rgan isationai buyer - Ltcences outparts Structure etc Resources etc etc etc Fnrrinmpnt Property, Means of developmenl Buildings Branches - Experimental lields Means of transporl Infrastructure Equipment - Test equipment Means ol production Means ol transport etc etc etc etc Research statf Designers Draughtsmen etc Staff - Insidestaff - Outside stalf Professional staff Auxillary stalf etc etc Stafl - Inside staff 0utside staff etc Budget; long-term linance g r r r t ' - 1 .T1y. p e sa n da r e a so f a c o m p a n y 'pso t e n t i a la, f t e r [4.6] vr crrllcda searchfield 14.2,4.6].In the final determination of a searchfield. il lr bc nssssary to take severaladditional factors into account (seeFigure 4.2). Ic lroundariesof a searchfield depend strongly on the planning horizon-that | .n Ihc time scale (short-term or long-term) set for product planning. 'l'lrc s.me approach underlies the procedural plan shown in Figure 4.3. l..l l)iscovery of product ideas searchfor newproductideas. f cru* ol productplanningis the systematic ideasarc cssentially mcthodsfor findingsolutionsas )thodsol' discovcring ('hlptcr 5, anclshoukl be buscdon thc gcncral*-ting method in fsribcd lurrcd in 2,2,1. -r-r-.- , 1 P r o c l u c tp l a n n i n g a n d c l a r i l i c a t i o n o f t h e t a s k 49 Clarification ol' the task Therearecases,however,inwhichtheseproceduresalonedonotleadtothe discoveryofconcreteproducticleas.Insteadtheythrowupinterestingquestions for whicir solution ideas must first be elaborated' - Environmental control regulations Salety regulations Factory - Standards demands Market - Market potential - Market structure - Market share Competition measures Economic Financial Polic\/ policY Fiscal - Customs agreements regulations Export field Search {ields of search Determination Procedure - Methods - Materials , Aimsof enterprise Lilecyclesof Products Costs,Protits Enterprise Potential Limitsof Potentlal expansion situatton andlicence Patent - lnternationalrsatlon - Population growth - Education develoPments Military I r l r r r e4 . 3 . P r o d u c t p l a n n i n g .1.1.5 Product definition 'I Fisure 4.2. External and internal iniormation for the determination of searchfields, atler [4.6] 4.1.4 Product selection The selection of promising product ideas is of great importance becauseof t expenseof subsequentdevelopments.Such selectioncannot, of course,be mo t than rough and ready at this stage, when still relatively little is known about to product theof properties or the idea product ihe of impleme-ntation develooed from it. Here, too, it may be advisableto make a feasibility st before beginning the search for a concrete solution' With a great many product ideas it is advisableto make a selectionfor purpose oT identifying t6or" product ideas that seem to fit in bcst with I'url .o-puny'r or othei generalobjectives. This step should be followcd by a l i s tcd evaluatitn (see 5.8) based on the technicalancl ccononriccritcriit , 4.e] [4.1 lr, last step of product planning, namely 'product definition', involves the rg,,t il'ication of the most important features and requirements of the final frr,rtluct.Such definitions or proposals are usually submitted to the company lrr',1111 $efs1e they are acted upon. They are best presented in the form of a rrrrrlrlified specification or requirements list (see 4.2) which must later be , r r r p l c t e da n d e l a b o r a t e db y t h e d e s i g nd e p a r t m e n t . .2 Clarification of the task l. | 'I'he importance of task clarification t l es i g t t c r ' sw o r k s t i t r l sw i l l t i t p i r r t i c r r l r r r ' p r o b l c[r' n . v. c r vt a s ki n v o l v c sc c r t l r i n t t l t i t t l s t l u t t t t t i r vc l t i r n g cw i l h t i r r r c l r r r l r r r r r s hl c l r r l l y u r r d c r s t r l o ri l t h c I t t t t t t ts r t l t t t i r l tits t o l t c l o t u t t l , l j r o n t l h c v c r y ( ) u l $ c l .t l r c i c t i r r c .t l r e t l r s k r n t r s l r l c l i t t c d i r s l u l l v i u t t l c l c i r r l yi t s p o r u i b l cf f r t h $ t u m p l i l ' i c i r l i o r r r ; r r rcrol r r c c l i o r r s ,l Product planning and clarificationof the task 50 To that during its subsequentelaborationcan be confinedto the most essential' (requirements end, ind also as a basisfor subsequentdecisions,a specification in list) [4.10,4.11]shouldalwaysbe drawnup and consulted.It is indispensable the caseof originaldesigns. -presented to the designor developmentdepartmentin The task is generally one of the following forms: -aSadevelopmentorder(fromoutsideorfromtheproductplanning department); definiteorder; or -asarequestbasedon,forinstance,suggestionsandcriticismbysales' itself. research,test or assembiystaff, or originatingin the designdepartment and one hand the on proposer or Without closecontactbetweenthe client can solution optimum no other, the thosein chargeof the designdepartmenton often department, design the to presented be expected6".u,,*" the pioblem, as further data does not contain all the necessaryinformation. A phase of following the answer must phase This collection must then be initiated. questions: - What is the ProblemreallYabout? - What implicit wishesand expectationsare involved? - Do the specifiedconstraintsactuallyexist?and - What paths are open for development? formulation Fixed iolution ideas or concreteindicationsimplicit in the task function required the Only outcome. final the on often have an adverseeffect should constraints task-specific the and outputs with the appropriateinputs and be must questions following the purpose that For be specifiedrigtrt at the start. asked: -Whatobjectivesistheintendedsolutionexpectedtosatisfy? - What proPertiesmust it have?and - What propertiesmust it not have? departme eny geneial requirementsnot specifiedby the product planning department' T must be assessedin terms of information collected by the design that end, the following factors should be examined: :1.2Clarification of the task 3. - 51 Standards and guidelines International recommendations. National standards. Expert advice. 4. Future developments -Allowing for changesin requirements ancl fashion. - observing new projects so as to determine the trends of technical and economic developments. - Developing ideas that best meet customers'wishes. once all the necessarydata have been collected, it is advisableto combine them into a system basedon the establishedstepsof the designprocess. For that l)urpose a general specification should be drawn up, namely, a more detailed rcquirements list than the one supplied by the customer. 4.2.2 The specification(requirementslist) t Contents Whcn preparing a detailed specification it is essential to state whether the irrrlividualitems are demands or wishes. I)amands are requirementsthat must be met under all circumstances,in other $o1{5, requirements without whose fulfilment the solution is not acceptable(for irrrtrrncesuch qualitative demands as 'suitable for tropical conditions', ,splashprrol'' etc). Minimum demands must be formulated as such (for examole l' ' 20 kW; L < 400 mm). ll'i,rftesare requirements that should be taken into considerationwhenever possible,perhaps with the stipulation that they only warrant limited increasesin r"t It is advisable to classify wishes as being of major, medium or minor l ) ( ) r t a n c [e4 . 1 3 ] . lhc distinction between demands and wishes is also important at the ,rlrrirtionstage, sinceselection(see 5.6) dependson the fulfilment of demands, rlt' L'vuluation (see 5.8) bears on only such variants as already meet the t;rttcls l. Possible com7anY shortcomings -Evaluationofenquiriestothesalesdepartment.Thisevaluationclarifiest customer'srequirements.Itisimportanttoevaluateenquiriesratherth firmorders,whichalreadyrepresenttheselectionofaspecificproduct' - Customers' comPlaints. -Assembly and test reports. 2. - Stateof technologY ComPetitors' Programmes' tcchnical jtltrrnitls ancl ntttlttt Acc<luntstlf sinrilitr stllutions in tcxtbotlks, S t u c l Yt t l P a t c n t s ' l'r'crr bcfore a certain solution is adopted, a list of demandsand wishesshould rlurwr.rup and the quantitativeand qualitativeaspectstabulated.only then I tlrt' resultinginformationbe adequate: ttrty: All data involving numbers and magnitudes,such as number of items rccluired, maximum weight, power output, throughput, volume flow rittc ctc. ity: All data involving permissiblevariationsor specialrequirementssuch ils watcrproof, corrosi<tnproof, shock proof etc. G g u i r c n t c n t ss h o u l c l .i f p o s s i b l c ,b c q u a n t i f i c c la n c l ,i n a n y c a s e ,c l e f i n e di n c l c s r c s t p t l s s i b l ct c r m s . S p c c i a l i n d i c u t i o n so l ' i n r p o r t i u r t i n f l u c n c c s .i n t c n - or procedurcsmuy llso be includcdin thc rpccificltion,which is thus an 52 4 Product planning and clarification of the task internal digest of all the demands and wishes expressed in the language of the various departments involved in the design process. As a result, the specification not only reflects the initial position but, since it is continually reviewed, also serves as an up-to-date working document. In addition it is a record that can, if necessary,be presented to the board and the sales department so that they may make their objections known before the actual work is started. 2 Format 1.2 Clarification ol the task 53 crnphasismust always be entered in the specification,which will then reflect the progressof the project at any one time. Responsibility for this work is vested in the chief designer. The updated specification should be circulated among all departments concerned with the tlevelopment of the product (management, sales,accounts, researchetc). The specificationcan only be changed or extended by decision of those in charge of lhe overall project. 3 Listing the requirements For a recommended layout of a specification,see Figure 4.4. The format of the specification should be agreed with the company's standards office so that it can be used, elaborated and adopted in as many departments as possible. Figure 4.4 is thus no more than a suggestionthat can, of course, be modified at will. veor orooeily Oblecl wilhquanl kl veand quar tal!e dala I necessry spil nloslb{yslems (l!nclonsor assemb les) of baseon checkiht headings Figure4.4. Layout of a specification It may prove useful to draw up the specificationin order of sub-systems (functions or assemblies)where such can be identified, or else by checklist headings(see4.2.2.3).With established solutions,in whichthe assemblies to developed or improved are already determined, the specificationmust arranged in accordance with these-special design groups are usually put i charge of the development of each assembly. With motor cars, for instance, t sp'ecificationcan be subdividedinto engine, transmissionand bodywork velopment. In the case of essentialand also of less obvious reouirements it is extre useful to record the source of specific demands or wishes. It is then possible to back to the proposer and to enquire into his actual motives. This is partic important when the question arises of whether or not the demands can phangedin the light of subsequentdevelopments. Such change.rin, and additions to" the original task as might rcsult from b c t t c r u n d c r s t a n d i n g o f s o l u t i < l n p o s s i b i l i t i c so r f r o m p o s s i b l c c h t n g c s As a rule designers have some difficulty in drawing up their first specification. l:xperience, however, will greatly facilitate the compilation of subsequentones. It is useful to head all specificationswith a description of the overall task and some characteristicdata, for example'Induction motor, rating 63 kW,4-pole'. 'l'his helps to convey some idea of the nature and scope of the problem. Further data are collected with the help of a checklist reflecting the general anclspecific objectives and constraints.By applying this checklist to the task in hrrnd and then asking what questions he needs to have answered,the designer rrrayelicit a most beneficial associationof ideas. Franke [4.3] has drawn up a very detailed checklist,basedon a searchmatrix. ('hccklists and questionnairesare particularly usefulif they cover no more than a lirnited field, if they do not date too quickly, and if they can be taken in at a glirnce. In this book we shall deliberately refrain from presenting detailed tlrrcstionnaires-it is our considered opinion that easily memorised checklists rrith regular headings will help the designerto hit upon the essentialquestions ;rrrlonatically, and without laborious aids. 'I'lrc first step in the clarification of the task is the elucidation of the necessary Irrrtctionsand task-specificconstraints. This is done by reference to the followtrru lrcadings: geometry-kinematics-forces-energy-material-signals. The lorrrbination of the relevant concepts produces a welcome redundancy and Ircrrccan important check that nothing essentialhas been forgotten. I'lrc remaining general or task-specificconstraintscome under the headings frstcrfin 2.1.6 and must be taken into account time and again. ( )ncc the data have been gathered, they must be combined in a sensibleway. 'lo thlrt end, numbering of individual items may prove useful. lrr thc light of the arguments advanced in this chapter, the following general Itcllrocl of compiling specificationscan now be recommended: ('onrpilc the requirements l'ly attcntion to the main headingsof the checklist(Figure 4.5) and determine l l t c c p r a n t i t a t i v ea n d t h e q u a l i t a t i v ed a t a . Ask: Whlt objcctivcsmust thc solutionsatisfy'? Whlt propcrtics must it havc'/ Whutpropertics mustit not havc? furthcr information Compile '*-- 55 .l Product planning and clarificationol the task 54 Mainheadings Examples ;SUEd:NOV VEPAG PACKING CO. Specification Geometry arrangement, number, spacerequirement, length,diameter, Size,height,breadth, extensron. connection, Kinematics Forces velocity, acceleratl0n. Type ofmotion, direction ofmotion, stiff ness, weight, load, detormation, lrequency, magnitude of force, Direction of force, res0nance. inlertia f0rces, elasticity, Energy pressure, heating, temperature, ventilati0n, state, loss,friction, elficiency, 0utput, c0nve'si0n. capacitV. supply. storage. cooling, Material Flowandtransport of materials. propertres auxl arymaterials, Physica andchemical oltheint a andf nalproduct, prescr atl0ns etc). als(toodregu bedmater Probablelolerance: t I mm control equipment. Inputs lorrn, display, andoutputs, (Project y Bll allowl0raul0malic leed'inln duecOurs€ conference Cardb0ard sediOns fedif mafual lgi70) , nu165 Signals Satety protecti0n Operational andenvrronmental salety. Direct systems, Ergonomics height, clearness 0t layout, relationship, typeol operation, operating Man-machine compatibility. lighting, shape sitting comfort, pr0ducti0n prelerred rnethods, possible I mitations, maximum dimensrons, Factory qualtyandtolerances, wastage. achievable means ofproduction, Producti on control 0uality regulations andstandards. application ofspecial Possibilities oftesting andmeasuring, Assembly foundatlons. installation, siting, regulations, Special Transport (height nature andweight), clearance, means 0ftransport dueto liftinggear, LimitatiOns andconditions ol despatch. 0peration (lorexample, sulphurous marketing area, destinati0n wear, special uses, 0uietness, tropical conditi0ns). atmosphere, Maintenance painting, (il any),inspecti0n, cleaning exchange andrepatr, intervals Servicing Costs permissible investment anddeprecrati0n. manufacturing costs, costoftools, lvlaximum les Schedu project planning date. andc0ntro1, delivery Enddateol development, for Sub-task: assemble cartons )hanges D Requirements W Resp. Assemble andgluel5 carlons/mln ai?o ^f hn"^hr ^, I ca^ri^nc Alternatives 500 x 500mm 400 x 400mm 450x 450mm (onlyl0%) if anyol lhree dlrect ons Carl0ns capable 0l benq rem0ved 151121 r970 bar Available airoressure:6 lheassembled cartons COunler required f0rcOuftlng quick y m0veable Machine wllhout lurther adjustmenl ngthe Gluenq 0n leavngthemachine 0l bear th€giuemuslhave setandthecarl0ns mustbecapable lull load. c leedmechan sm in Outpul n withaul0mal Workng prnclple musla lowincrease t0 30 carlors/m min!l€s 20i70) c0r{cfcfcc Maxm!mprodrcl 0nc0slsDl/ I 5000 (Pr0jecl End ol developmenl Schedule: Planfeddeliverydate Once the task has been adequatelyclarified and the relevant departments itre s a t i s f i e ctlh a t t h c l i s t e d r e q u i r e m e n t sa r e t e c h n i c a l l ya n c l c c o n o m i c i t l l yi t t t i t i i r b l c . t h c w i t v i s c l c i r r l i l r t h c c o r t c c p t u u lc l c s i g nl t h i t s c . Smith s gr0up placed 0nlher base Assembled cartons 0n cofveyor belt1yln0 ll00rleve:300mm Heighl 0f convey0r beltab0ve F i g u r e4 . 5 .C h e c k l i sf to r d r a w i n gu p a s p e c i f i c a t i o n - Specify demands and wishes clearly. -If possible, rank wishes as being of major, medium or minor importance. 2. Arrange the requirements in clear order, as follows: -First define the main objective and the main characteristics; -then split into identifiable sub-systems, functions, assemblies etc, or in accordancewith the main headings of the checklist. 3. Enter the specification on standard forms and circulate among interested departments"licensees,directors etc. 4. Examine objections and amendments and, if necessary,incorporate them in the specification. Page:1 311311971 1l7 11911 21,'1i71 m f!les2ill P0nt2) c0nlerence ncw Ihsetln0limcls (Prolecl fl0lmeI gluenqmach lll/ I,'7l putchased cOstD[/ 6000 ftom0!tsideAdditional equlpment Gl!cing I )t'11 (induslria safety) wllh2 handcOftrol can0fly bestafted Opefation 8 )ttl Provide slop emerq€ncy li'ol,nr'ri ,jsr/' r/ I r g t r r e ' l , t t , l ' ; 1 1 l o ts I e c i t i c l t t r o r r l i r r i r ( ' i r t l o n ; r r s e t t t l r l t t t l r c l t i n c ' ( l ) c t t u r t t t l s / ) l l t tvteo t I t ' t ' t r r r r t l i t; r t t . t l) , 56 ' 1 P r o c l u c tp l a n n i n g a n d c l a r i f i c a t i o n o f t h e t a s k 4 Examples our first example concerns the subsidiary task 'assemble cartons' in the new design of a packing machine (Figure 4.6). Quatitative and quantitative considerations were taken into account, for instance on the specification of the cardboard sections.A first change in the specification(15 December 1970)was made following the discovery that, though the nominal rating of the compressed air system was 8 bar, no more than 6 bar could be counted on with certainty. closer study also showed that glueing would prove more costly than was originally anticipated, so that the maximum costswere corrected on 29 January I97I, the extra expenditure having first been approved. rn 5.2, Figure 5.2, and in 5.9, Figures 5.65 and 5.81, complete specifications based on the above recommendationsare provided as further examples. 5 Further applications Even when the design is not original and the solution principle as well as the layout are fixed so that nothing more than adaptations or dimensional changes have to be made in a familiar area, orders should neverthelessbe executed on the basis of specificationswhich can then take the form of printed forms or questionnaires.They should be constructed in such a way that information for electronic data processing and quality control can be read off directly. As a result, specificationsbecome sourcesof information for direct action. Beyond that, specifications once compiled, are an invaluabl e store of information about the required or desired properties of the product, and hence extremely helpful for further developments, negotiationswith suppliersetc. The examination of a specification during project conferences or before assessingvarious designs is an extremely useful procedure. All those involved are put in possessionof all the available information and all salient evaluation criteria are brought home to them. Conceptualdesign Conceptual design is that part of the design process in which, by the identification of the essential problems through abstraction, by the establishment of function structures and by the search for appropriate solution principles and their combination, the basic solution path is laid down through the elaboration of a solution concept. From Figure 3.3 we can see that the conceptual phase is preceded by a decision based on the following considerations: - Has the task been clarified sufficiently to allow development of a solution in the form of a design? - Must further information about the task be acquired? - Is it possible to reach the chosen objective within the given financial restrictions? -Is a conceptual elaboration really needed, or do known solutions permit direct progress to the embodiment and detail design phases? - If the conceptual stage is indispensable, how and to what extent should it be developed on systematiclines? 5.1 Stepsof conceptualdesign According to the procedural plan outlined in 3.2, the conceptual design phase succeedsthe clarification of the task. Figure 5.1 shows the steps involved; they rrre correlated in such a way as to satisfy the general principles set out in 3.1. The reasons for the individual steps have been examined in Chapter 3 and nced not be further discussedhere. It should, however, be mentioned that rcfinements of any one of the steps by reiteration on a higher information level slrould be made whenever necessary.The loops involved have been omitted lrom Figure 5.1 for the sake of greater clarity. 'l'he individual steps and the appropriate working methods will now be c x u m i n e di n d e t a i l . 5.2 Abstracting to identify the essentialproblems 5.2.1Aim of abstraction Solutionprinciplcsor dcsignsbasedon traditionalmethodsare unlikelyto providc optimum ilnswcrswhcn ncw tcchnologics, proccdurcs,matcrials,and 58 5 C o n c e p t u a lc l e s i g n I I I I lnlormation Abstract to identifv the essentlal Droblems I Definitiion II ishfunction Estab structures - sublunctions lunction Overall I Creation princlples for solution Search to fulfilthesubjunctions 6 a principles Combine solution I lunctron to fulfiltheovera 6 O Firmupintoconcept variants variants against Evaluate concept lechnical and economic criteria Check i Decision I Figure5.1. Stepsof conceptual design also new scientific discoveries, possibly in new combinations, hold the key to better solutions. Every industry and every design office is a store of experiences as well as of prejudices and conventions which, coupled to the wish to minimise risks, stand in the way of better and more economic but unconventional solutions. In his search for an optimum solution, the designer, far from allowing himself to be influenced by fixed or conventional ideas, must therefore examine very carefully whether novel and more suitable paths may not be open to him. To that end, he should have recourse to abstraction, which means ignoring what is particular or incidental and emphasising what is general and essential. Such generalisation leads straight to the crux of the task. If it is properly formulated, then the overall function and the essentialconstraints become clear without in any way prejudicing the choice of a particular solution. As an example, consider the improvement of a labyrinth seal in accordance with a specification. In the abstracting approach, the crux of the task would not so much be the design of a labyrinth seal as that of a shaft seal without physical contact, due regard being paid to certain operating and spatiai constraints,and also to cost limits and delivery times. Specifically,the designerwould have ttl ask h i m s e l fw h c t h e r t h e c r u x w a s : 5 . 2 A b s t r a c t i n g t o i d e n t i f y t h e e s s e n t i a lp r o b l e m s 59 -to -to -to -to -to improve the sealing quality or to improve operational safety; reduce weight or to reduce the spacerequirement; lower costs; shorten delivery time; or improve production methods. New developments involving a proven solution principle, coupled to modifications in production methods, are often imposed by the need to lower costs and shorten delivery times. All the requirements listed above might have to be satisfied by the overall solution, but their importance may differ from case to case. Nevertheless,due regard must be paid to each of them, since any one is likely to provide the impetus for the discovery of a new and better solution principle. Once the crux of the task has been clarified to some extent, it becomesmuch easierto formulate the overall task in terms of the essentialsub-problemsas they cmerge. Thus if, in the example we have mentioned, an improvement in the sealing properties were the crucial requirement, new sealing systemswould have to be lirund. This would mean studying the flow of fluids in narrow passagesand, from the knowledge acquired, providing for better sealing properties, while also sirtisfyingthe other sub-problemswe have mentioned. lf, on the other hand, cost reduction were the crucial point, then, after an rrrralysis of the cost structure, one would have to see whether the same physical cllccts could be produced by the use of cheaper materials, by reducing the rrrrmberof components or by using a different manufacturing technique. It is the identification of the crux of the task with the functional connections rrrtdthe task-specificconstraintsthat throws up the essentialproblems for which r o l u t i o n sh a v e t o b e f o u n d 1 5 . 1 6 .5 . 3 5 . 5 . 5 7 1 . .q.2.2 Abstraction and problem formulation I lrc clarrification of the task with the help of a specification will have helped to lot rrs the designer's attention on the problems involved and will have greatly Ilt rcased his particular level of information. Elaborating the specificationmay lhrrs be said to have prepared him for the next step. When he setsout to tackle Irr rrssignment,the designer has no ready solution, or only an inadequate one. )('l)cndingon his knowledge, experienceand familiarity with previous designs, ' lrrsk will be more or less new to him. llis first step is to analyse the specification in respect of the required function csscntial constraints. Roth [5.50] advisesthat the functional relationships ttaincd in thc specificationbe formulated explicitly and arrangedin order of 'l'hlt w i l l r e v e a lt h e s e n e r a l abstraction, analysis,couplcdto a stcp-by-stcp ;ts and cssentialfcaturcsof the task.as follows: l. Eliminatcpcrsunalprcfercncct. - 5 C o n c c P t u a ld c s i g n 5 . 2 A b s t r a c t i n g t o i d e n t i f y t h e e s s e n t i a lp r o b l e r n s Step2. Omit requirements that have no direct bearing on the function and the essentialconstraints. Step3. Transform quantitative into qualitative data and reduce them to essential statements. Step 4. Generalise the results of the previous step. Step 5. Formulate the problem in solution-neutral terms. Depending on either the nature of the task or the size of the specification, or both, certain steps may be omitted. Tabl'e 5.1 illustrates abstraction based on these steps using the specification shown in Figure 5.2.The generalformulation makes it clear that, with respectto the functional relationships, the problem is the measurement of quantities of motor vehiclefuel gaugebasedon specification Table5.1. Procedureduringabstraction: givenin Figure5.2. Specification Fuelgauge D V o l r m e : 2 0l16 0 l Shape llxedor unspec lied(rigid) D l,4atsria : sleelor p astlc Connectlon lo conta ner: Fiange connection Resultof stepsI and 2: - V o l u m e s2: 0 I t o 1 6 0I - Shapeof container:fixed or unspecified(rigid) - Top or side connection - Height of container:150mm to 600 mm - Distancebetweencontainerand indicator:I Om, 3m to 4m - Petroland diesel,temperaturerange:-25"C to 65'C - Output of transmitter: unspecifiedsignal - Externalenergy:(DC at 6V, 12V, 24V, Variation-I5Vo to +257o) - Output signalaccuracyat maximum+3Vc (togetherwith indicatorerror *5".) - Responsesensitivity:\o/oof maximumsignaloutput - Possibilityof signalcalibration - Minimum measurable content:3aloof maximumvalue Result of Step3 - Various volumes - Various containershapes - Various connections - Various contents(liquid levels) - Distancebetweencontainerand indicator:I Om - Quantity of liquid varieswith time - Unspecifiedsignal - (with outsideenergy) D Topconnecliol D Sideconn€ct on H : 1 5 0m m - 6 0 0m m d:gTlmmh=20mm D D slancc lromconla ner10indcalor + 0 m ,3 m - 4 m lm-20m 2 C0nlenls, lemperalure rangemalerial Llquid range operatln0 Sl0raqe envirOnmenl Pelrol 0r diesel -25"C t0 +65'C -40'C l0 +100'C 3 Signal, energy 0ulput0f transmilter: electrlc slgfal(v0ltage change w th quantity chanqe) Availabl€ s0urc€ 0f energy: DCal 6V,12V,24V Vollage varatlon-15% Ia +25'k Oulput signaaccrracy al maxa3% !2"/" (toqelher withndlcalor err0rt5%) undernormal c0ndllions hOrizonlal leve v = cOnsl ablel0 wllhsland shocks 0l n0rmal drivifg Res00nse sensitiv ly: I % 0f maximum oulpul s gnal Result of Step 4 - Various volumes - Various containershapes - Transmissionover variousdistances - Measurecontinuouschangesin quantity of liquid Resultof Step5 (Problem formulation) - Measurecontinuouslychangingquantitiesof liquid in containersof unspccificd from thc conlitittc at variousdistances and shapeand indicatethe measurements 0 5% 01maxmumoulput signal S gfal unallected byanqleol iquidsurtace Possb I ly ol signal calibralion Replaces lirs!issue ol 14i5i19f3 l ' r g u r c . 5 . 2S . p c c i l i c a t i o n r: r r o t o rv c h i c l c f u c l g l u g e 6t 62 5 Conceptualdesign 2rd issue21 6,1973 Specifrcatton Fuegauqe for 63 5 . 2 A b s t r a c t i n g t o i d e n t i f y t h e e s s e n t i a lp r o b l e m s liquid, and that this is subject to the essentialconditions that the quantity of liquid is changing continuously and that the liquid is in containers of unspecified size and shape. This analysis thus leads to a definition of the objective on an abstract plane, without laying down any particular solution. 5.2.3 Systematic broadening of problem formulation Changes D W Bespansible Requirenents Possibil withIu I c0ntaifer ty of siOnacalibralion D Llinimum measurable conlent: 3% 0l maxmumvalue Reserve lankc0nlents byspeclal signa 4 0peralinq condlions D Forward acceleral on 1 l0m/s2 D Sideways accelerat on t10m/s2 0 tlpward acceleral 0n (vibrali0n) up lo 30m/s2 Sh0cks in l0MarddirectiOn wilh0ul damaoe up lo 30m/s2 D FoMard t ll up to 130" D Sideways lill max45' D (ventilated) Tankn0lpressrrised Once the crux of the task has been identified by correct problem formulation, a step-by-stepenquiry must be initiated to discover if an extension of, or even a change in, the original task might lead to promising solutions. An excellent illustration of this procedure has been given by Krick [5.29]. The task he used as an example was an improved method of filling, storing and loading bags of animal feed. An analysisgave the situation shown in Figure 5.3. It would have been a grave mistake to begin immediately by thinking of possible improvementsto the existing situation. By proceedingin this way one is likely to ignore other, more useful and more economic solutions. In principle, the following problem formulations are possible,each representing a higher level of abstraction (broader formulation) than the last: l. Filling, weighing, stitching and stacking bags of feed. .1. Transferrins feed from the mixins bin to stacked bass in the warehouse. 5 Testrequiremsnls D SaltspraytestsfOrinsdeandoulside components accordlng l0 clefl s requremenls D Pressure testfof c0nlaiflef 30kN/m2 6. Lifeexpeclancy dufabll ly 0l conla ner D Lileexpeclancy ln respecl 5 years 0l corrOslOt dueI0 cOflenls andcondensali0n D l\.4ust c0nlorm withheavy vehlcle reg!lalions Stack of empty sacks / Product on -A B0 Key P Prepare L) | ransporl Q Process I Check c0 Wagon \iia\ S Tply To0lied 0 L:ld"lpre'L0.ldrner on maintenance 8 Operat lnstallalion by nonspec a isl 0 #lT:rql lvuslbe repaceab e andmant€nance lree Vstore 9. 0uantily 10,000/day 0l theadiuslable lype5000/day 0l them0stp0puar type 10.Cosls l\lanulactrrinq cOsts< Dl\13 00 each Replaces firslissutol 14i311973 Figure -5.2(continued) D E Stacked sacks await filling Q ManA Iiftsempty sacklromstack andplaces it under spout forfilling. $ ManA f llsthesackbygravrty leed, manually controlling therateof frow. @ ManA hands thebagto manB ManB checks theweghtandadds.orrem0ves r'raleil,i wh(-.n necessary to adjrsl 'heweighl Vv t, lVlan B hands lhcbaqto manC. o o w c a|(isllch0s M;ur(l ftrlrl:; lhclop0l lit0harl Q' tr,tan D takesthebagandloadsit onwagon ioaoeO wagonis pushed to warehouse. by menE andF. aaqsarestacked Baqs awartrng sale. - arestOred Bags areloaded onwaiting lruck, twoo' tnree al a trmebyfanolruck (jclrvrrrrtrr llrorr r0 c0nsumer 'l'he l ' t g u r e5 , . 1 . P l c s c r r ln t c t l t ( x lo l l i l l i r r g ,s l o r i l t g ,n n t l l o n r l i r r gl r i r g ro l l i , c t l .r r l t er l 5 . J t ) l 5 Conceptualclesign 3. 4. 5. 6. 7. 8. Transferringfeed from the mixing bin to bagson the delivery truck. Transferringfeed from the mixing bin to the deliverytruck. Transferringfeed from the mixing bin to a deliverysystem. Transferringfeed from the mixing bin to the consumer'sstoragebins. Transferringfeed from ingredientbins to consumer's storagebins. Transferringfeed ingredientsfrom their sourceto the consumer. Krick hasincorporatedsomeof theseformulationsin a diagram(Figure5.4). StateA = N Feed m i x i nbgi n E StateA EI C '- rj =o E Feed bin mixing L A State = - E Feed bin mixing E A State E o r StateB B c WFeed ingredients at their s0urce Stacked bagsol leedin warehouse B State 'ffi Bagsof leed loaded on delivery truck StateB in Feed c0nsumer's orns storage State B D L,t/ Feed in consumer's bins storage illustrating of the feeddistribution_problem, Figure5.4.Alternativeformulations of a problem,after[5'29] progressively broaderformulations A : initial state;B : final state It is characteristicof this approachthat the problem formulation is made as broad as possiblein successivesteps. In other words, the current or obvious which formulation is not acceptedat face value but broadenedsystematically, Th may conflict with decisionsalreadytaken, but opensup new perspectives. formulation 8 above is the broadest. the most general and the least circu cribed. The crux of the task, in fact, is the transport of the correct quantity and quali of feed from the producer to the consumer and not, for instance, the method of closing or stacking bags, or moving them into the warehouse.With broader formulation, solutions may appear that render the filling of bags a storing them in the warehouse unnecessarv H o w f a r t h i s p r o c c s so f a b s t r a c t i o ni s c < l n t i n u e dd c p c n d so n t h c c o n s t r a i n t s . 5 . 2 A b s t r a c t i n g t o i d e n t i f y t h e e s s e n t i a lp r o b l e m s 65 the case under consideration, Formulation 8 must be rejected on technical, seasonaland meteorological grounds: the consumption of feed is not confined to harvest time; for various reasons the consumer will not want to store feed for a whole year; moreover, he may be reluctant to mix the required ingredients himself. However, the transfer of feed on demand, for instance, with delivery trucks taking it directly from the mixing bin to the consumer's storage bin (Formulation 6), is more economical than intermediate storage in a warehouse and the transport of smaller quantities in bags. In this connection, the reader might recall the development in a different field which culminated in the delivery of ready-mixed concrete direct to the building site in special vehicles. We have tried to show how comprehensive problem formulation on an abstract plane opens the way for better solutions. This approach, furthermore, helps to raise the influence and responsibility of the engineer by giving him an overview of the problem and thus involving him in, for instance,environmental p r o t e c t i o na n d r e c y c l i n g . At this stage, at the very latest, all but genuine constraints ought to have been climinated. Unfortunately, fictitious constraints often continue to impede the designeror proposer, albeit unconsciously.Thus a solution of the feed-transfer problem in accordance with Formulation 6 would have been impossible had the cngineer concerned set himself the fictitious constraint that the transfer might only be made in bags. Another example: technical staff of a company that has been exclusively rnaking or using hydraulic control systemswill all too easily accept the fictitious constraint that, in the future as in the past, all technical control problems must hc solved on the hydraulic principle. This restriction only becomes a genuine eonstraint if it is decided, after due consideration and a deliberate decision, that the hydraulic control system must be retained, for instance in order to increase tlre firm's turnover or in order to simplify storekeeping and maintenance rrroblems. In principle, all paths must be left open until such time as it becomes clear *'hich solution principle is the best. Thus the designer must question all the t onstraints he is given and work out with the client or proposer whether or not they should be retained as genuine restrictions.In addition, the designer must It,rrrnto discard fictitious constraints that he himself has come to accept, and to thrrt end ask critical questionsand test all his presuppositions.Here he may find rt useful to ask the questionswe mentioned in connectionwith the compilation of thc specification, namely: What properties must the solution have? What properties must the solution not have? Abstraction helps to identify fictitious constraints and to eliminate all but licnuincrcstrictions. Wc shall concludc this section with a few useful examplesof abstractionand p r o b l c m l i r r t n u l a t i o n: l)o ntlt clcsigna gilritgc cloor, trut ltxrk lirr lnclrls of sccuringa garagein such a wny that the cur is protcctcd from thievcs und thc wcitthcr, 5 Conceptualdesign not design a keyed shaft, but look for the best way of connecting gear _Do wheel and shaft. Do not design a packing machine, but look for the best way of despatching a product safely or, if the constraints are genuine, of packing a pioduct compactly. Do not design a clamping device, but look ior u -"uns of keeping the workpiece firmly fixed. From the above formulations-and this is very helpful for the next step-the final formulation can be derived in a way that ioes not prejudice the solution and at the same time turns it into a function: 'Seal shaft without contact'-and not 'Design a labyrinth seal,. quantity of fluid continuously'-and noi .Guug" height of liquid -'Measure with a float'. 'Measure out feed'-and not .Weieh feed in sacks'. 5 . 3 E , s t a b l i s h i n gf u n c t i o n s t r u c t u r e s 5.3.2 Breaking down into sub-functions Depending on the complexity of the problem, the resulting overall function will in turn be more or less complex. By complexity we mean the relative lack of transparency of the relationships between inputs and outputs, the relative intricacy of the necessaryphysical processes,and the relatively large number of assembliesand components involved. Just as a technical system can be divided into sub-systems and elements (2.I.1), so a complex or overall function can be broken down into sub-functions of lower complexity. The combination of individual sub-functionsresults in a function structure representing the overall function (see Figure 5.6). Energy Material 5.3 Establishingfunctionstructures 5.3.1 Overall function Accordingto 2'r.3, the requirements determinethe function,that is, the relationshipbetweenthe inputsand outputsof a plant,machineor assembly. In 5.2we explainedthat problemformulaiionobtainedby abstraction doesmuch the same'Hence,oncethe crux of the overallproblemhasbeenformulated, it is possibleto indicate an overall function thai, based on the flow of energy, materialor signalscan,with the useof a block diagram,expressthe relationship betweeninputs and outputs independentlyof thi solution.That relationship must be specifiedas preciselyas possible. In our exampleof a fuel gauge(Figure 5.2), quantitiesof liquid areintroduced . into and removedfrom a container,and the p.bbl.m is to measureand indicate the quantityof liquid found in the containerat any one time. The result, in the liquid system,is a flow of materialwith the funciion .storeliquid' and, in the measuringsystem, a flow of signalswith the function ,measureand indicate quantityof liquid'. The secondis the overallfunctionof the specific taskunder consideration, that is, the development of a fuel gauge(Figur"i.sy. That overall function can be divided into sub-functionsin a iurther stlp. System: Function: t system I Measuring I ----_-- + *LrJl|=. t andrndrcate I I lveasure otliourdI I- Quantity t Figure5.5.overall functionsof the systems involvedin measuring the contentsof a container Figure5.6. Establishing a functionstructureby breakingdownan overallfunction intosub-functions The object of breaking down complex functions is: - the determination of sub-functions facilitating the subsequent search for solutions; and - the combination of these sub-functions into a simple and unambiguous function structure. The optimum method of breaking down an overall function-that is, the optimum number of sub-function levels and also the number of sub-functions pcr level-is determined by the relative novelty of the problem and also by the rncthod used to search for a solution. In the case of original designs, neither the individual sub-functions nor their rclationships are generally known. In that case, the search for, and establishrrrcnt of, an optimum function structure constitute some of the most important stcps of the conceptual design phase. In the case of adaptive designs,on the othcr hand, the general structure with its assembliesand components is much bcttcr known, so that a function structure can be obtained by the analysisof the procluct to bc developed. Depending on the special demands of the specification, that l'unction structure can be modified by the variation, addition or o r n i s s i o n o f i n c l i v i c l u a ls u b - f u n c t i o n st l r b y c h e r n g e si n t h e i r c o m b i n a t i o n . l"unction structurcs arc of grcat importuncc irt thc clcvcl<lpmcntof moclular $ystcms, For ihis typc ol' vuriunt dcslgn, thc physical structurc-that is. thc 6u 5 conccptual design assemblies and individualcomponentsused as buildingblocksand also their relationships-must be reflectedin the function structuie. A further advantageof settingup a function structureis that it allowsa clear defin-itionof existingsub-systems oi of thoseto be newly developed,so that they can be dealt with separately. If existingassemblies can be assigneddirectlyas complexsub-functions, the subdivisionof the function structurecan be discontinuedat a fairly high level of complexity.In the caseof new assembliesor thoserequiring fuithei development,however,the divisioninto sub-functions of decreaiingclmplexitymustbe continueduntil the searchfor a solutionseemspromising.As a iesult, function structuresmay savea greatdeal of time and money. Apart from helpingin the searchfor a solution,functionstructures or their sub-functionscan also b,eusedfor purposesof classification.Exampres are the 'classifying criteria'of classification schemes(see5.4.3)and the subdivisionof designcatalogues. I.t may prove expedientnot only to set up task-specific functions,but alsoto . elaboratethe function structure from geniraily viia sub-functions. The latter recur in. technical systemsand may be helpful in the slarch for a solution inasmuchas they may lead to the discoveryof task-specific sub-functions (see 5.3.3)or as designcatalogues may list sorutions for them.Figure5.7, whichis Characteristic Generally valid functions Symbols Explanations (l)/0utput Input (0) Type Change *z- Magnitude Vary --K*E- Number Connect tE* Place Channel --tr] Placeofll0 Placeoll:0 Time Store *KIF Timeofll0 --<: -=- Typeand outward formol I and0 differ l<0 t>0 Numberofl>0 Numberofl<0 Figure 5.7. Generally valid functionsderived from the characteristicstype, magnitucie, number, place and time in respectof the conversionof energy,material and sigials basedon a suggestion by Krumhauer[5.30],on our own teachingexperience and on a numberof familiarproposals 5.44,5.50], givesa list<ifgenerally valid 15.26, functions. 5 . 3 E s t a b l i s h i n gf u n c t i o n s t r u c t u r e s 69 5.3.3 Logical considerations The logical analysis of functional relationships starts with the search for the cssential ones that must necessarily appear in a system if the overall problem is to be solved. It may equally well be the relationshipsbetween sub-functionsas those between inputs and outputs of particular sub-functions. Let us first of all look at the relationships between sub-functions' As we pointed out in 2.1.3, certain sub-functions must be satisfied before 'if-then' relaanother sub-function can be meaningfully introduced. So-called tionship helps to clarify this point: if sub-functionA is present, then sub-function B can come into effect, and so on. Often several sub-functions must all be satisfied simultaneously before another sub-function can be put into effect. The co-ordination of sub-functions thus determines the structure of the energyl material or signal conversion under consideration.Thus, during a test of tensile strength, the first sub-function-'load specimen'-must be satisfied before the 'measure deformation'-snn be deother sub-functions-'measure force' and ployed. The last two sub-functions,moreover, must be satisfiedsimultaneously. Attention must be paid to consistency and order within the flow under consideration, and this is done by the unambiguous combination of the sub-functions. Logical relationships,moreover, must also be establishedbetween the inputs lundoutputs of a particular sub-function. In most casesthere are several inputs and outputs whose relationshipscan be treated like propositionsin two-valued logic (for example, true/untrue, yes/no,in/out, fulfilled/not fulfilled, present/not present) and expressedby Boolean algebra. We distinguish between AND-functions, OR-functions and NOT-functions, rrnd also between their combination into more complex NOR-functions (OR rvith NoT), NAND-functions (AND with NoT) and storagefunctions with the help of flip-flops [5.10, 5.38, 5.44]. All these are called logical functions. In the caseof AND-functions, all signalson the input side must have the same validity if a valid signal is to appear on the output side. In the caseof OR-functions, one signalonly on the input side must be valid if a valid signal is to appear on the output side. In the case of NOT-functions, the signal on the input side is negated so that thc negated signal appearson the output side. All these logical functions can be expressedby standard symbols.The logical virlidity of any signal can be read off from the truth table shown in Figure 5.8, in which all the inputs are combined systematicallyto yield the relevant outputs. I'lre Boolean equations have been added for the sake of completeness.Using Iogical functions it is possible to construct complex switchings and thus to incrcasc thc safety and reliability of control and communication systems. shows two mechanical clutches with their characteristiclogical Figurc -5.c.) 'l'ltc w r t r k i n g so f t h c c l u t c h o n t h c l c f t c a n b e r e p r e s e n t e db y a s i m p l e luncti<lns. A N I ) - l u n c t i o n ( s i g n a ln r u s tb c s c n t i l n ( l c l u t c l tc n g l g c c lb c f o r c t h c t t t r q u c c a n b e t r l n s r n i t t c d ) . ' l ' h c c t u t c l ro n t l r c r i g h t h a o b c c n s o c o n s t r u c t c dt h i t t , w h c n t h c " pV /-* "'1 I lo .5 Cionceptual design ANDJunction (Conjunction) Designation Symbol 0RJunction (Dislunction) x1 A1 t7 A2 N0TJunction (Negation) / l0 1 Truth table / T-lo Boolean arge0ra (Function) y={2Ax1 y = x 1 ux 7 77 5 . 3 E s t a b l i s h i n gf u n c t i o n s t r u c t u r c s operating signal is given, the clutch is disengagedso that X1 must be negative if the torque is to be transmitted. In other words, only X2 may be present or positive if the desired effect is to be produced' As our next example, we shall consider the catch mechanism of a car door (Figure 5.10). Here, too, the logical relationship is a simple-AND-function, b"iuur. the catch operating lever C can only be activated by the input force lc a c t i n go n t h e l e v e r A i f t h e l o c k i n g l e v e r B i s a t ' 1 ' . S h o u l d t h e s p e c i f i c a t i o n Mechanism v.x LCVETS A,B,C Baseplate 0 spring Spiral E ot leverA surtaces 0 , 0 , c , 0 Working totheSystem F Force applied Figure 5.8. Logical functions X i n d e p e n d e n t s t a t e m e n t( s i g n a l ) ;Y d e p e n d e n ts t a t e m e n t 'l 0 ' , ' 1 ' v a l u e o f s t a t e m e n t e, g 'off', 'on' Kinematic diagram F r 1 . lnputforce x2 t; (First inputvariable) B lever of locking Positron (Second inputvariable) leverC of catchoperating Positron (0utput) of thesystem lunction Logical x1 y=x1A),7 ^? ol1 'v = y' . 'n\ 'y' -/ y . x1 A xt X t - 1 l o rF > 0 x 1 =0 l o rF : 0 rotation X 2 -0 LeverB prevents of leverA ol leverB X ? - 1 Posltion of leverA allowsrotation Y = 0 Doorcatchclosed Y - 1 DoorcatchoPen }/(Torque transmitted) F i g u r c 5 . 9 . l , o g i c a ll i r n c t i o r ro f t w o c l u t c h c s l r i g r r r c5 . l ( ) . ( ' i r l c l rr r r c c h l r r r i sor rl rl r c u r t k x r t i t l l e t 1 5 .l f r l r v i t l rt t t c c h t t t t i s t tkti.l t c t l t l t t i t r l i i r g r l r r irr r t r lh r g i c l r l t t t t c l i r r tttt l t l t c s v s l c l t t 72 5 Conceptualdcsign contain further demands affecting the logical connection, then the function structure will grow correspondinglymore complex [5.16]. Figure 5.11 shows a logical system for monitoring the bearing lubrication system of a multi-bearing machine shaft involving AND- and OR-functions. Every bearing position is monitored for oil pressure and oil flow by a comparison of a specifiedor target value with the actual value. However, only one positive value for each bearing position is needed to allow the system to operate. | _') 5 . 3 E s t a b l i s h i n qf u n c t i o n s t r u c t u r e s 5.3.4 Physicalconsiderations Apart from logical considerations,the demands and wishes of the specification also reflect the physical relationships of the conversion of energy, material and signals. These must be represented by an appropriate function structure (see Figure5.12). ol {low: Types t M Bearing 1 Pt v, Bearing 2 Pt i) Bearlng 3 Pt v3 s Bearing 4 Pt vL anddirection Flowof energy anddirection Flowofmaterial anddirection Flowofsignals System: boundary System Function: Figure5.12.Symbolsfor representing sub-functions in a functionstructure Figure5.11.Logicalfunctionsfor monitoringa bearinglubricationsystem.A positive signalfor everybearing(oil present)sufficesto permit operation.Monitor pressurep; monitor oil flow V - - To decide whether or not a losical function structure mav help in the sea for a solution, it is useful to estaLrlishwhich demands and wishes in t specification have logical contents. It is very important to determine whet these logical contents can be satisfied by logical functions with the AND, OR and NOT relationships alone,or whetherthey mustbe combi into more complicatedsystems.Only in the lastcaseis it usuallynecessary examinethe logical relationshipsfurther. - Logical structuresare evaluatedand optimisedwith thc help of Bot alsebra. Auxiliary function iL i- - - l It is useful to begin by establishingthe main flow of a structure,if clearly present.Once a simple function structurewith its most important relationships hasbeenfound, then it is mucheasier,in a furtherstep,to considerthe auxiliary l'lowswith their correspondingsub-functions,and also to proceedto a further rrnalysis of complexsub-functions. At this point, it often helpsto envisagea lrreliminary solution concept for the simplified function structure without, however,prejudicingthe final solution. Fisures5.13and 5.14showthe functionstructureof a tensiletestinsmachine Our examples make it clear that: Logical relationships are derived directly from the corresponding requi ments of the problem (specification). Such requirements may conce operation,safety,reliability or preventionof faults.They are determined the constraintsof the system. - It is helpful to inquire into the logicalcontentof demandsin the specificati by meansof if-then' propositions. - It is possibleto createthe prerequisitesfor different solutions. - Experiencehas shownthat clarificationof the logicalrelationshipsfacilital the searchfor solutions. Mainlunction Lload E Lde{ormalion Specimen Specimen611o,r16 c_ ul0rce Sdeformation F Lload .t ,l I M e a s u rl e l l'--l I deformation l-l-.rl to, i fdeformation men Speci Specimen6rlo,rr6 D ) v. c r l l l f u n c t i o n( a ) a n ds u b - f u n c t i o (nnsr a i nf u n c t i o n s()b ) o f a t e s t i n g l ; i g u , r c . 5 . 1( 3 mrchinc 14 5 C o n c e p t u a ld e s i g n F.. -auxrl SF tarqet Sarrrsa E Dough Additives -t t"gul----i-er,*rn reotp,', l*r --li ---------Ll[il': - I E LLoad s Specimen l-;',;;1. ;-- !':'lru$1 ir----.i--L-qs--I energy Change intoforceand m0vement 15 5.3 Establishins function structures F, -toss Shaped dough L_ Dispense o L0a0 specimen Fdelormation Waste Separate Feedoul o @ V Waste Figure 5.14. Completed function structure for the overall function set out in Figure 5.13 with a relativelycomplexflow of energy,materialand signals.In this type of overall function, the function structure is built up step by step from subfunctions,attentionat first beingfocussedon essentialmain functions.Thus, on a first functional level, only suchsub-functionsare specifiedas directly satisfy the required overall function. In addition, such complex sub-functionsas 'changeenergyinto force and movement'and 'load specimen'are alsoformulatedat the start, becausethey help to establisha simplefunctionstructure. In the problem under consideration,the energy and signalflows are of roughly equivalent importance in the search for a solution, while the flow of materialthat is the exchange of specimens-is only essential for the holding function which was added to Figure 5.14. Hence it is impossibleto specify just one main flow. In Figure 5.14 an adjusting function for the load magnitudes and, at the output of the system,the energy lost during the energy flow were added because both clearly affect the design. The energy required to deform the specimen is lost. Moreover, the auxiliary functions 'amplify measurements'and 'compare target with actual values' proved indispensable for the adjustment of the energy level. As a further example let us consider the function structure of a dough-shaping machine used, for instance,in the manufacture of biscuits. To satisfy the ove all function of this conversion of material in accordance with Figure 5.15 appropriate or necessarysub-functions of the main flow have to be found The most important can be deduced, often without great difficulty, from t technology or manufacturing techniques of the processinvolved-in this ca the manufacture of confectionery. For the rest, the function structure can completed by asking appropriate questions. Thus a negative answer to question 'May the waste be mixed directly with new dough?' introduces t necessarysub-function 'prepare'. The operation under consideration was fclu to involve seven sub-functionswhich could be combined into a number function-structure variants.Fieure5.15showsfurtherhow severalfunctions be fused together to provide what are often simple and economic solutions. T h e r e a r e , h o w e v e r ,s o m ep r o b l e m si n w h i c h v a r i a t i o no f t h c m a i n f l o w c i r n n o t l c a c lt o i t s o l u t i < t nh c c a u s ca u x i l i a r y f l < l w sh a v c a c r u c i i t l b c i r r i n qo n t lrrgure5.1-5.Overallfunctionandfunctionstructurevariantsof a dough-shaping machine Ior the manufactureof biscuits(in respectof the mainflow only) tlcsigrr. As an cxample. let us consider the function structure of a potato I t l t r v c s t i t t gt r r i t c h i n c .F i g u r c . 5 . 1 6 as h o w s t h e o v e r a l l f u n c t i o n a n d t h e f u n c t i o n t l r u c t u r c b i t s c do n t l t c l ' l o wo l ' n t i r t c r i a l( t h c m a i n f l o w ) a n c lt h e a u x i l i a r yf l o w s o f c l l c r g y a n d s i g n i t l s .I n F i g u r c . 5 , l 6 / ) .b y c o r n p a r i x r t t .l h c l u r r c t i o ns t r u c t r r r ei s 76 5 C o n c e P t u a ld e s i g n Potatoes potatoes Reject Leaves etc. Soil, 71 - 5 . 3E s t a b l i s h i n gf u n c t i o n s t r u c t u r e s System elements Tasks of system elements Accept input torque ano signal Iakeup FandS -L* Turn 1 Spindle ( h\ valid SubJunctions N0. Generally luncti ons Passliquid I t _l Std.S."-t 0usn It II fl )otatoes rn heground Reject J Coarse Leaves Fineparticles Beari ng bush Locate spindle axially and radially bear Ing bush Sealbetween housing andspindle Housi ng n Reject Coarse Leaves Finepartlcles D machine Figure5.16.(a) Functionstructureof a potatoharvesting (b) For comparison: diagramwith generallyvalidfunctionsbasedon Figure5.7 represented by means of generally valid functions, to emphasise the clear interrelationship of the different flows. When generally valid functions are used, the separation into sub-functions is as a rule more pronounced than it is in the case of task-specific sub-functions. Thus, in the present example, the sub' 'connect energ,y function 'separate' is replaced with the generally valid functions 'connect'). with material mix'and'separate material mix' (the reverse of Our next example is meant to illustrate the derivation of function structures the analysisof existing systems.This method is particularly suitable f developments in which at least one solution with the appropriate functit structure is known, and the main problem is the discovery of better solutit Figure 5.17 shows the stepsused in the analysisof a flow control valve (a typi on-off switch). showins the individual tasks of the various elements itn(l I s u b - f u n c t i o n ss u t i s f i c c lb v t h e s v s t e m . F r o m t h e s u b - f u n c t i o n s ,t h c t t t t t c t i s t r u c t u r cc i r n l r c t l c r i v c c l r n c lt h c n v a r i c c lf o r p u r p o s c so l i t t t P r o v i n gt h e p r r l t l lll*,"* lilr-- I I Locate spindle I Threaded axially and + beannq radially I bush Channel l-11J--tiouio Supporl Potatoes liquid tl srrrug,t-*--l liquid seal , - I f* Pass ilquro I Vary llow ptrrrr,rl Provide axial on sealing bush i Channel flowol energy andsignals Connect flowof energy andsignals Connect flowof andmaterial ? signals Varyfiowof material 3 Donotchannel flow ofmaterial(stop) t, Channel flow of material 5 Channel flow ol energy II tl -lrt t , r z # n z IT l i r|l lff1'---!Ei.n | f'ril i"- :--#. ' L--.- -- -t I :igure5.17. Analysisof a flow control valvein respectof its functionstructure The function structure examined in Section5.9 shows clearly that the study of Irrnction structures may prove extremely useful, even after the physical effect lrrs been selected, in determining the behaviour of the system at a very early ',trrgeof its development, and hence in identifying the structurethat best suits the problem under consideration. .5.-1.5Practical usesof function structures \\'hcn cstablishingfunction structures,we must distinguishbetween original and .rtLrptivcclcsigns.In the caseof original designs,the basisof a function structure rs tlrc .r7.rcci/'it'ution and the abstract .formulation of the problem. Among the r l ( ' l l l i t r l ( lisr t t c lw i s h c s ,w c a r e a b l e t o i d e n t i f y f u n c t i o n a lr e l a t i o n s h i p so, r a t l e a s t l l t c s t t b - l t t t t c l i r l t si r l t h c i n p u t sa n c lo u t p r . r t os f a f u n c t i o n s t r u c t u r e .I t i s h e l p f u l t o \ r r l t c ( ) t l ll h c l t r r t c l i o r r r lc l i r l i r l n s h i pi rsr i s i r r lgr o n r t h c s p c c i f i c a t i oinn t h c f o r m o f r t ' t t t c t l c c si t t t t l t o i r r n u l l . l ct l t c s c i n l l t c r l r t l e ro l t l r c i r l r n l i c i l t i r t c ci rl n p < l r t i r n c<cl r i n r r l n c o l l l c r k r g i c i r lo t t l c r ' , 78 5 Conccptual dcsign In the case of adaptive designs, the starting point is the function structure of the existingsolution obtained by the analysis of the elements. It helps to develop variants so as to open the path for other solutions, for subsequentoptimisation and to develop modular products. The identification of functional relationships can be facilitated by asking the right questions. In modular systems, the function structure has a decisive influence on the modules and their arrangement. Here, the function structure and that of the assembly is affected not only by functional considerations, but also, and increasinglyso, by manufacturing needs. Anyone setting up a function structure ought to bear the following points in mind: 1. First derive a rough function structure with a few sub-functions from what functional relationships you can identify in the specification, and then break this rough structure down, step by step, by the resolution of complex sub-functions. This is much simpler than starting out with more complicated structures. In certain circumstances, it may be helpful to substitute a first solution concept fot the rough structure and then, by analysisof that first concept, to derive other important sub-functions. It is also possible to begin with sub-functionswhose inputs and outputs cross the assumedsystem boundary. From these we can then determine the inputs and outputs for the neighbouringfunctions, in other words, work from the system boundary inwards. 2. If no clear relationship between the sub-functions can be identified, the searchfor a first solution principle may, under certain circumstances,be based on the mere enumeration of important sub-functions without logical or physical relationships,but if possible, arranged in order of increasingcomplexity. 3. Logical relationships may lead to function structures through which the logical elementsof various working principles (mechanical,electricaletc) can be anticipated. 4. Function structures are not complete unless the existing or expected flow of energy, material and signalscan be specified. Nevertheless,it is useful to begin by focussingattention on the main flow because,as a rule, it determinesthe designand is more easilyderivedfrom the requirements.The auxiliaryflows then help in the further elaborationof the design, in coping with faults, and in dealing with problems of power transmi sion, control etc. The completefunction structure,comprisingall flows and thei relationships,can be obtained by iteration, that is, by looking first for t structureof the main flow, completing that structureby taking the auxili flows into account, and then establishing the overall structure. 5. In settingup function structuresit is helpful to know that, in the conversion energy,material and signals,severalsub-functionsrecur in most structures should therefore be introduced first. Essentiallv. these are the senerally functions of Figure 5.'7, and they can prove extremely helpful in the search task-specificfunctions. C o n v e r s i o no f e n e r g y : - C h a n g i n g c n c r g y - f o r i n s t a n c e ,e l e c t r i c a li n t o m c c h i t n i c a cl n e r g y . 5 . 3 E s t a b l i s h i n gf u n c t i o n s t r u c t u r e s -Varying energy components-for instance,amplifying torque. - Connecting energy with a signal-for instance,switchingon electricalenergy. - Channelling energy-for instance, transferring power. - Storing energy-for instance, storing kinetic energy. Conversion of material: -Changing matter-for instance, liquefying a gas. -Varying material dimensions-for instance,rolling sheet metal. -Connecting matter with energy-for instance,moving parts. - Connecting matter with signal-for instance,cutting off steam. -connecting materials of different type-for instance, mixing or separating materials. - Channelling material-for instance, mining coal. -Storing material-for instance, keeping grain in a silo. C o n v e r s i o no l s i g n a l s : -Changing signals-for instance, changing a mechanical into an electrical signal, or a continuous into an intermittent signal. -Varying signal magnitudes-for instance,increasinga signal'samplitude. -Connecting signalswith energy-for instance,amplifying measurements. - Connecting signalswith matter-for instance,marking materials. - Connecting signals with signals-for instance, comparing target values with actual values. - Channelling signals-for instance, transferring data. -Storing signals-for instance, in data banks. 6. From a rough structure, or from a function structure obtained by the analysis trf known systems, it is possible to derive further variants and hence to optimise the solution, by: -Breaking down or combining individual sub-functions; - Changing the arrangement of individual sub-functions; - Changing the type of switching used (series switching, parallel switching or bridge switching); and by making Shifts in the system boundary. Because varying the function structure introduces distinct solutions, the sctting up of function structures constitutes a first step in the search for solutions. / Function structures should be kept as simple as possible, so as to lead to sirnple and economical solutions. To this end, it is also advisableto aim at the ( ()rnbination of functions for the purpose of obtaining integrated function ( lrrriers.There are, however, some problems in which discretefunctionsmust be ;rssignedto discrete function carriers, for instance, when the requirements rlcrnandseparation, or when there is a need for extreme loading and quality. In tltis conncction, the reader is referred to our discussionof the division of tasks ( s c c( r . 4 . 2 ) . ti. f n tlrc scitrch lilr it solution, nonc l'tvl ltronti:;ittgJunc'tion structuresshould be t t t l r o t l t r c c r l ,l i r r w h i c h p u r p o s c i t s c l e < ' l i t t tl tt r u t c a d r r r a( s c c . 5 . ( r ) s h < l u l c lb c c t t t p k r y c d .c v c l r a l t h i s c i r r l y s t a g c . u0 5 Conceptualdesigrr 81 5.3 Establishingfunction structurcs + I I t r.l H H H H {fl H H H H E H t . h1 -;ril -11:j I rtl lE= | I l--r !*l i=t*ll1 L--J I I I I * >t? E .9p .9 E E e E E .qo o-E .. s:-I F-.',8 o="=!* E Eo : 93 = e ! 2: 9{> i A '''_ 9o >* q a E E gI I n . = 6 Y.YE= .=;..= = E 6RE ts - 'i 35 3 3.- E Figure 5. 1ll. Dcve krpment of a function structure for a fuel gaugc 3=.=; g-= ts c;:: E: = x - j4 fl =5x5EFS ggE€;:F 6 E Y \ = o . e . = -5 Er-cI?> - . rO _ = -u E= ; c q = 6 * a j^9 5T oI != & >:..o:h E::E -= - 3 H € 5 a R c o 6 _> o -9E=,== = hi 6: =! = = S l c p - b y - s t c pt t c v c l o p r r r c r sr tt i r r t i r r gl l . o r l t h c p r o b l c r nl i r l r r r u l i r t i o n = F " ' - > € a Ee " c q :i = >9:Y @tLd -ll 5 Conceptualdesign 82 9. For the representation of function structures it is best to use the simple and informative symbols shown in Figure 5.12, supplemented with task-specific verbal clarifications. Function structures are intended to facilitate the discovery of solutions: they are not ends in themselves.It dependsvery much on the novelty of the task and the experience of the designer to what degree he will develop them. Moreover, it should be remembered that function structures are seldom completely free of physical or formal presuppositions, which means that the number of possible solutions is inevitably restricted to some extent. Hence it is perfectly legitimate to conceive a preliminary solution and then to develop and complete the function structure and its variants by working through the process as many times as necessary. Let us return to the example of the fuel gauge(5.2.2). Figure 5.18 shows the development and variation of a function structure in accordance with the suggestionspresented in this section. The flow of signals has been treated as the main flow. Associated subfunctions are developed in two steps. Since the specification also provides for measurementsin containers of different sizes,holding varying initial quantities of liquid, an adjustment of the signal to the respectivesize of the container is expedient, and is accordinglyintroduced as an auxiliary function. Measurements in containers of various unspecified shapes will, in certain circumstances, demand the correction of the signalas another auxiliary function. The measuring operation may require a supply of external energy, which must then be introduced as a further flow. Finally, consider the system boundary. If existing indicating instruments are to be used, the device will have to emit an electric output signal. If they are not, then the sub-functions 'channel signal' and 'indicate signal' must be included in the search for the solution. An important sub-function that must be satisfiedfirst, and on the working principle of which the others clearly depend, is 'receive signal'. The solution to this will largely decide to what extent individual sub-functionscan be chansed round or omitted. 5.4 Searching for solution principles to fulfil the sub-functions Solution principles have to be found for the various sub-functionsand t principles must eventually be combined. A solution principle must reflect t physical effect needed for the fulfilment of a given function and also its design features (2.1.5). In many cases,however, it is not necessaryto look special physical effects, the form design being the sole problem. Moreover, the search for a solution it is often difficult to make a clear mental distincti between the physical effect and the form design features. Theoretical i a b o u t t h c n a t u r c a n c lf o r m o f t h e f u n c t i o n c a r r i e r sa r c u s u a l l yp r e s c n t c db y o l d i a g r a n t so r I r c c h i r n d s k c t c h c s . a 5 . 4 S e a r c h i n gf o r s o l u t i o n p r i n c i p l e s It should be emphasisedthat the step we are now discussingis intended to lead to several solution variants (a solution field). A solution field can be constructed by variation of the physical effects and of the form design features. Moreover, to satisfy a particular sub-function, several physical effects may be involved in one o r s e v e r a lf u n c t i o n c a r r i e r s . The following aids and methods are useful in the search for solutions, not only during the conceptual design phase, but also during the embodiment design phase that follows. An attempted solution based on conventional aids may, if necessary,be developed further by methods with an intuitive or a discursive bias. If, in what follows, we distinguish between conventional aids and methods with an intuitive or discursive bias, we do so for purely didactic and systematic reasons.The different approachesdo not exclude and, indeed, often complement, one another. Which precise path should be followed in individual cases depends on the problem, the state of information and the nature of the preliminary work. 5.4.1 Conventionalaids I Literature search I'or the designer, up-to-date technological data provide a wealth of important information. Such data can be found in textbooks and technical journals, in Patent files and in brochures published by competitors. They provide a most rrsefulsurvey of known solution possibilities.Increasingly,this type of information is fed into computer data banks and stored for future use. 2 Analysis of natural systems l lrc study of natural forms, structures,organismsand processescan lead to very rrscful and novel technical solutions. The connections between biology and tt'chnology are investigatedby bionics and biomechanics.Nature can stimulate tlrc creative imagination of the designer in a host of different ways [5.21]. I'cchnical applications of the design principles of natural forms include lr{lrtweight structures employing honeycombs,tubes and rods, and the profiles ,rl ircroplanes, ships etc. Of great importance are lightweight structuresin the Irrr'111 ,r1 thin stems (Figure 5.19). Another technical application is sandwich t'ottstruction.Figure 5.20 shows a few derivations of the natural principles that lrrrvcprovcd useful in the building of aircraft. 'l'ltc ho<lksttf a burr provided the solution incorporated in the Velcro fastener ( l ; i g u r c s. 5 . 2 1a n c l- 5 . 2 2 ) . Anulysisof existingtechnicalsystems ..1 'l'hc anitlysisof cxisting tcchnical systcm$is onc of thc rnost irnportant mcans of ncwor improvcdsolutionvrrlrnb rtcp by titcp. 3cneruting u4 -5 ('onceptual design 5 . J S c r r c h i n gI o r s o l u l i o np r i n c i p l e s 85 ^t\ F i g u r e5 . 1 9 .W a l l o f a w h e a t s t e m [ 5 . 2 1 ] Transverse direction + lil /)Yil ,\,\..\ tt ll ttl \,2\,2\,/ ill .YY) lil \,2\./\,/o, lll o E € )\ \\)) ltl ill lll'il w\ry L A -T-J- o a tlilitq.qJLE} Layeflng Extruded n0neyc0m0 Layeflng F i g u r e5 . 2 1 .H o o k so f a b u r r .a f t e r[ 5 . 2 1 1 Figure -5.22.Velcro fastener,after [5.21] identification of the physical effects involved which, in turn, might have suggestednew solutions principles for corresponding sub-functions.It is also possibleto adopt solution principles discoveredduring the analysis. Existing systemsused for analysismight include: - Products or methods of competing companies. - Older products and methods of one's own company. - Similar products or assembliesin which several sub-functionsor parts of the function structure correspond with those for which a solution is being sought. Becausethe only systemsto be analysedare those having some bearing on the ncw problem as a whole or on parts of it, we may call this method the systematic cxploitation of proven ideas, or of experience.It will prove particularly helpful irr finding a first solution concept as a starting point for further variations. It rrrust,however, be said that this approach carries the danger of causing the tlcsigner to stick with known solutions instead of pursuing new paths. { Analcgies c Figure 5.20. Sandwichconstructionfor lightweight structures[5.22] (a) A few honeycomb structures(b) Completed honeycombstructure (c) Sandwichbox girder This analysis involves the mental or even physical dissection of fin products. It may be considereda form of structure analysis(2.2.I) aimed at t discoveryof related logical, physicaland form designfeatures.Figure -5.17sht a n c x a m p l c o f t h i s t y p e o f a n a l y s i s .H e r e , s u b - f u n c t i o n sw c r c d c r i v c c lf r o m c x i s t i n g c o n l ' i g u r l l i ( ) n , I i r o m t h c r n , f u r t h c r i t n i l l y s i sw o u l d h i t v c l c d t o l n the search for solutions and in the analysis of system properties it is often rrscl'ul to substitute an analogous problem (or system) for the one under considcration, and to treat it as a model. In technical systems,analogiesmay be o l r t r r i n e d ,f o r i n s t a n c e , b y c h a n g i n g t h e t y p e o f e n e r g y u s e d [ 5 . 4 , 5 . 5 2 ] . Artirkrgieschosen from the non-technicalsphere may prove very useful as well ( s c c. 5 . 4 . 1 . 2 ) . llcsiclcshclping in the searchfor a solution, analogiesare also most helpful in tlrc stuclyttl' thc l'rclritviourof a systcm ch.rringan early stage of its development h y l t t c i t n s< t l ' s i t t t u l i t t i oann c lr n o c l c lt c c h n i t l u c s i,r n c li n t h c s u b s c q u e n itd e n t i f i c a l i o n t t l ' c s s c t t t i i t ln c w s u b - s o l u t i o n su n c l / t t rt h c i r r t r t x l u c t i o no l ' c a r l y o p t i m i s a Iton$, fl(r 5 Conceptualclcsign If the model is to be uppii.O to systemsof markedly different dimensionsand conditions, a supportive similarity (dimensional) analysisshould be undertaken ( s e e7 . 1 . 1 ) . 5 Measurements and model tests Measurements on existing systems, model tests supported by similarity analysis and other experimental studies are among the designer'smost important sources o f i n f o r m a t i o n [ 5 . 2 ] . R o d e n a c k e r [ 5 . a 3 ] i n p a r t i c u l a r l a y s g r e a t s t r e s so n t h e importance of experimental studies, arguing that design can be interpreted as the reversal of physical experiment. In the precision engineering and mass production industries, experimental investigationsare an important and establishedmeans of arriving at solutions. This approach has organisational repercussionssince, in the creation of such products, experimental development is often incorporated within the design a c t i v i t y( 1 . 1 . 1 ) . 5 . 4 S e a r c h i n gf o r s o l u t i o n p r i n c i p l e s 87 - Because of inadequate information, new technologies or procedures may fail to reach the designer'sconsciousness. These dangersincreasewith specialisation,the division of tasks and with time pressure. There are several methods of encouraging intuition and opening new paths by the association of ideas. The simplest and most common of these involves critical discussionswith colleagues.Provided that such discussionsare not allowed to stray too far and are based on the general methods of persistent questions, negation, forward steps etc (2.2.1), they can be very effective and helpful. Methods with an intuitive bias such as brainstorming, synectics,the Delphi method, Method 635 and many others involve group-participationtechniquesto generate the widest possible range of ideas. Most of these techniques were originally devised for the solution of nontechnical problems. They are, however, applicable to any field that demands new. unconventional ideas. I Brainstorming 5.4.2 Methodswith an intuitive bias The designer often seeks and discovers the solution for a difficult problem by intuition-that is, the solution comes to him in a flash after a period of search and reflection. As ProfessorJohn Galtung of the International PeaceResearch Institute in Oslo has put it: 'The good idea is not discoveredor undiscovered;it comes,it happens'. It is then developed, modified and amended, until suchtime as it leads to the solution of the problem. Good ideas are always scrutinisedby the subconsciousor preconsciousin the light of expert knowledge, experienceand the task in hand, and often the simple impetus resulting from the association of ideas suffices to force them into consciousness.That impetus can also come from apparently unconnected external eventsor discussions.Frequently, the designer'ssudden idea will hit the bull's eye, so that all he need do is to make changesor adaptations that lead straight to the final solution. If that is, indeed, the caseand a successfulproduct is created,then the designerhas followed an optimum procedure and can rightly be satisfied. Very many good solutions are born in that way and successfully developed. A good design method, far from trying to eliminate this process should rather serve to back it uo. An industrial concern should neverthelessbeware of exclusiverelianceon t intuition of its designers,nor should the designersthemselvesleave everythi to chance or rare inspirations. Purely intuitive methods have the followi disadvantages: -The right idea does not always come at the right time, since it cannot forced. - C u r r c n t c o n v c n t i o n sa n d p e r s o n a lp r e j u d i c e sm a y i n h i b i t o r i g i n a l nlcnts, Brainstorming can be describedas a method of generatinga flood of new ideas. It was originally suggestedby Osborn [5.34] and provides conditions in which a group of open-minded people from as many different spheres of life as possible bring up any thoughts that occur to them and thus trigger off new ideas in the minds of the other participants [5.63]. Brainstorming relies strongly on stimulation of the memory and on the association of ideas that have never been considered in the current context or have never been allowed to reach consciousness. For maximum effect, brainstorming sessionsshould be run on the followins lines: Composition of the group - The group should have a leader and consist of a minimum of five and a maximum of 15 people. Fewer than five constitute too small a spectrum of opinion and experience, and hence produce too few stimuli. With more than 15, close collaboration may decline because of individual passivity and withdrawal. The group must not be confined to experts.It is important that as many fields and activities as possibleare represented,the involvement of laymen adding a rich new dimension. The group should not be hierarchicallystructured but, if possible,made up of cqr.ralsto prevent the censoring of such t'houghts as might give offence to s u p c r i o r so r s u h o r t l i n a t e s . I tudershilt rt.l'thc grotrlt 'l'hc l c a d c r o l ' t h c g r o u l ' r s h o u l d o n l y t t k c t h c i r r i t i i r t i v ci n c l c i r l i n gw i t h o r g u n i s u l i o n u lp r o b l c m s ( i n v i t a t i o n , a o m p o t i t i o n . d u r i t t i o n a n d c v l l u i r t i o n ) . 88 5 (irnceptual design Before the actual brainstorming sessionhe must outline the problem and, during the session, he must see to it that the rules are obsefved and, in particular, that the atmosphereremains free and easy. To that end he might itart the sessionby expressinga few absurd ideas, or mentioning an example from another brainstorming session. He should never lead in the expression of ideas. On the other hand, he can encourage the flow of new ideas whenever the productivity of the group slackens. The group leader must ensurethat no one criticisesthe ideas of other participants.He should appoint one or two participants to take minutes. Procedure -All participants must try to shed their intellectual inhibitions-that is, they should avoid rejecting as absurb, false, embarrassing,stupid or redundant any ideas expressedspontaneouslyby themselvesor by other members of the group. --No participant may criticise ideas that are brought up, and everyone must refrain from using such killer phrasesas'We've heard it all before','It can't 'It will never work' and 'It has nothing to do with the problem'. be done', New ideas are taken up by the other participants, who may change and develop them at will. It is also useful to combine several ideas into new proposals. -All ideas should be written down, sketched out, or spoken into a tape recorder. - All suggestionsshould be concrete enough to allow the emergence of specific solution ideas. -The practicability of the suggestionsshould be ignored at first. -A session should not generally last for more than 30 to 45 minutes. Experience has shown that longer sessionsproduce nothing new and lead to unnecessaryrepetitions. It is better to make a fresh start with new ideas or with other participants later. Evaluation - The results are reviewed by experts and if possible classified, graded in order of feasibility and developed further. - The final result should be reviewed with the entire group to avoid possible misunderstandingsor one-sided interpretations on the part of the experts. New and more advanced ideas may well be expressedor developed during sucha reviewsession. Brainstorming is indicated whenever: -No practical solution principle has been discovered. - The physical process underlying a possible solution has not yel identified. - T h c r e i s a g c n c r a l f c c l i n g t h a t d e a c l l o c kh a s b c c n r c i l c h c ( l . - A r a d i c a ld c p a r t u r c l r o t r r t h c c o n v c n t i o n a la p p r o u c hi s r c q u i r c t l . 5 . 4 S e a r c h i n gl o r s o l u t i o np r i n ei p l e s Brainstorming is useful even in the solution of sub-problemsarisingin known or existing systems. Moreover, it has a beneficial side effect: all the participants are supplied with new data, or at least with fresh ideas on possible procedures, applications,materials, combinations etc, becausethe group representsa broad spectrum of opinion and expertise (for instance, designers, production engineers, salesmen, materials experts and buyers). It is astonishing what a profusion and range of ideas such a group can generate. The designer,for one, will remember the ideas brought up during brainstorming sessionson many future occasions. Brainstorming triggers off new lines of thought, stimulates interest and representsa break in one's normal routine. It should, however, be stressed that no miracles must be expected from brainstorming sessions.Most of the ideas expressedwill not be technically or economically feasible, and those that are will often be familiar to the experts. Brainstorming is meant first of all to trigger off new ideas, but it cannot be expected to produce ready-made solutions becauseproblems are generally too complex and too difficult to be solved by spontaneousideas alone. However, if a sessionshould produce one or two useful new ideas, or even some hints in what direction the solution might be sought, it will have achieved a great deal. An example of a solution obtained by brainstorming will be found in 5.9.2, which also shows how the resulting ideas were evaluated and how classifying criteria for the subsequent search for solutions were derived from them. 2 Method 635 Brainstorming has been developed into Method 635 by Rohrbach [5.45]. After familiarising themselves with the task and after careful analysis, each of six participants is asked to write down three rough solutions in the form of kcywords. After some time, the solutions are handed to the participant's rrcighbour who, after reading the previous suggestion, enters three further solutions or developments. This processis continued until each original set of thrce solutions has been completed or developedthrough associationby the five rrther participants. Hence the name of the method. Method 635 has the following advantagesover brainstorming: A good idea can be developed more systematically. It is possible to follow the development of an idea and to determine more or lcss reliably who originated the successfulsolution principle, which might prove advisable for legal reasons. 'l'lrc problem of group leadership hardly arises. l h c r c l a t i v c d i s a d v a n t a g eo f t h e m e t h o d i s : llccluccclcrcativity by the individual participants owing to isolation and lack ol stinrulirti<ln i n t h c a b s e n c eo f o v e r t g r o u p a c t i v i t y . J l)clphi method I n t h i s m c l h o d . c x p c r t s i n t ; r a r t i c u l a rf i e l d u r c u s k c d k r r w r i t t c n o p i n i o n s 1 5 . 9 1 , 90 5 Conccptual design The requeststake the following form: First round: What starting points for solving the given problem do you suggest?Pleasemake spontaneoussuggestions. round: Second Here is a list of various starting points for solving the given problem. Pleasego through this list and make what further suggestionsoccur to you. Third round: Here is the final evaluationof the first two rounds. Pleasego you consider through the list and write down what suggestions most practicable. This elaborateproceduremust be plannedvery carefullyand is usuallyconfined to generalproblemsbearingon fundamentalquestions or on companypolicy.In the field of engineeringdesign,the Delphi method should be reservedfor fundamentalstudiesof long-termdevelopments. 4 Synectics Synectics is a technicalneologismand refersto the combinationof variousand apparentlyindependentconcepts.Synecticsis comparableto brainstorming, with the differencethat its aim is to trigger off fruitful ideas with the help of analogiesculled from non-technicalor semi-technical fields.The methodwas with first proposedby Gordon [5.18].It is more systematic than brainstorming its arbitraryflow of ideas.For the rest, both methodscall for completefrankness and lack of inhibitionor criticism. A synecticsgroup shouldconsistof no more than sevenmembers,lest the ideas expressedrun away with themselves.The leader of the group has an additionaltask: he must help the group to developthe proposedanalogiesby guiding them through the following steps: - Presentation of the problem. -Familiarisation with the problem(analysis). - Graspof the problem. - Rejection of familiar assumptionswith the help of analogiesdrawn from other spheres. -Analysis of one of the analogies. - Comparisonof the analogywith the existingproblem. -Development of a new idea from that comparison. -Development of a possiblesolution. If the result is unsatisfactory,the processmay have to be repeated with a different analogy. An examplemay help to illustrate this method. In a seminarset up for the purposeof discoveringthe best method of removing urinary calculi from tho human body, severalmechanicaldevicesfor gripping, holding and extracting these stones were mentioned. The device would have to stretch and open inside the urethra. The keywords'stretch' and 'open up' suggestedthe idea of umbrellato one of the participants(Figure5.23). Qirestion:how can the umbrellaprinciple (a) be applied?lly (b) dril 5 . - l S e a r c h i n gl o r s o l u t i o np r i n c i p l c s 91 throughthe stone,pushingthe umbrellathroughthe hole and openingit up? Not very feasible.By (.) pushinga tuLrethroughthe hole andblowingit up (balloon) behindthe stone?Drilling of hole not feasible.BV (d) pushingthe tube pastthe stone?When the tube is withdrawn the resistancemay seriouslvdamagethe 7A*'. / \ \\ ^, til tllil ,{ -'.i ',' tl \J \ -- J- v U2 obc Figure5.23.Step-by-stepdevelopmentof a solutionprinciplefor the removalof urinary calculibasedon an analogy urethra. By (e) adding a second balloon as a guide and by (f) embedding the stone in a gel between the two ballons and then pulling it out? This was found to be the best solution (Figure 5.24). Characteristic of this approach is the unrestricted use of analogieswhich, in the case of technical problems, are selected from the non-technical or semitechnical spheres. Such analogieswill generally suggestthemselvesquite spontaneouslyat the first attempt but, during subsequentdevelopment and analysis, they will generally be derived more systematically. I i g u r e5 . 2 4 . developedin Figure5.23 5 (lombination of methods ,,\ny one of these methods taken by itself may not lead to the required goal. l ' . r p c r i e n c eh a s s h o w n t h a t : 'I'hc group leader of, or another participant in, a brainstorming sessionmay, whcn the flow of ideas dries up, introduce synectic procedures-deriving irnirlogics,systemzrticnegation etc-to releasea new flood of ideas. A ncw idca or an analogy may radically changethe approach and ideas of the 8roup. A surnmary of what has been agreed so far may lead to new ideas. 'destroystone' I n t h c s c m i n a r w c n r e n t i o n e d ,t h e p r c s c n t a t i o no f t h e i d e a producctl a host ttf ncw suggcstions such as tlrilling, smashing, hammering, Ultraronic disintegrution and so on. Whcn ths flow of idcas cvcntually dricd up, - 92 5 Conceptualdcsigrr 93 5 . 4 S c a r c h i n g 1 ' o rs o l u t i o n p r i n c i p l e s 'How doesnature destroy?'which immediately evoked a the group leader asked, number of new suggestionsincluding weathering, heating and cooling, decay, putrefaction, bacterial action, ice expansion and chemical decomposition. A 'clasp stone' and 'destroy stone' provoked the combination of the two principles question, 'What else?'. This produced the answer'contact', which in turn threw up such new ideas as sucking, glueing, and applying various contact forces. The different methods should be combined so as best to meet particular cases. A pragmatic approach ensuresthe best results. 5.4.3 Methods with a discursivebias Methods with a discursive bias provide solutions in a deliberate step-by-step approach. Discursive methods do not exclude intuition, which can make its influence felt during individual stepsand in the solution of individual problems, but not in the direct implementation of the overall task. I Systematicstudy of physical processes If the solution of a problem involves a known physicaleffect representedby an equation, and especially when several physical variables are involved, various solutions can be derived from the analysisof their interrelationships-that is, of the relationship between a dependent and an independent variable, all other quantities being kept constant. Thus, if we have an equation in the form y : / (u,v,w) then, according to this method, we investigatesolution variants for the r e l a t i o n s h i p s . I r: f @ , y , w ) , y 2 : f ( u , v , w ) a n d y 3 : f ( u , v , w ) , t h e u n d e r l i n e d quantities being kept constant. Rodenacker has given several examples of this procedure, one of which concernsthe development of a capillary viscometer[5.42]. Four solution variants can be derived from the well known law of capillary action q * Ap'/l(V't) ( t h e y a r e s h o w n s c h e m a t i c a l l yi n F i g u r e 5 . 2 5 ) . 1. A solution in which the differential pressure/p servesas a measure of thc viscosity: rl 6 lp (V,r and / : const.). 2. A solution based on the diameter of the capillary tube: q 6 / (V,/p and / = c o n s t .) . 3. A solution based on changesin the length of the capillary tube: ry cr lll (/p.V and r : const.). 4. A solution based on changesin the volume flow rate'.n cr IIV (Ap.r and I = c o n s t .) . Another way of obtaining new or improved solutions by the analysis of physicalequationsis the resolution of known physicaleffectsinto their lrrrli vidual components. Rodenacker, in particular, has used this approach in tho design of new devices or the development of new applications for cxisti clcvices. I l y w a y o l c x i t n r p l c ,l c t u s l o o k a t t h e c l c v c l o t . l r n c rot l i r l r i c t i o n i r l t h r c l o c k i r r gt l c v i c c . b l r s c t lt l r r t h c i r r r l r l y s i so l l l t c c t l t t t t l i o t tg r t v c t t t i r t g l l t c l o r I iulrre-5.25.Schematicrepresentationof four viscometers,after [5.421 / ('ontainer; 2 gear pump; ,l variable drive; 4 pressuregauge;5 fixed capillarytube; 6 , rrpillarytube with variable diameter; 7 capillary tube with variable length ,r,'cdcd to release a threaded fastener: r : pl(dt2) tan(Q"- 0) + @t2)pl (1) llrc torque given by equation (1) is made up of the following components; I r i c t i o n a lt o r q u e i n t h e t h r e a d : Tt* P(dl2) ano,: rrlrerc P(dl2')p, (2) tan@,: ptllcos(ol2): p, I r i t ' t i o n i r lt o r q u e o n t h e b o l t h e a d o r n u t f a c e : T,i: P(Dl2) tan@t: P(Dl2) p1 (3) l ( t ' l c l r s ct o r ( l u c o l t h c t h r e a d d u e t o p r e - l o a da n d t h r e a d p i t c h : 1 ' ,ll, l, I'(dl2)rln (-/l) - P . lL (4) t l r r c i r rpl i t c h ./ , . l r c l i xi r n g l c .r / * r n c i u rl h r e i r t l( t ) r l i i r r n c t c r/.' . . - p r c - k r i r r l . r r r el r r l r r e c( | ) t l i r r r t r t ' l c,rr.r , v a [ l u i l (l v ) c o c l l i c i en t o l l t i c t i ( ] ni n t l r c t l r r c l r t l . t)4 -5 Conceptualdesign It : actual coefficient of friction in the thread,lr : coefficient of friction on the head or nut face, a : flank angle, Q : angle of friction). To discover solution principles for the improvement of the locking properties of a threadedfastener, we must analysethe physicalrelationshipsfurther so as to identify the physical effects involved. The individual effects involved in equations (2) and (3) are: - The friction effect (Coulomb friction) 4: - 95 5 . ' l S e a r c h i n gl b r s o l u t i o n p r i n c i p l c s lor criterion Classifying thecolumns labelling tor Classifying criterion labelling therows ! ' P a n df i : P 1 P The lever effect Tt: Ft dl2 and Tt: \ Dl2 - The wedge effect p,, : prlcos (ctl2) The individual effects in equation (4) are: - The wedge effect F, n P tan(-B) - Classifying criterion for labelling therows The lever effect Tr: F, dl2 An examination of the individual physical effects will yield the following solution principlesfor the improvement of the locking properties of the fastener: - Use of the wedge effect to reduce the tendency to loosen by decreasingthe helix angle B. - Use of the lever effect to increasethe frictional moment on the head or nut face by increasing the mean face diameter D. - Use of the friction effect to increase the frictional forces by increasing the coefficient of friction trr. - Use of the wedge effect to increase the frictional force on the face by means of conical surfaces(P.1r1lsin 7 with included angle : 27). This method is used with automobile wheel attachment nuts. - Increase of the flank ansle a to increase the virtual coefficient of friction the thread. 2 Systematicsearch with the help of classification schemes In 2.2.1 we showed that the systematicpresentation of data is helpful in t respects.On the one hand it stimulatesthe searchfor further solutionsin vuri< directions; on the other hand it facilitates the identification and combination essential solution characteristics.Because of these advantases a numbcr classificationschemes have been drawn uo" all with a similar basic strr.rct D r c i b h o l z[ - 5 . 1 1h] a s p u b l i s h e da c o m p r e h e n s i vscu r v e yo f t h e p o s s i b l ci r p p l i c l i < l n so f s u c h c l a s s i f i c a t i o ns c h c m e s I ' h c r r s r r r rt lw o - r l i r r r c r r s i o l sn cl h c r n cc < l n s i s lrst l r o l s i r t t t l c o l t r r t t l tos l 1 n l c t c r sr r s e trl r st ' l r r s s i l v i ncgl i { c r i t r .I j i g r r r c5 , 1 6i l l u r t r n l c rl h c g c r t r ' r ' lsr tl r r r c l t r r c Figure 5.26. General structure of classificationschemes,after [5. I 1l r'lassificationschemes(a) when parameters are provided for both the rows and thc columns and (b) when parameters are provided for the rows only, because thc columns cannot be arranged in any apparent order. If necessary, the , llssifying criteria can be extended by a further breakdown of the parametersor ,lrrrracteristics(Figure 5.27), which process,however, often tends to confusethe r'('ncrAlpicture. By allocating the column parametersto the rows it is possibleto r r;ursformevery classificationschemebasedon row and column into a schemein rrlrich only the row parameters are retained, and the columns are merely r r r u n b e r e d( F i g u r e 5 . 2 8 ) . Srrchclassificationschemeshelp the design processin a great many ways. In l , , r r l i c u l a r ,t h e y c a n s e r v e a s d e s i g n c a t a l o g u e sd u r i n g a l l p h a s e so f t h e s e a r c h j(,r 'r \olut'on, and they can also help in the combination of sub-solutionsinto r , \ t ' r i r l ls o l u t i o n s ( s e e 5 . 5 . 1 ) . Z w i c k y [ 5 . 6 5 ]h a s r e f e r r e d t o t h e m a s ' m o r p h o l o r ' r , ; rrl r u r t r i c e s ' . l lrc clroicc ol classifying criteria or of their parameters is of crucial import. t t t t t ' I l l c s t i r b l i s h i n ga c l i t s s i f i c a t i o ns c h e m e i t i s b e s t t o u s e t h e f o l l o w i n g r t r ' p- l r y - 5 t c 1y.tl r o c c c l u r c : 5 l r ' p l : S o l r r t i o np r o l - r o s a lasr c c n t c r c c li n t h c r o w s i n r a n d o m o r d e r . \ t r ' p . l : l l t r . ' s c p t o p r r s l t l s l r r c l n l r livr sr tcht cl l i g h l o tl h c r n l r i n h c a c l i n g s ( c h a r a c t e r i s t i c s )s t r c l ti r s t v l ) eo l c t t c r g v .t t ' p c o l r r r o t i o nc t c ( s c c F i g u r c s5 . 2 9 u r r c l .s..1()), lltt'p l; 'l l t t ' \ ' i r r c el i r s s i l i tel i r r r r c c o r d u t t gw c a l h l l r er c l r el r r l i r r t s . 96 5 Conccptual clesign 9l 5 . 4 S e a r c h i n gf o r s o l u t i o n p r i n c i p l c s Classilying criteria: physical Types of energy, etfects andoutward appearances Headings. Mechan ical Hydraulic Pneumatic Electrical Examples; intertia, Gravitation, centrilugal torce Hydrostatic, hydrodynamic Aerostatic, aerodynamic Electrostatic, electrodynamlc, inductive, piezo-electric, capacitative, transtormation, rectlfication Maqnetic 0ptical Ferromagnetic, electromagnetic Thermal Expansion, bimetal effect, heatstorage, heat heat heatinsulation transfer, conduction, Chemical reduction, Combustion, oxidation, dissolution, combination, translormation, electrolysis, reaction exothermic andendolhermic Nuclear Biolog ical Radiation, isotopes, source ofenergy Rellection, relraction, interference, diffraction, polarisation, infra-red, visible, ultraviolet putrefaction, Fermentation, decomp0siti0n I'igure5.29. Classifyingcriteriaand headings(characteristics) for variationin the physical searcharea Figure 5.27. Classificationschemewith further subdivisionof parameters.after [5. ] 1l 2 3 4 5 R1 R2 C1 R3 R4 R1 R2 c2 R3 R4 R1 Figure5.28.Modified classification scheme,after [5.11] R2 C3 R3 R4 This procedure not only helps one to identify comDatiblecombinations m o r c i m p o r t a r r ts t i l l , c n c o u r a g e st h e o p e n i n g u p o f t h c w i c l c s tp o s s i b l cs o l u t l'ic Itls. Figure -5.31provides a simple example of a search for a solution to satisfy a sth-function Here the answer was obtained by varying the type of energy rrsainsta number of working principles. Figure 5.32 is an example of variation based on motions. Here, the complex Irrnction'coat carpet with syntheticbacking material'involves the two subI r r n c t i o n s ' m o v ec a r p e t ( s t r i p ) ' a n d ' m o v e a p p l i c a t o r ' f o r w h i c h v a r i o u st y p e s o f rrrotion and their combination serve as classifying criteria. Thanks to its \ ()lnprehensivenature the scheme covers every conceivablepossibility and can ,rlsobe used as a model for similar tasks. As an example of the variation of motions in three dimensions, let us now (()lrsider the function'form support wires for concrete reinforcing rods'. The ''lrrrpeof such support wires is shown in Figure 5.33, and the possiblemotions of t l r e l i r r m i n g t o o l s i n F i g u r e 5 . 3 4 a - c .T h e w i r e s a r e f o r m e d b y t h e c o - o r d i n a t e d rrrovcmentsof two tools (punch and die). By variation of the basic motions rlr.wn in (a) it was possible to correlate 20 punch and die motions (b), and by ,'rrrpkryingthe motions in and round all the co-ordinates to finish up with 239 p r r c l i c i r b l cc o m b i n a t i o n s ,s o m e o f w h i c h a r e s h o w n i n ( c ) . F i g u r e5 . 3 5s h o w st h e r r ' l c c l c r sl o l u l i o np r i n c i p l e s . l i i g t r r c , 5 . 3 ( rs h o w s t h c v a r i a t i o n o f t h c s u r f a c e si n t h e d e s i g n o f s h a f t - h u b t ' o t t t t c c t i t t t l s . ' l ' l t i t t t kl s< l s u c h i l r r i l n g c n r c r ) t st .h c r n u l t i p l i c i t yo f s o l u t i o n s o h l i r i r r c dl.i r r i n s l i r r r c bc y l h c n r c l h o do l ' l o r w l r t l s l c l l s( s c c 2 . 2 . F i g r r r c2 . 7 ) . c i r n I t c l t t t t i l t l o r l r r l c r i u r t l c o r r r p l lec t l . 9t3 - 5 C t r n c c - p t u i rdl c r i g n Classilying criteria Surfaces, motions properties andprincipal material Surtaces Headings Type Shape Exanples. Point, line,surface, body Curve, circle, ellipse, parabola hyperbola, Triangle, pentagon, square, rectangle, hexagon, octagon Cylinder, cone, rhomb, cube, sphere Symmetrical, asymmetrical Position Axial, radial, vertical, horizontal Parallel, sequential Size Small, large, narrow, broad, tall,low Number Undivided, divided Simple, double, multiple Motions Headings Type Nature Exanples Stationary, translati0nal, rotatr0nal Direction Uniform, non-uniform, lating oscil Plane orthree-dimensional In x.y.z direclion ard/oraoouly.y,z axis Magn itude Velocity Number 0ne,several, composite movements Principai properties materiai Examples Headings State gaseous Solid, liquid, Behaviou r plastic, Rigid, elastic, viscous Form grains, powder, Solid body, dust Figure5.30.Classifyingcriteriaand headings(characteristics) for variationin the form designsearcharea If solutions to severql sub-functions are sought, then it is advisable to start by 'l'he uEdLrllB these treating rucss suD-il.lncuons sub-functions as as Ine the classlrylng classifying cnterla criteria lor for the rows. The appropri the rows. uppropri. ate columns are then filled with possible solution principles and their characteris tics, in numericalorder. Figure5.37 showsthe basicstructureof this tvoe classification scheme.In the rows,solutionss1iare assigned to the sub_functir d. Depending on how concrete a level the search for solutions is conduc thesecan be physicaleffects,working or solution principles,function carric selectedcomponents, or merelycharacteristics of individualsolutions.Variati is usuallyfollowed by the combinationof solutionprinciples.combinati methodswill be discussed in sreaterdetail in 5.5. To sum up: the searchfor solutionprincipresfor sub-functionsshoulcl b a s c do n t h e f o l l < l w i nlsi n c s : 5 . 4 S e a r c h i n gf b r s o l u t i o n p r i n c i p l c s Typeof energy ' principle Workinq 1 99 mechanical J h. \ L)) hydraulic reservoir Flywheel , Hydraulic Batlery A o.Bladder | | b Pislon p c.Membrane .+ lV,+l ' (Pressure energy) (rot.) Moving t,T, MASS 7 (transl. ) 3 electrical , Liquid 1 (pot.enerr f' t __,_,t_ Pol energy ln-l -=-l _ t tr+-|fesefvOtf h Flowing liquid ,Fl thermal lVlass ns f-?^.,.1 77777777. vdpdLr tul (electr. field) ,IC Magnet (magn. field) Heated liquid Superheated steam ??r+ Metalspring t, '!:1 F tr 5 6 r0t.+transl.+oot Other springs (compr. against fluid+ gas) l.f€l-ap,nz lrigure5.31.Differentworkingprinciplesto satisfythe function'storeenergy'byvarying tl)etypeof energy Preference should be given to such main sub-functions as determine the principle of the overall solution and for which no solution principle has yet bcen discovered. ('lussifying criteria and associated parameters (characteristics) should be rlcrived from identifiable relationships between the energy, material ancl/or signal flows or from associatedsystems. ll thc physical working principle is unknown, it should be derived from the physical cffccts and, for instance, from the type of energy. If the working principlc has bccn determined. appropriate form design features (surfaces, t t l o t i o t t si t n c l I n i t t c r i a l s )s h o u l d b e c h o s e na n d v a r i e d . C h e c k l i s t ss h o u l d b e t r s c dt o s t i r n u l i r l cn c w i c l c a s( F i g u r c s- 5 . 2 9a n c l- 5 . 3 0 ) . ( ' l i r s s i l i c i t t i t sl tct l t c n t c s h o u l r b l c b u i l t u l ) s t c l )b y s t c p l r n c a l sc < l m p r c h c n s i v c l y i r s p o s s i l r l c .l r r c o r r r l l i r t i b i l i t i cssh o u k l b c t l i s c t r r t l c r li .r n r l < l n l vl h c r n o s l 1 t r < l n r i s i t t g s o l u l i o t l ' p r o p o s i r l sp u r s u c t l . I n s o d o i n g , t l r c t l e s i g r r c rs l r o t r k l t r y t o 100 5 Conceptualdesign 5 . 4 S e a r c h i n gI o r s o l u t i o n p r i n c i p l e s 101 ,.8 .+ o "tJ Figure5.33.Supportwire for concretereinforcingrods[5.23] rz lower rods b upperrod c supportwire F ,.9 ,N .N determine which classifying criteria contribute to the discoveryof a solution, and to examine further variations of these. -The most promising solutions should be chosen by a special selectionproced u r e ( s e e5 . 6 ) . -The same procedure should be followed for other important sub-functions, attention being paid to their incompatibility with previously elaborated sub-functions. -Solution principles should be combined in accordancewith 5.5. If possible, the most comprehensiveclassificationschemesshould be drawn up, that is, schemesfor repeated use, but systemsshould never be built for systematics'sake alone. tl I ffit .l+nrl 'Er o ,o c,) E .l 'l'he use of design catalogues(manuals) c -Y I -E <? 0) tr F-l <.9 Oci lool lool lool Fo 6i - = @.9 c '5 c -! c c io 6= s @+ o;+ o cio r)d O^ 'ti a | )csign catalogues are collections of known and proven solutions to design l,roblems. They contain data of various types and of distinct levels of embodirrrt'nt. Thus they may cover physical effects, solution principles, solution |.nccpts, machine elements, standard parts, materials, bought-out components t lt. In the past, such data were usually found in textbooks and handbooks, ( r)nll)'rnycatalogues, trrochuresand standards.Some of these contained. aoart lr,rrr purely objective data and suggestedsolutions, examples of calculation, r.l1;1i1v11 methods and other design procedures. Design catalogues should lrtrviclc: ' ()trickcr, more problem-orientated accessto the accumulated solutions or tlrrlir. l ' h c n r o s tc o n r p r c h c n s i v er a n g e o f s o l u t i o n sp o s s i b l e ,o r , a t t h e v e r y l e a s t ,t h e t t t o s tc s s c r r t i i rol n c s , t o w h i c h t h c r c s t c a n l a t e r b e a d d e d . 'l'ltc g r c i r t c s lp o s s i b l er i r n g c o f i n t c r d i s c i p l i n a r ya p p l i c a t i o n s . l ) i r l i r l t l r c ( ) t l v c t l t i ( ) r l i rcl l c s i g r tp r o c c d u r c s i t s w c l l i r s l i l r c o r r r p u t c r - a i c l c c l mclltrxls, 102 -5 Conceptualdesign basicmotions Possible Rotation O andTranslation . : diefixed 5 . 4 S e a r c h i n gf o r s o l u t i o n p r i n c i p l e s No. S o l u t i opnr i n c i p l e 03 No principle Solution (a) .-.-------\ Number ol basic motions Variants ,] motion 1 basic Punch Diefixed 2 Punch 2 basic motions Diefixed 3 motion Punch I basic motion Die1 basrc t, Punch 2 basic motions motion Die1 basic 5 11 2 3 5 L l 6 -. I . ._ 't,] 17-2 ._ ^ Punch 2 basicmotions Die2 basicmotions .------ <) 7= a) _ a> (b) No. 1 11-? 11.-1 1 . 1 .3 1.7.-1 ,L,.4, A ,A., .4, ,-K, ,-sQ, ,4, ,A ,ffi, l zz rl' I ,-\, 21 3 31-7 1 rl' --\, 3 . 1 .3- .l J.- I 33 2 | l zz fl ' nl L\L ,-\, 4L 1 1 2.- 3 17.-2 ,rrtt-&t 1 , . 5 ,1 t Y *'gt 4 . 6 . 1- ll n' ^g'lr(, 51,1 5 . 2 .1- 5 . 4 .1- lz 5 lt n' ilt ?lr rt3 I I lt [r h ll+ Lii r-+ 3 3 -2 |-|+ ui t-{ 5 . 71 \ffi ,-r\r, /n\ Figure 5.34. Variation possibilitiesfor motions of tools for forming support wircs for concretereinforcing rods, after [5.23]; ( a ) P o s s i b l eb a s i cm o t i o n s (b) Classificationschcme for possiblcrnotior.rs of purrchanrl rlic ( c ) S o n r co l t l t c p r r r c l i c i t l l l c o r t r b i r l r t i o nos l 1 - r u r r cl nht l c l i cr n o l i o r r s ffi F{*:*}L ffi r c 5 . . 1 5S . c l c c t c t ls o l t t l i o t tp r i n c i p l c sl i t r l i r r r r r i r rsgu l ) l x ) r tw i r c s t i r r c o l c r c t c o t c t l t St r t t l r o t t l l t e l r i r s i so l t o o l r r r o l i o r risr r l C C o r d n t r cwc i t h l r i g u r c5 . . 1 { c l,r t t c r l5,l.j r04 -5 Conceptualdesign Variant Characteri stic 3 7 5 l- 6 @@@@@ &w Position @@@ M Shape wffi Size Number @@& @ Fisure 5.36. Variation of surfacesfor shaft-hub connections,after [5.441 g'q;;trdT m 2 1 F1 Srt Srz Sri Stt 2 F2 Szr szz 52j s;; n F, Sir srz sl sm ,Fn Snt Snz Sni q "nm Figure5.37.Basic structureof a classificationscheme with the sub-functions of an overallfunction and associated solutions The construction of design catalogueshas been studied, above all, by Rot and collaborators [5.46, 5.51]. Roth suggeststhat a design catalogueof the typ shown in Figure 5.38 is most likely to satisfy all the demands listed above' Roth also attaches utmost importance to classifying criteria. They influen the easewith which cataloguescan be handled and reflect the level of complexit of particular solutions, and also their degree of embodiment. In the conceptu 5 . 4 S e a r c h i n gf o r s o l u t i o n p r i n c i p l e s 105 design phase, for instance, it is advisable to select as classifying criteria the functions to be fulfilled by the solutions. This is because the conceptual design is basedon the underlying sub-functions.For classifyingcharacteristicsit is beit to choose generally valid functions (5.3), which help to elicit the most productindependent solutions. Further classifying criteria might include the type and characteristicsof energy (mechanical, electrical, optical etc), of materials or signals, of surfaces, of motions and of physical effects. In the caseof design cataloguesintended for the embodiment design phase, useful classifying criteria include the properties of materials and the characteristics of particular machine elements, such as couplings and clutches.By contrast, characteristicsof solutions, such as dimensions, noise etc, should not be used as classifyingcriteria becausethey are of different importance to different users of the catalogue. In the actual solutions column, it is possible to include physical equations, sketchesof solution principles, layout drawings,names of materials,illustrations etc, depending on the complexity of the catalogue.The type and completeness of the information given once again depends on the intended application. Great importance in the choice of solutions attachesto the columns covering the selection characteristics.Such characteristics may involve a great variety of properties-for instance, typical dimensions, reliability, response, number of clementsetc. They help the designerin the preliminary selectionand evaluation of solutions and, in the caseof computer-basedcatalogues,they can also be used rn the final selection and evaluation. Another important requirement of design cataloguesis that they should have rrniform and clear definitions and svmbols. The more concrete and detailed the stored information, the more direct but ;rlsothe more limited is the application of a catalogue.With increasingdegree of t'nrbodiment, data for a given solution become more comprehensive,but the t lurncesof arriving at a complete solution spectrum decreases.Thus it may be lrossibleto provide a full list of physical effectsfulfilling the function 'channel', lrtrt it would hardly be possibleto list all the potential embodiments of bearings {thanelling a force from a rotating to a stationary system). Ewald [5.12] has I'rrblisheda collection of design cataloguescompiled by himself and by other turrlhors[5.46,5.51]. There are several cataloguescovering generators,transfornr('r.s.gears, bearings, shafts, couplings and springs. Koller has published erlrrlogues of physical effects to fulfil the functions'change type of energy', lrrrrrgc type of signal'and 'vary physicalmagnitudes'15.271. lay0ut drawrngs, physical ons, equati etc sketches Basic Figure-5.38. structureof a clesign l ronr[ 5.'161 catalog,uc, or Assessmenl descnption ol thesoluttons 0r elemenls lrt what follows we include just a few examplesof, or extractsfrom, available signcataktgues. lrigurc -5.39shows a catalogueof physicaleffectsassociatedwith the functions ngc cncrgy'ancl'varycncrgycomponent'. It is basedon Koller 15.27land 'l'he ttiltirucr[.5..1{)1. cirtaloguc mlkcs it possible to derivetheseeffectsfrom c l l s s i l ' y i n gc r i t c r i i r .t h a t i s , ' i n p u t s a n c lo u t p u t s ' .T h e c h a r a c t e r i s t i cosn w h i c h s c l c c l i o n i s b l s c d r n u s t b c c l c r i v c dl ' r o m t h c t c c h n i c a ll i t c r a t u r e . F'igurc.5.4(lshowsa catnkrgucol mcchlnicalsolutionprinciplcsto f'ulfilthc 5 Conceptualdesign 5 . . 1 S e a r c h i n gf o r s o l u t i o n p r i n c i p l e s Physical ellects E^",n y - ,4 t^,,n 1,,' l- Equation for locking direclion I a n ds l i d i ndgi r e c t i o Sn F00ke (Tensioni compTessionl 0en0nq) t= to na . s i n @ .c o s @ -atl,L I 0 nq - l r \ / PI -IRT-"1 I+L I 'o lrt..llt E O E o l-v1-L = \, / 6, -'\ e\ -d \'')f f6) 9 6 -= € '- .E I \l-- + 'd L: FI .= t1tt1 1-plt/ll ; = . = s= l 1 1 / 1 3 ) F .E 6 ttFc E lrt..IA, - P)tona l , !- 1 loia-p1\ - l l ' l o n c z - l r|\ / + c 41 iona.prl r9#3 -v2n*tt -'---tzl . o ',1N, E .9 I = 6 'o 113<.117:117<111 ; p,< Ian a ,l 6 ;> ;l E e E . I lllton d lJ I tan a p.\ - lona /1? t, pr\" .= to^a,ir)', g li L: ', t. u,lt/lt 1 1 ,! , t , 1 p,l,/1,| l, ,, ,), ,,.ffi1,.hn fu,tlr '- t1 I,F1'1' !,' S, t1 1 ut hl 2 L o c k i ncqo r d i t i o n (without friction): s l f& . _ f< J-)sinBf i .a EE - :r/2 wedge.1) J l r ', Fl o n a , l tcn(r(l 4j pt) pr cq + h , >o gE -.= lljr(ll)a-Y I F i g u r e- 5 . 3 9D . e s i g nc : r t a l o g u eo f p h y s i c a cl f f c c t sb a s c t lo n 1 5 . 2 7 5. .j t ) l f i r r t h c g c r r c r ; r l l y a p p l i c a b l et u n c t i o r r s ' c h : r n gccn c r g y ' a n c l ' v a r vc n c r g v c ( ' n l l ) r r n c n t A ' .l s o l r p p l i fkrw ttf sigttlls lnt tt ll = 1t,1t,),p,,1t,' c 5 J ( l l l x t r i r e tl t o t t l 1 1 1 s 1 gt '1r t1t r t l o g t tocl t l t c l r r t t c t r o n ' P r ( ) \ , r (ol rcr c - w l r v axirrl r r r ' , l t o r r1r5 , { h l 5 Conceptualdesign 108 function'provide one-way axial motion'. In this, unlike the previous catalogue, the solutions are concrete enough, thanks to the specification of the form design features, for the embodiment design phase to start with a scale layout drawing' 5.5 Combining solution principles to fulfil the overall function The methodsdescribedso far were primarily intendedto help in the searchfor solution principlesand in the constructionof a solutionfield for sub-functions. To fulfilihe overall function, it is now necessaryto elaborateoverall solutions from the combination of principles (system synthesis).The basis of such combinationsis the establishedfunction structurewhich reflectslogicallyand/or of the sub-functions' possibleor usefulassociations physically The methodswe have beendescribing,and particularlythosewith an intuitive but there bias.are intendedto lead to the discoveryof suitablecombinations, In directly. more are also special methods for arriving at such syntheses the principles with principle, they must permit a clear combinationof solution iretp of the issociated physical quantities and the appropriate form design features. The main problem with such combinations is ensuring the physical and geometricalcompatibility of the solution principlesto be combined,which in the smooth flow of energy, material and/or signals.A further i.r.n "nsures problem is the selectionof technicallyand economicallyfavourablecombina' iions of principlesfrom the large field of theoreticallypossiblecombinations. at greaterlengthin 5.6. This aspectwill be discussed 5.5.1 Systematiccombination For the purpose of systematic combination, the classification scheme to wl 'morphological matrix' (seeFigure 5.37) is particu Zwicky [5.05] reters as the ty useful. Hire, the sub-functionsand the appropriate solutions (solut principles) are entered in the rows of the scheme. If this schemeis to be used for the elaborationof overall solutions'then least one solution principle must be chosenfor every sub-function(that is' every row). To provide the overall solution, these principles must then com6ined systematically into a function structure. If there are rn 1 solul principles for the sub-function Fr, mz for the sub-function F2 and so on, nn thettreti ifter iomplete combination we have ly' tp1 . t z. ft4 possibleoverallsolutionvariants. The mainproblemwith thismethodof combinationis to decidewhichsol p r i q c i p l c s a r c c O m p a t i b l e ,t h a t i s , t o n a r r o w d o w n t h c t l t c o r c t i c a l l y scarch ficld to thc pritctically possiblc search ficld. -rrrft! 5.5 Combining solution principles 109 The identification of compatiblesub-solutionsis facilitatedif: - The sub-functionsare listed in the order in which they occur in the function structure, if necessaryseparatedaccordingto flow of energy,material and signals. - The solution principles are suitably arrangedwith the help of additional column parameters,for examplethe type of energy. - The solution principlesare not merely expressedin words but also in rough sketches. - The most important characteristicsand propertiesof the solutionprinciples are recordedas well. This method of combinationis depictedin Figure 5.41. In Figure5.42 it is appliedto the overallfunctionstructureshownin Figure5.16.The shadedareas Combinations of principles I igure5.41.combiningsolutionprinciplesinto combinations of principles ( o n t b i n a t i o1n: S , , + S 2 2+ . . . + , S n 2 ( o r n b i n a t i o2n: S , , + , S 2 + r . . . + Snr r('l)resent the selected combination of principles, the sub-function ,separate \l()r)cs'involves the application of two solution principles in succession [5.3]. lrurther examples of this method of combination will be found in 5.9 (Figures I { r l ' i -. 5 . 6 9 5 , .90 and 5.91). l'hc verification of compatibilities,too, is facilitated by classificationschemes. ll trv. sub-functions to be combined-for instance, 'change energy, and .vary nrcchirnicalenergy component'-are entered respectivelyin the column and row herrtlingof a matrix with their characteristicsin ihe appiopriate boxes, then the rnPirtibility of the sub-solutionscan be verified more easily than it could be rc such examinations confined to the designer'shead. Figure 5.43 illustrates \ t y p c o f c o m p a t i b i l i t ym a t r i x . ()ttly contbinccompatiblcsub-functions. only pursucsuchsolutionsils meetthe demancls of the specification and look likc fallingwithin rhc prop.scdbu<lgct(sccsclcctionpiocedures in 5.6). Concentrtttctln prontisingcornbinations und cstablishwhy thcscshoulclbe pttfcrred above thc rcst, 5 Conceptual desig n 110 3 ? 1 L iubJunctions\ Lift 7 sitl ---__Change Vary Xetn, nrecnan. energy \ c{)mponent \ Chaindrive ,lmu* ' #;h Separate leaves A Spurgeardrive B grid Sifting 3 5.5 Combiningsolutionprinciples (: wheel Sifting Sb,4t/tt Itese drive C rctionwheel 0 rve Plucker lI1 spiral in 0scillatingBimetal solenoid hotwater Electric motor 7 I il A capable ol rolating slowmolion yes slowrolati0n onlythrough addlt onal (frtrwheelin0 elemenls elc.)dltlicull ro reverse direclion yes lookoutlor shock loads Separate stones 5 Sort p0tat0es 6 Collect by hand byfriction tl checksize (inclined plane) { (holegauge) Conveyor mass cnecK (weighing) Sack{illing devrce Combination ol principles Figure5.42.Combination of principles for designing a potato harvesting machine in with the overall function structure shown in Figure 5.16 accordance In conclusion, it must be emphasised that what we have been discussingI S generally valid method of combining sub-solutions into overall solutions. method can be used for the combination of solution principles during conceptual phase, and of sub-solutions or even of components and assembl during the embodimentphase.Becauseit is essentiallya method of informat processing,it is not confinedto technicalproblemsbut can also be usedin developmentof managementsystemsand in other fields. 5.5.2 Combining with the help of mathematical methods Mathematical methods and computers should only be used for the combina clf solution principles if real advantagescan be expectedfrom them. Thus, at r c l a t i v c l y i r b s t r i t c tc o n c c p t u a lp h a s e ,w h e n t h c n a t u r e t l l ' t h c s o l u t i o n i s n t l t f u l l y u n c l c r s l o o d .l t g u a n t i t a t i v c l a h r l r a t i o n - t h n t i s , u m t t l l r c n r i t t i c acl o m l:J very oifricutt (doml pursue l0 appiy fu(her) f] mnontV beapplied under certain circumslances (deler) witlra rackand swivel, butonly for lowpiston speeds Gearsegments sutlice, depending on angleof rolalon - Lever withsliding blockbulonly tor lowpiston _/ speeds Largetorcesbecause ol torque dlringsiow movemenl imprecise posllioning see82 | additional leverlinkage bulonlyfor ow pislonspeeds yes(when angieof rolalon is small,lever w lh sl d ngblock) \./ yes L yes see82 .,./ t 3 Oscillating piston hydraulic seeD3 \ | | | Irigure 5.43. Compatibility matrix for combination possibilities of the sub-functions 'change 'vary energy' and m e c h a n i c a l e n e r g y c o m p o n e n t ' , f r o m [ 5 . 1 1] tron along with an optimisation-is quite out of place and can be misleading.The ( \ceptions are combinations of known elementsand assemblies.for instance in r:rriant design. In the case of purely logical functions, combinations can be 1 , , ' r . f o r m ewd i t h t h e h e l p o f B o o l e a n a l g e b r a[ 5 . I 5 , 5 . 4 4 ] i n , s a y , t h e l a y o u t o f .',rlctysystemsor the optimisation of electronic or hydraulic circuits. lrr principle, the combination of sub-solutionsinto overall solutions with the lrt'lp of mathematical methods calls for knowledge of the characteristicsor l'rrrllcrtiesof the sub-solutionsthat are expectedto correspondwith the relevant l , r , r p c r t i e so f t h e n e i g h b o u r i n gs u b - s o l u t i o n sT. h e s ep r o p e r t i e sm u s t b e u n a m b i grrousand quantifiable. In the formation of solution concepts, data about the Itlrt'sicalrelationships may be insufficient, since the geometrical relationships Itr.rv lrave a limiting effect and hence may, in certain circumstances,lead to Itrr'orrr;ratibilities. In that case,physical equation and geometricalstructure must ftrsl bc rnatchcd mathematically, and this is not generally possible except for llstcrrrs of klw complexity. For systemsof higher complexity, by contrast, such f t n c l l r l i o n s o l ' t c n l r c c o r n ca m b i g u o u s , s o t h a t t h e d e s i g n e r m u s t o n c e a g a i n f h r x r s c b c t w c c n v i r r i a n t s .w c m a y , a c c o r c l i n g l ys, p c a k o f d i a l o g u es y s t e m si n t l r c l ) r ( ) c c s sr l l c o l n l r i n l r t i o nc t l n s i s t so l r r r i r l h c r r u r t i c lar ln c lc r c i r t i v cs t c p s . fhich J ' l ' h i s n t i t k c s i l c l c i t r t l t i r t t h o u g h , w i t h i n c r c a s i n gp h y s i c i r lr c a l i s a t i o no r p h o d i m c n t o l ' u s o l u l i r t np r i n c i p l c . i t b c c o m c rr i m p l c r t o c s t l b l i s h q u a n t i t a t i v c t12 . 5 C t t n c e p t u u ld e s i g n 5 . 6 S e l e c t i n gs u i t a b l e c o m b i n a t i o n s combination rules, the number of properties increasesand with them the number of constraints and optimisation criteria, so that the mathematicaleffort required becomesvery great. Combining with the help of mathematicalmethods c a l l sf o r t h e u s e o f c o m p u t e r s[ 5 . 6 ,5 . 8 , 5 . 2 6 ,5 . 3 0 ] . 5.6 Selectingsuitablecombinations For the systematicapproach, the solution field should be as wide as possible.By paying regard to all possibleclassifyingcriteria and characteristics,the designer is often led to a larger number of possible solutions. This profusion constitutes the strength and also the weaknessof the systematicapproach. The very great, theoretically admissible but practically unattainable, number of solutions must be reduced at the earliest possible moment. On the other hand, care must be taken not to eliminate valuable solution principles, becauseoften it is only in their combination with others that an advantageousoverall solution will emerge. While there is no absolutely safe procedure, the use of a systematic and verifiable selection procedure greatly facilitates the choice of promising solutions from a wealth of proposals [5.37]. This procedure involves two steps, namely elimination and preference. First, all totally unsuitable proposals are eliminated. If too many possible solutions still remain, those that are patently better than the rest must be given preference. Only these solutions are evaluated at the end of the conceptual design phase. If faced with numerous solution proposals, the designer should compile a selectionchart (Figure 5.44).In principle, after every step-that is, even after establishingfunction structures-only such solution proposalsshould be pursued AS: - are compatible with the overall task and/or with one another (Criterion -fulfil the demands of the specification(Criterion B); realisablein respect of performance, layout etc (Criterion C); and -are expected to be within permissablecosts (Criterion D). Unsuitable solutions are eliminated in accordancewith these four cri applied in the correct sequence. Criteria A and B are suitable for decisionsand their application posesrelatively few problems. Criteria C and often need a more quantitative approach, which should only be used o criteria A and B have been satisfied. Since criteria C and D involve quantitative considerations,they may lead only to the elimination of proposed solutionswith too small an effect or too a cost, but also to'preferencesbased on large effects,small spacerequire and low costs. solut if, arnongthc vcry largcnumhcrol'possiblc is iustificcl A prcfcrcncc thcrc urc $omcthlt: -rjtr.- r c 5 , 4 4 .S y s t c n r l t i cs c l c c t i o nc h a r t : a l , b l . c t c a r c s o l u t i o nv a r i a n t so f t h e p r o p o s a l s c i n ' l ' a h l c . 5 . 1 , ' l ' h cc o l u n t r r c s c r v c r l{ i r r r c n l r r k s i l s l sl c i r s o n sf o r l u c k o f i o no r c l i n r i n i r t i o n 114 5 Conceptualdesign 5 . 6 S e l c c t i n gs u i t a b l e c o m b i n a t i o n s 115 - incorporate direct safety measures or introduce favourable ergonomic conditions (Criterion E); or - are preferred by the designer's company, that is, can be readily developed with the usual know-how, materials, proceduresand under favourable patent conditions (Criterion F). It must be stressed that selection based on preferential criteria is only advisablewhen there are so many variants that a full evaluation would involve too much time and effort. In the caseof function structuresand solution principles,criteria A and B will usually suffice. Only after combination of the solution principles need criteria C and D be applied, followed, if necessary,by the application of criteria E and F. If, in the suggested sequence, one criterion leads to the elimination of a proposal, then the other criteria need not be applied to it there and then. At first, only such solution variants should be pursued as satisfy all the criteria. Sometimes, however, it is impossible to settle the issue because of lack of information. In the case of promising variants that satisfy criteria A and B, the gap will have to be filled (see 5.7) by a re-evaluationof the proposal, which will ensure that no good solutions are passedover. The criteria are listed in the order shown above as a labour-savingdevice, and not in order of importance. The selectionprocedure has been systematisedfor easierimplementation and verification (Figure 5.44). Here, the criteria are applied in sequeneeand the reasonsfor eliminating any solution proposal recorded. Experience has shown that the selectionprocedure we have describedcan be very quickly applied, that" it gives a good picture of the reasonsfor selection,and that it provides suitable documentation in the form of the selection chart. If the number of solution proposalsis small, elimination may be bapedon the same criteria. but less formally recorded. The example we have chosen concerns solution proposals for a fuel gauge in accordance with the specification in Figure 5.2. An extract from the list of proposalsis given in Table 5.2. A further example is shown in Figure 5.45. The specification of a test r for gear couplings demanded an axial displacementin the test coupling so t the forces which then appear could be measured. The possible position of t Figure 5.45. Sketch showing the principle of a test rig for gear couplings. 1 drive; 2 gearbox;3 high-speed shaft; 4 test gear coupling; 5 adjustable bearing block for s et t i n g t h c a l i g n n t c n t ;6 d c v i c c f o f lpplying l()rquc Solution principle a1 a2 a3 l. Mechanical, static I.I. Liquid Weight of liquid Mass attraction Use of solvent in the liquid a4 Matter suspendedin the liquid b1 1.2. Gas Gas-filled bladder on top of the liquid DZ Volume of gas in sealed container cl c') c3 1.3. Reproduction (Analogue) Similar quantities of liquid in a container with similar cross-sections Similar volumes of gas (similar cross-section) Container-shapedbody with p j. p1iuu1,1 Indication by: Suspend by spring + displacement Attach source of light to floating body displacement of light beam Signal Force Force Concentration Residue (of dissolved substance) Concentration Displacement Pressure (of gas) Pressure (of gas) Force Displacement Pressure (of gas) Force 2. Mechanical, dynamic 2.1. Liquid Exploit the inertia of liquid (acceleration) Oscillation of mass of liquid Force (moment) Frequency, time interval Pump liquid from one container into another Time Irleasure quantity pumped across Enefgy (electrical) Measure differences in the inflow and the outflow Quantityof in the container rnatter(+) Tap the container and measure response Frequency A t t a c h o s c i l l a t o rt o c o n t a i n e rTime (oscillation) response dependent on mass of liquid 2.2. Cas Pump gas into a sealed container or pump out gas until a fixed pressure or a fixed quantitv of gas is reached Pressure(of gas) -1. F,lcctricul ll : as ohmic rcsistance (volumedependent) Ohmic resistance l-it1u1.! Liquiclas diclcctric(volumedependent) Capacitv r r l t l c5 , 1 . l l x t r i r c t I r o r r rt h c l i s l o l s o l u t i o r rp r o p o s i r l sf i r r t l c s i g r r i r rugl i r c l g u t r g c ll6 5 Conceptualdesign displacement (classifying criterion of the rows) and the axial force input (classifying criterion of the columns) were combined into the classification scheme shown in Figure 5.46. The various combinations were checked and unsuitable variants were eliminated for a number of not immediatelv obvious reasons. Axialforce input \ 3 v --lli:- ---T-D<XI hydrodyn. Position \ of disolacemenlaxialbearing t1 1 Right-handI prnr0n -* Inerm. expansion -Fo T$ry , !E* l]-___*Jrn-$ z ,,tv. ., 2 Right-hand 1 2 sleeve 1d*___llhilfrs. rolling bearing +tlll.' --tt l L -w- 3 Adlustable l3 shaft tHHFili$ D 4 Lett-hand sleeve :-#H_l-ht_l-$ 5 lntermediateI5 shaft t1fi ,/,(t IL o. ,.,W _FF _l _T t-- ,8.-.. \ +- --l1.. r rf llt' 25 F,H ffi Figure 5.46. Systematiccombination and elimination of variantsthat are unsuitablein principle Combination 21: Fo too great (life of rolling bearingstoo short) Conrbination23: 2 . Fo, hence life of rolling bearingstoo short Combination 22,24: Peripheral speedtoo great (life of rolling bearingstoo short) Combination 31-34: Thermal lencth too small 5.7 Firming up into conceptvariants The principles elaborated in 5.4 and 5.5 are usually not concrete enough to le to the adoption of a definite concept variant. This is because,as the searchtirr solution is based on the function structure, it is aimed, first and foremost, at f u l f i l m e n t o f a t c c h n i c a l f u n c t i o n . A c o n c e p t m u s t , h < l w c v c r ,a l s o s a t i s f y c < l n c l i t i o t tl si t i d d o w r t i n 2 . 1 . 6 , a t l c a s t i n c s s c n c c .l i r r o n l y t h c r r i s i t l l o s s i h l e c v i t l u a t c i t . B c l i r r c c ( ) l l c c p tv i r r i a n t sc l n l ) c c v i t l u u t e d .t h e y n r u s t l l c l i r r n e d 5.7 Firming up into conceptvariants ITl which, as experience has shown, almost invariably involves a considerable effort. The selection process may already have revealed gaps in information about very important properties, sometimes to such an extent that not even a rough and ready decision is possible, let alone a reliable evaluation. The most important properties of the proposed combination of principles must first be given a much more concrete qualitative, and often also a rough quantitative, definition. Important aspectsof the working principle (such as performance and susceptibility to faults) and also of the embodiment (such as spacerequirements,weight and service life) and finally of important task-specific constraints must be known, at least approximately. More detailed information need only be gathered for promising combinations. If necessary,a second or third selection must follow the collection of further information. The necessary data are essentially obtained with the help of such proven methods as: -rough calculationsbased on simplified assumptions; - rough sketches or rough scale-drawings of possible layouts, forms, space requirements, compatibility etc; -preliminary experimentsor model tests to determine the main properties, or approximate quantitative statements about the performance and scope for optimisation; -construction of models to aid analysis and visualisation (for example, kinematicmodels): - analoguemodelling and systemssimulation, often with the help of computers; - further searchesof patents and/or the literature with narrower objectives; and - market researchof proposed technologies,materials, bought-out parts etc. With these fresh data it is possible to firm up the most promising combinations of principles to the point at which they can be evaluated (see 5.8). The properties of the concept variants must reveal technical as well as economic lcatures so as to permit the most accurate evaluation possible. It is therefore rrclvisablewhen firming up into concept variants to bear possible later evaluation criteria (see 5.8.3) in mind, the better to elaborate the information in a purposeful way. An example will show how it is possible to firm up solution proposals into ( ()ncept variants. To that end, we return once more to our fuel gauge. Figure 5.47 shows a possible solution principle. Estimates of the weights and rrrertia forces form the basis of the firming up procedure. 'lirtal force of 20 to 160 litres of the liquid (static): F , , ,:,6 . r . g . V : 0 . 7 - 5 x 1 0 x ( 2 0 . . . 1 6 0 ) : ( 1 5 0 . . . 1 2 0 0 )N ( f u e l ) i\rlclitional lirrccs cluc to acceleration + 30m/s2 (only the liquid is taken into c r l n s i d c r a t i o)n: F r r k=r n t ' o = ( 1 5 . . . 1 2 0 )x 3 ( X t ) = t ( 4 5 ( ) .. . 3 6 ( X ) ) N 118 5 Conceptual design If the force is convertedinto movementit can be detected,for instancewith the help of a potentiometer. (The suppressionof movements resulting from accelerationforces calls for considerabledamping.) It is possibleto obtain the total force, and hence the quantity of liquid, statically, either directly by measuringthree bearing forces or indirectly by measuringjust one bearingforce (seeFigure5.47). Result:developsolutionfurther, provide damping,seekappropriatesolutions and firm them up by means of rough scale drawings.Figure 5.48 showsthe result. Once the necessaryparts and their arrangementare drawn, the proposal can be evaluated. 4gl Figure5.47.Solutionprinciplea1 (Table5.2)measureweightof liquid (Signal= force) Jt t \ l | F-tr-E 3 force-measuring devrces {'-_l -t: I A-----E 1 force-measuring devrce Venlilalion screw soldered 0 i lf i l l i n g fordamping Flowrestrlctor Gaslilling (Pressure balancing duringtemperature changes) Figure5.48.Embodimentof the solutionprincipleshownin Figure5.47 5.8 Evaluating concept variants against technical and economiccriteria I n t h c n c x t s t c p . t h c s o l u t i o n p r o p o s a l s ,n o w f i r n r c r l u p i r r t o c o n c c l . l vt i r r i i r t n u s t b c c v a l u a l c d s o i t s t o p r o v i c l ca n o b j c c t i v c b u s i sf o r d c c i s i o n s . ' l ' l r c r c 5.8 Evaluating concept variants 119 special evaluation procedures to fill this need, all of them so constructedas to lend themselves not only to the evaluation of concept variants, but quite generally of solution variants in every phase of the design process. 5.8.1 Basicprinciples An evaluatior is meant to determine the 'value', 'usefulness' or 'strength' of a solution with respect to a given objective. An objective is indispensablesince the value of a solution is not absolute, but must be gauged in terms of certain requirements. An evaluation involves a comparison of concept variants or, in the caseof a comparison with an imaginary ideal solution, a 'rating' or degree of approximation to that ideal. An important element of design practice is cost analysis. h involves value analysis (VA) [5.19, 5.58, 5.61, 5.62], that is, the determination of 'function costs', by the assignmentof function carriers to the various sub-functions,and the determination of their manufacturing costs. The main problem here is to disentangle functions from components since a single component may carry several sub-functions or a single function may be fulfilled by several components, which leads to an ambiguous distribution of costs. Moreover, costing presupposes the availability of considerable design documentation. Finally, if the evaluation and choice of solutions is basedpurely on production costs,there is the danger that essential technical criteria and other economic considerations-for instance the market reaction to the product-which often cannot be t;uantified in absolute amounts of money, will be ignored. Hence there is a need for methods that allow a more comprehensive evaluation, or in other words cover a broad spectrum of objectives(task-specific rccluirementsand general constraints).These methods are intended to elaborate rrot only the quantitative, but also the qualitative, properties of the variants, thus rrrrrkingit possible to apply them during the conceptualphase, with its low level ,rl cmbodiment and correspondingly low state of information. The result must be r t'liuble, cost-effective,easily understood and reproducible. The most important rrrethods to date are use-valueanalysis* (UVA) based on the systemsapproach l:.(r4l and the combined technical and economic evaluation technique specified rrr (iuideline VDI2225 [5.60], which essentiallygoes back to Kesselring 15.24]. lrr what follows, we shall outline a basic evaluation procedure incorporating f lre conccpts of use-valueanalysisand of Guideline VDI 2225. I ldentifying evaluation criteria 'l lrc lirst stcp in any evaluation is the drawing up of a set of objectives from wlrich cvalr.rationcriteria can be derived. In the technical field, such objectives orc rnirinly l'rascclon thc rcquirements of the specification and on the general C o n s l r i l i n l(s2 . I . 6 ) . t ' l ' h cl c l t t ' u s c - v l l t r ci r n i r l v s i s ' ius d i r c c tt r l t t s l i l l i ( t lt l o r r tt h c o r i g i n a(l i c r n r a nt c x t . A | | m i l n r ,l r u l n r o r cg e n c r ; r lt,c r r nl i r r t h i s t y l t co f t c c h n i q u ci s ' c o s t - b c r t c li r tr t l l v s i s ' . t20 5 Conceptualdesign A set of objectives usually comprises several elements that not only introduce a variety of technical, economic and safety factors, but that also differ greatly in importance. A range of objectives should satisfy the following conditions: -The objectives must cover the decision-relevant requirements and general constraints as completely as possible, so that no essentialcriteria are ignored. - The individual objectives on which the evaluation must be based should be as independent of one another as possible-that is, provisions to increase the value of one variant with respect to one objective must not influence its values with respect to the other objectives. - The properties of the systemto be evaluatedmust, if possible,be expressedin concrete quantitative or at least qualitative (verbal) terms. The tabulation of such objectives depends very much on the purpose of the particular evaluation-that is, on the design phase and the relative novelty of the product. Evaluation criteria can be derived directly from the objectives. Because of the subsequent assignment of values, all criteria must first be given a positive formulation, for example: 'low noise' not 'loudnesslevel' 'high not 'magnitudeof losses' efficiency' 'low not'maintenancerequirements' maintenance Use-value analysissystematisesthis step by meansof an objectivbstree, in which the individual objectives are arranged in hierarchic order. The sub-objectives are arranged vertically into levels of decreasing complexity, and horizontally. into objective areas-for instance,technical, economic-or even into major and minor objectives (Figure 5.49). Because of their required independence,subobjectives of a higher level may only be connected with an objective of the next lower level. This hierarchical order helps the designer to dete\mine whether or not all decision-relevant sub-objectives have been covered. Moreover, it simplifies the assessment of the relative importance of the sub-obiectives. The Areas e.g.economrc etficiency of engine I low runn ing costs _? t' I I I ' .,--1 lOw ,---\.' ( 012)repair ( 02,) f ..f'' 3 costs Y 6@G6 ,,^.. consumption consumption I"igurc5.49.Slructurcof an objcctivcstrcc -*.,- @ 5 . 8 E v a l u a t i n gc o n c e p tv a r i a n t s t21 evaluation criteria, (called objective criteria in use-valueanalysis),can then be derived from the sub-objectives of the stage with the lowest complexity. Figure 5.76 gives a concrete example of such an objective system. Guideline VD|2225, on the contrary, introducesno hierarchicalorder for the evaluation criteria, but derives a list of them from minimum demands and wishes and also from general technical properties. 2 Weighting evaluation criteria To establish evaluation criteria, we must first assesstheir relative contribution (weighting) to the overall value of the solution, so that relatively unimportant criteria can be eliminated before the evaluation proper begins. The evaluation criteria retained are given 'weighting factors' which must be taken into consideration during the subsequent evaluation step. A weighting factor is a real, positive number. It indicates the relative importance of a particular evaluation criterion (objective). It has been suggestedthat such weightings should also be assignedto the wishes recorded in the specification [5.49], but that is only possible if such wishes can be ranked in order of importance when the specification is first drawn up. That, however, rarely happens at this early stage-experience has shown that a whole series of evaluation criteria emerges during the development of the solution, and that their relative importance changes. It is nevertheless most helpful to include rough estimates of the importance of wishes when drawing up the specification, because, as a rule, all the persons concerned are available at that time (4.2.2). In use-value analysis. weightings are based on factors ranging from 0 to 1 (or lrom 0 to 100). The sum of the factors of all evaluation criteria (sub-objectives at the lowest stage) must be equal to 1 (or 100) so that a percentageweighting can be attached to all the sub-objectives. The drawing up of an objectives tree Llreatlyfacilitates this process. Figure 5.50 illustrates the procedure. Here the objectives have been set out on four levels of decreasing complexity and provided with weighting factors. The evaluation proceeds step by step from a lcvel of higher complexity to the next lower level. Thus the three sub-objectives Oq1, O12 and O1j of the second level are first weighted with respect to the ob jective 01 (in this particular case the weightingsare 0.5, 0.25, and 0.25). The srrrnof the weighting factors for any one level must alwaysbe equal to Zwl: 1.0. Ncxt comes the weighting of the objectivesof the third level with respectto the rrrb-objectivesof the second level. Thus the relative weights of O111and O112 rvith respect to the higher objectives O11 were fixed at 0.67 and 0.33. The r c n t i r i n i n go b j c c t i v c sa r e t r e a t e d i n s i m i l a rf a s h i o n .T h e r e l a t i v ew e i g h t i n go f a n ohjcctivc at a particular level with respect to the objective 01 is found by r r r r r l t i p l i c a t i o no 1 ' t h c w c i g h t i n g f a c t o r o f t h e g i v e n o b j e c t i v e l e v e l b y t h e w e i g h t i r t gl ' l c t o r s o l t h c h i g h c r o b j c c t i v c l c v c l s . T h u s t h e s u b - o b j e c t i v eO t r r r , w h i c h h a s i r w c i g h t i n g o l 0 . 2 . 5w i t h r c s p c e tt o t h c s u b - o b j c c t i v eC ) 1 1 1 r )t h f e next h i g h c rl c v c l . h u s i r w c i g h t i r r g o l 0 , 2 .x5 ( r . 6 7 x0 . 5 x | = 0 . 0 9 w i t h r c s p c c t o O 1 . 122 5 Conceptualdesign 5.8 Evaluating conccpt variants tz3 6 LevelI 1 e9= = E E = C . E E E E E E t E 6:= 6: 0.09 + 0 . 2 5 + 0 . 1+ 6 0 . 0 9 + 0 . 0+4 0 . 0 4 + 0 . 0+8 0 . 2 5 : X w:i1 . 0 s-- = Figure5.50.Objectivestree with weightingfactors 6F S E >= - E E Such step-by-step weighting generally produces a realistic ranking because it is much easier to weight two or three sub-objectives with respect to an objective on a higher level than to confine the weighting to one particular level only, especially the lowest. Figure 5.76 gives a concrete example of the recommended procedure. Guideline VDI2225 tries to dispensewith weightings and relies instead on evaluation criteria of approximately equal importance. Weighting factors (2x, 3x) are, however, used for pronounced differences. Kesselring [5.24), Lowka. [5.31] and Stahl [5.56] have examined the influencesof such weighting factorson the overall value of the solution. Their conclusion was that they exert a significant influence whenever the variants to be evaluated have very distinct properties, and whenever the corresponding evaluation criteria have a great importance. :69= d = tn- I o€> _> S- = o 3F -i q) F ! C) -= 69> c;= (d €)s' AS? S C ; \ S = o o E O 3 Compiling parameters The settingup of evaluationcriteriaand the determinationof their importanceis followed,as a next step, by the assignment to them of known (or analytically determined)parameters.Theseparametersshouldeitherbe quantifiableor, if that is impossible,be expressed by statements framedasconcretelyas possible. It hasprovedvery usefulto assignsuchparametersto the evaluationcriteriain = c E E >5 seethat the verbalformulationof the evaluationcriteriastronslvrcscnrblcs t 5 c) EB ad '- E oa >F L an evaluation chart before proceeding to the actual evaluation. Figurc 5.5 shows an example of such a chart for an internal combustion engine. appropriate magnitudes entered in the relevant variant columns. The rcuclcr of the parameters. I n G u i c l c l i n cV D l 2 2 2 5 , b y c o n t r a s t ,e v a l u a t i o nf o l k r w si m n r c d i i t t c l yu p o n s c t t i n gu p o l c v a l u a t i o nc r i t c r i i r . -l= ol= += .sf .-. @6 N = = s 5 " 2..1l ! r U .9 =5 >9 E c :U 2 E .j v? c-. qJ =i EO ad ( oo IJ. 124 -5 Conceptualdcsign values 4 Assessing The next step is the assessmentof values and hence the actual evaluation. These 'values' derive from a consideration of the relative scale of the previously determined parameters, and are thus more or less subjective in character. The Valuesare expressed by points. Use-value analysisemploys a range from 0 to 10; Guideline VDI 2225 a range from 0 to 4 (Figure5.52). The advantageof the wider range is that, as experiencehas shown, classificationand evaluation 125 5.8 Evaluating conceptvariants Before anyone can assignpoints to the parameters of the variants, he must at 'value least be clear about the assessmentrange and the shape of the so-called function' (see Figure 5.53). A value function connectsvalues (v) and parameter magnitudes(lz), and its characteristicshapeis determined either with the help of the known mathematical relationship between the value and the parameter or, more frequently, by means of estimates [5.20]. 1.0 1,0 Value scale ysis Use-value ana Pts. I 0 Meaning absolutely useless solution UDI2225 I Guideline Meanins nr.l I 0 lncreasing linear lunction function linear Decreasing lncreaslng exponential function lunction exponenlial Decreasing unsatislactory veryinadequate solution 2 weaksolution 3 tolerable solution L adequate solution 5 satisfactory solution 6 goodsolution with fewdrawbacks 7 goodsolution B v. goodsolution 9 solution exceeding therequirement 10 idealsolution ,] justtolerable 7 adequate 3 good t very9000 (ideal) l ; i g u r e5 . 5 3 .C o m m o nv a l u ef u n c t i o n sf ,r o m [ 5 . 6 4 1 ; :x m ; i , y : v ; i It is useful to draw up a chart in which the parameter magnitudes are t'orrelated step by step with the value scale. Figure 5.54 shows such a scheme, rrrcorporatingthe point system of use-valueanalysisand also of YDI 2225. All in all, therefore, the assignment of a value and the setting up of an Figure 5.52. Points awarded in use-valueanalysisand guideline vDr2225 are greatly facilitated by the use of a decimal system and percentages. The advantagesof the smaller range is that, in dealing with what are so often no morc than inadequately known characteristicsof the variants, rough evaluationsare sufficient and, indeed, may be the only meaningful approach. They involve the following assessments: -far below average -below average - average - above average - far above average It is useful to begin with a search of variants with extremely good ancl q u a l i t i e sa n d t o a s s i g nappropriate points to them. Points 0 and 4 (or 10) sht o n l y b e a w a r d e di f t h e characteristicsare really extremc-that is, unsatislitc ()r vcry gootl (itlcirl).Oncc thcse extremc ptlints ltitvc llccn rrssignccl, r c n r i r i r t i t t gv i t r i i t t t l sl t r c r c l i r t i v c l vc a s v t o l ' i t i l t , Value scale Parameter magnitudes Fuel per Mass VOl2225 Use-value I I Simplicity Service I consumptionunilpower components anarysrs I lol I I km s/kWhlkg/kwll Pts I Ptt 0 0 360 3/,0 7 3 {. 5 6 7 B r{1 400 380 3.5 3.3 3,1 2,9 2 320 300 2,7 25 3 280 260 7,3 2,1 7 t0 1,9 7?0 1,7 100 I 'r extremely complicated complicated average simple 7A rn3 30 40 60 BO 100 r20 140 200 extremely simple 300 s 0 0. 1 0 t parilmctcrmagniludcrwith vnlucscitlcs 5,54.('hnrt corrclulirrg I -() 5 Conceptualdesign assessmentschememay involve strong subjective influences.Caseswith a clear, or even experimentally verified, correlation between the values and the parameters are few and far between. one such exception is the evaluation of machine noise, where the correlation between the value (that is, the protection of the human ear) and the parameter (noise level in dB) is clearly defined by ergonomics. The values v;j of every solution variant established in respect of every evaluation criterion are added to the list shown in Figure 5.51 (Figure 5.55). Whenever the evaluation criteria are of different importance to the overall value of a solution, the weighting factors determined during the second step must also be taken into consideration.To that end, sub-valu€v,; iS multiplied bv t h e w e i g h t i n gf a c t o r w i ( w v i i : w r . v i ) . F i g u r e 5 . 7 7 g i v e s a p r a c t i c a le * a m p t . - . r21 5.8 Evaluating concept variants = \ E '- E E = = E E E o=> o;s E E E f E * E E > oq oE \ >a E E E i .g E E E F i :s >*> ,E E E j- * \ 5 Determining overall value E -_?> s ;'s= 6= = ci Ef -- i Ss Socc j- E !o sE- o E vl D Unweighted: O V i' : \- v,; i: l N J;'9= N \ o d '<,= nn Weighted: - +E E ^ =s >-ro< d> ts> The sub-valuesfor every variant having been determined, the overall value must now be calculated. For the evaluation of technical products, the summation of sub-valueshas become the usual method of calculation but can only be considered accurateif the evaluation criteria are independent. However, even when this condition is only satisfied approximately, the assumption that the overall value has an additive structure seems to be iustified. The overall value of a variant i can then be determined. js s_ =S O W V' 1 : I w r . r ; 1= I w v ; 1 i-f O i:r 5"e E ___ .s l 6 Comparing concept variants N E O On the basis of the summation rule it is possible to assessvariants in several ways. -l = = -t= Determining the maximum overall value E In this procedure that variant is judged best which has the maximum overull value'. oo l1 E 0 E o@ L E 92] >F =x F C cd --. a6 OVi---> max or OWV1-+ max what we have here is a relative comparison of the variants. This fact is madc o f i n u s e - v a l u ea n a l y s i s . = N O <i I f f }a'l s Irl io Determining the rating I f i t r e l a t i v c c o m p i t r i s o no f t h e v a r i a n t si s c o n s i c l c r c ct ol l t c i n s u f f i c i c n ti r n d i t l r s t r l u l cn r t i r t l o 4 l i t v r t r i l u . tht a s t o [ r c c s t a b l i s h c d ,t l r c r rt l r c o v c r i r l l v i r l u c r n r r s l ;:l 'f G "l E (r, - - : c ir, c = r; \4. O ri 128 -5 Cionccptual design referred to an imaginary ideal value which results from the maximum possible value. n !rr.. L'll Unweighted: R, : r=l #: In somecasesit is usefulto derive the overallrating from thesepartial ratings and to expressit in numericalform, for instancefor computerprocessing.To that end, Baatz[5.1] hasproposedtwo procedures,namely: -the straightline method,basedon the arithmeticmean n nWV, ""ir v-u".I,w1 W e i g h t e dW : Rj: 2 i=1" n f.o*. I l1r'i and -the hyperbolic method which involves multiplying both ratings and then reducing to values between 0 and 1: R:VR,xR" The two methods have been combined in Figure 5.57. Where there are great differences between the technical and economic ratings, the straight-linemethod can be used to compute a higher overall rating than is the casewith low but balancedpartial ratings. Becausebalancedsolutions should be preferred, however, the hyperbolic method is the better of the two; it helps to balance great differences in rating by its progressive reduction effect. The greater the imbalance, the greater the reduction effect on the lower overall values. l?r,;1'z c ",faT t ou c Rr+Re R: Iw; .v1; If the available information about the properties of all the concept variants allows cost estimates,then it is advisableto proceed to a separatedetermination of the technical rating R1 and the economic rating R". The technical rating is calculatedin accordancewith the rule we have given-that is, by division of the technicaloverall value of the given variant by the ideal value-and the economic rating is calculated similarly, but by reference to comparative costs. The latter procedure is suggested in VDI 2225, which relates the manufacturing costs determined for a variant to the comparative manufacturing costs C,,. In that case, the economic rating becomesR": (ColCvariant). It is possibleto put, say, C, : 0 8 X Cadmissiblc or Co : 0.8 x C-ini-r. of the cheapest variant. If the technical and economic ratings have been determined separately, then the determination of the 'overall rating' of a particular variant may prove useful.For that purpose, Guideline VDI 2225 suggests a so-called s-diagram (strength diagram) with the technical rating R, as the abscissaand the economic rating R" as the ordinate (see Figure 5.56). Such diagrams are particularly useful in the' appraisal of variants during further developments, because they show up the effects of design decisions very clearly. 1.0 t29 5.8 Evaluating conccpt variants 4 E n nA ,<. E o Bol c ilh c \ 04 4\ 'rl -o*,.uli jdvetopmen' . sreps _ \ =--\ 0.2 Figure 5.56. Rating cliagranrlroln 02 04 06 Tcchnical raling B, 0B 10 vDr2225 02 04 06 Tcchnical ralittqfr' -----* 0B 10 I ) c l c r t t t i t t i r t i . to t l o v c r l t l l r l r t i r r gh v s l r l i g l t t - l i r r cl r r t t l l v p c r h . l i c t t t el h ' t l s ' 130 -5 Conceptualdesign Rough comparison of solution variants The method we have described relies on differentiated value scales. It is useful whenever the 'objective' parameters can be stated with some accuracy and whenever clear values can be assignedto them. If these conditions cannot be satisfied, relatively fine evaluations based on a differentiated value scale constitute a questionable and expensive method. The alternative here is a rough evaluation involving the application of a particular evaluation criterion to two variants at a time and the selection of the better in every case. The results are entered in a so-called dominance matrix [5.13] (Figure 5.58). From the sum of Varianl 112 r l 0 0 0 1 0 I I ,l I 0 I 0 4lsl6l7 0 I ,l 0 1 0 0 0 0 0 l 0 0 0 0 ,l 0 0 0 Sum 3 1 5 2 t 6 t 0 t 4 t 1 Rank 1 1 2 s r 1 l 7 l 3 l 6 1. better 0 . notbetter Figure5.58.Binaryevaluationof s o l u t i o nv a r i a n t sa, f t e r[ 5 . 1 3 ] the columns it is possible to establish a ranking order. If such matrices of individual criteria are combined into an overall matrix, an overall ranking order can be established, either by addition of the preference frequencies or by addition of all the column sums. While this method is comparatively easy and quick, it is not nearly as informative as the other procedures we have discussed. 7 Estimating evaluation uncertainties The possibleerrorsor uncertainties of the proposedevaluationmethodsfall i two m_aingroups, namely subjective errors and inherent shortcomings of t proceoure. Subjective errors can arise through: - Abandonment of the neutral position, that is, through bias and partiality. bias may be hidden from the designer,for instancewhen he comparcshis d e s i g nw i t h t h a t o f a r i v a l . H e n c e a n e v a l u a t i o n b ys u ' c r u l p c r s o n s ,i f p r from variuusdcpartnrcnts, is alwaysadvisablc, to lt is cquullyinrportant 5.8 Evaluating concept variants 131 to the different variants in neutral terms, for instance as A, B, C, rather than as 'Smith's Proposal' etc, since otherwise unnecessaryidentifications and emotional overtones may be introduced. Extensive systematisationof the procedure also helps to reduce subjective influences. - Comparison of variants by the application of evaluation criteria not equally suited to all the variants. Such mistakes arise even during the determination of the parameters and their associationwith the evaluation criteria. If it is impossible to determine the parameter magnitudes of individual variants for certain evaluation criteria, then these criteria must be reformulated or dropped lest they lead to mistaken evaluationsof the individual variants. - The evaluation of variants in isolation insteadof successively by application of the establishedevaluation criteria. Each criterion must be applied to all the variants in turn (row by row in the evaluation chart) to eliminate any bias in favour of a particular variant. -Pronounced interdependenceof the evaluation criteria. -Choice of unsuitable value functions. -Incompleteness of evaluation criteria. This defect can be minimised if one of the checklistsfor design evaluation appropriate to the relevant design phase is f o l l o w e d ( s e e5 . 8 . 3 a n d 6 . 7 ) . Procedure-inherentshortcomings of the recommended evaluation methods are the result of the almost inevitable 'prognostic uncertainty' arising from the fact that the predicted parameter magnitudesand also the values are not precise,but subject to uncertainty and to random variation. These mistakescan be greatly rcduced by estimatesof the mean error (scatter). With regard to prognostic uncertainty it is therefore advisable not to express lhc parametersin figures unlessthis can be done with some accuracy.otherwise, rt is preferable to use verbal estimates(for instancehigh, average,low) which do rrot claim to be precise. Numerical values, by contrast, are dangerousbecause t hcy introduce a false senseof certainty. A more detailed analysisof evaluation proceduresfor the purpose of judging tlrcir reliability and also for comparative purposes has been carried out by lelclmann [5.13] and Stabe [5.55]. The latter has also provided an extensive I'ibliography. If there is an adequate number of evaluation criteria, and if the rtrlr-valuesof a particular variant are fairly balanced, then the overall value will lrt' sLrbjectto a balancing statisticaleffect, and partly too optimistic and partly kro pcssimisticindividual values will more or less balance out. t Scarching for weak spots Wclrk spots can be identified from below-averagevalues for individual evaluallott critcriu. Carcful attention must be paid to them, particularly in the case of t i s i r r gv i r r i i r r r t sw i t h g o o d o v e r a l l v a l u e s , a n d t h e y o u g h t i f p o s s i b l et o b e t i t l c d d u r i n g l i r r t h c r c l c v c k r p m e n tT. h c i d c n t i f i c a t i o no f w e a k s p o t sm a y b e a t c d b y g r l ; r l r s o l ' t h c s u b - v a l u c s - l i r r i n s t a n c c , b y t h c s o - c a l l c c vl a l u e illustrntcdirt liigurc.5..59, In it. thc lcngth*ol thc barscorrcspond to thc -5 Conceptualdcsign l -)L value and the thickness to the weightings. The areas of the bars then indicate the weighted sub-values,and the cross-hatchedarea the overall weighted value of a solution variant. It is clear that, in order to improve a solution, it is essentialto improve those sub-values that provide a greater contribution to the overall value than the rest. In Figure 5.59, this is the casewith the evaluationcriteria that have an above average bar thickness (great importance) but a below average bar length. Apart from a high overall value, it is important to obtain a balanced value profile, with no serious weak spots. Thus, in Figure 5.59, variant 2 is better than variant 1, although both have the same overall weighted value. Varrant2. )WV: 2 Variant ]WV' r I A N1. II: U WVI: t-I- liqf 'n,,.. ll uy1.'..' 1l ' wvz2 lf l,t ,'ll:' " WVn. I tyya: ... ' ll t' llVql ./ ' '".) ' ||vot ' ]I ,wv62, i I t I t: I I t . 10 = =- l0 of two variants()w, : l) Figure5.59.Valueprofilesfor the comparison There are also casesin which a minimum permissiblevalue is stipulatedfor al} sub-values-that is, any variant that does not fulfil this condition has to be rejected, and all variants that do are developed further. In the literature this 'determination of satisfactorysolutions' procedure is described as the [5.64]. l -tJ 5.8 Evaluating concept variants Use-value analysis VDI Guideline2225 Identification of objectives or evaluation criteria for the evaluation of concept variants with the aid of the specification and a checklist Construction of a hierarchicallyrelated system of design objectives (objectivestree) based on the specification and other general requlrements Compilation of important technical characteristicsand also of the minimum demandsand wishes of the specification Analysis ol the evaluation criteria for the purpose of determining their weighting to the overall value of the solution. If necessary, determination of weighting factors Step by step weighting of the objective criteria (evaluation criteria) and if necessaryelimination of unimportant criteria Determination of weighting factors only if evaluation criteria differ markedly in importance Compilation ol parameters applicable to the concept variants Construction of an objective parameter matrix Not generally included Assessment of the parameter magnitudes and assignment of values ((l-10 or (I-4 points). Construction of objective Assessmentof value matrix with the helo characteristicsbv points (0-,1 points) of a points system or value functions; 0 10 polnts Determination of the overall value of the individual concept variants, generally by reference to an ideal solution (rating) Constructionof a use value matrix with due regard to the weightings; determination of overall values by summation D e t e r m i n a t i o no f a t e c h nical rating by summation, with or without weightingsbased on an ideal scllution.If necessary determination of a n e c o n o m i cr a t i n g b a s e d on manufacturing costs Comparison ol' (oncept varmnts Comparison of overall use-values Comparison of the technical and economic ratings. Construction of an s-(strength) diagram Ii.slitrtuliott of evulttuliort tutccrtuinlies Estimation of objective parameter scatter and use value distribution Constructionof usevalue proliles Not explicitly included Step 5.8.2 Comparison of evaluationprocedures Table 5.3 lists the individual steps in the evaluation procedures we have described and also the similarities and differences between use-value analysis and Guideline VDI 2225, which are based on similar principles. The individual steps of use-value analysis are more highly differentiated and more clear-cut. but involve more work than those of Guideline YDI 2225. Tha latter is more suitable when there are relativelv few and roughly equivalc evaluation criteria, which is frequently the case during the conceptual phasc ( 5.S.3), and also for the evaluation of certain form design areas during embodiment phase (see 6.7). 5.8.3 Evaluation during the conceptualphase 'l'irblc . 5 . 3s u l l l s u l ) t h c g c n c r a l c v u l u a t i o np r r l c c r l t t r cl.) r r r i t t gl h c c o t t c p h i r s c .t l t c i t t d i v i t l t r r rsl t c p s s h o t t l t lb c i t s l i r l k t w s : ,\curclt .l?tr wcuk spots l i r r t h c p u r p o s eo f i r r r l l r o v i n gs e l e c t c c l v ; r ti i r t t t s Identification of characteristicswith a few points only 1 l r l 1 .\ I l r r r l i l i t l r r l r ls t t ' p s i n t ' r ' : r l t r : r t i o r tn, t t r l t ' o t t t l r i t t i r r n rl r t ' t w t ' r ' r tt t s c - v ; r l r t ci t n i t l v s i sl r t t t l Iurrlt'ltttt'Vl )l l.ll5 134 5 Conceptualdesign I dentfy ing evaluation criteria This step is based, first of all, on the specification.During a previous selection procedure (5.6) unfulfilled demandsmay have led to the elimination of variants that were found to be unsuitable in principle. Further information was gathered subsequently by the firming up of solution proposals (combinations of principles) into concept variants. Hence it is advisable, with all the newly acquired information, to establish first of all whether all the concept variants to be evaluated still satisfy the demands of the specification. This involves a new yes/no decision-that is, a new selection. It is only to be expected that, even at the present, more concrete, stage,this decision cannot be made with certainty for all the variants unless much further effort is applied, which the designermay not be able to provide at this stage.At the given state of information, it may only be possibleto decide how likely it is that certain requirements can be fulfilled. In that case, the requirements in question may become evaluation criteria. A number of requirements are minimum requirements. It has to be established whether or not these should be exceeded. If they should, further evaluation criteria may be needed. For evaluation during the conceptual phase, both the technicaland also the economic characteristicsshould be consideredas early as possible [5.28]. At the firming-up stage, however, it is not usually possibleto give the costs in figures. Nevertheless,the economic aspectsmust be taken into consideration, at least qualitatively, and so must industrial and environmental safety requirements. It has been suggested[5.37] that a seriesof headingsbe adopted in accordance with the embodiment design checklist (see 6.2), and also in keeping with other proposals, and that evaluation criteria be derived from them (Figure 5.60). Every heading in the checklist must be assigned at least one evaluation criterion. These criteria must, moreover, be independent of one another in terms of the overall objective, so as to avoid multiple evaluations. Consumer criteria are essentially contained in the first five and last three headings; producer criteria in the headings'embodiment', 'quality control', 'assembly'and 'costs'. Evaluation criteria are accordingly derived from: l. Requirements of the specification - Probability of satisfying the demands (how probable, despite what difficul- ties). - Desirabilityof exceedingminimumrequirements(how far exceeded) -Wishes (satisfied,not satisfied,how well satisfied). 2. General technical and economic characteristics(to what extent prc how satisfied). (see Checklist headings for design evaluation during the conccptual Figurc -5.60). I ) u r i n g t l t c c o n c c p l u i r lp h a s c ,t l r c t < l t a ln u r r r b c ro l ' c v i r l r r i r l i o n c r i t c r i i rr r r u s l b c l o t t l r i g l t t t t r t l 5 c r i l c r i i r i r r c u s u l l l y c t t o u g l t( s c c l r i g u r c 5 . ( ) 6 ) , 5.8 Evaluating concept variants 135 Mainheadings Examples Function Characleristics ol essential auxiliary lunction carriers thatlollowofnecessity lromthe principle chosen s0lution orfromthec0ncept variant principle Characteristics ol theselected orprinciples inrespect ofsimple andclear-cut lunctioning, adequate elfect, lewdisturbing factors principle Working Embodiment Safety Small number ol components, lowcomplexity, lowspace requirement, n0special problems withlayout orformdesign Preferential treatrnent (inherenily 0f directsalety techniques safe), noadditional safety measures needed, industria guaranteed andenvironmental safety Ergonomics Satislactory man-machine relationship, n0strain good 0rimpairment ofhealth, form 0esrgn Production production Fewandestablished methods, noexpensive equipment, smaI number of s mplecomp0nents Quality control Fewtests andchecks needed, procedures simple andreliable Assembly Easy, convenient andquick, nospecial a dsneeded Transport Normal means ol transport, norisks 0peration Simple operation, longservice life,lowwear, easy andsimple handling Maintenance Little andsimple upkeep andcleaning, easy inspection, easy repair Nospecial running orolher associated costs, noschedul ngrisks Costs Irigure5.60.Checklistwith mainheadings for designevaluationduringthe conceptual 'hase i l| t i ghting the evaluqtion criteria l hc evaluation criteria adopted may differ markedly in importance. During the (()nceptualphase, in which the level of information is fairly low becauseof the rt l:r{ive lack of embodiment, weighting is not generally advisable. It is much more advantageous,in the selectionof evaluation criteria, to strive l.t ittt ilPProximate balance, ignoring low-weighted characteristicsfor the time l', irts. As a result, evaluation will be concentratedon the main characteristics .rrrrlhcnce be clear at a glance. Absolutely distinct features, however, which t,rrrnot be ignored until later, must be introduced -with the help of weighting l , t tt o r s . l',rt 11111i,,f puratneters tl ltirs provccl uscful to list the identified evaluation criteria in the sequenceof lhc clrccklist hcadings (Figurc -5.60)and to assignthe parametersof the variants t l l l ( ' r l l . W h i r t c v c r c l u a n t i t a t i v ei n f o r m a t i o n i s a v a i l a b l ea t t h i s s t a g es h o u l d a l s o t t t c l r r t l t ' cS l .t t c l tr l t r i r t t l i l r r t i vccl a t ag c n c r a l l yr c s u l tf m m t h e s t e pw e h a v e c a l l e d l l F t l l ) t r l t ( )c o l ) c c l ) lv i t r i l u t t s ' .I l o w c v c r . s i r r c ci t i s i r n p o s s i b l et o q u a n t i f y a l l t ' l t l r r i t c l c r i s i i rcl st t r i r t gt l t c c o r t c c p l u i rplh i r s c l.h c t l r r l r l i t i r t i vi rcs p c c tssh o u l cbl c i n l o w o r d s i r r r t lc o r r c l i r l c r lw i t h t h c v r l u c * - u l c , r36 5 Conccptual clesign Assessing values Though the attribution of points raisesproblems, it is not advisableto evaluate too timidly during the conceptual phase. Those using the 0-4 system proposed in VDI Guideline 2225 may feel the need to assignintermediate values,particularly when there are many variants,or when the evaluatinggroup cannot agree on a precisepoint. It may prove helpful in such casesto attach a tendency sign ( J or t ) to the point in question (see Figure 5.96). Identifiable tendencies can then be taken into account when estimating the evaluation uncertainties. The G-10 scale, again, may suggesta degree of accuracythat does not really exist. Here, argumentsabout a point are often superfluous. If there is absolute uncertainty in the attribution of points, which happens quite often during the evaluation of concept variants, the point under considerationshould be provided with a question mark (seeFigure 5.96). During the conceptual phaseit may prove difficult to put actual figures to the costs. It is not therefore generally possible to establishan economic rating R" with respect to the manufacturing costs. Nevertheless, the technical and economic aspectscan be identified and separatedqualitatively, to a greater or lesserextent. The'strength diagram'(see Figure 5.56) can be used to much the same effect (see also Figures 5.61 to 5.63 which are for the test rig shown in Figure 5.45). In many casesa classificationbasedon consumers'and producers' criteria has a l s op r o v e d u s e f u l .S i n c et h e c o n s u m e r s ' c r i t e r i au s u a l l yi n v o l v et e c h n i c a lr a t i n g s while the producers' criteria involve economic ratings, it is possible to proceed \ariant tecnn\ cfltefla ,ffi t ) S m a l l d i s t u ri l-) 0ance0lc0upIingkinematics 2)Simple operation 3)Easyexchange of coupling 4)Functional salety 1) s)Simple construction Total ^ ", Total 20 ,iz_ r37 .5.8Evaluating concept variants 3)Shorttesting time 't Possibility ol manulacturi ng tn own w0rKsn0p Total ^ "e Total 16 'u,nT, J lLl FI_T-I rL 'fr( *'ilS 1)lowmaterial costs 2)lowreassembll costs t 2- -T* 2 ? 3 2 2 l/. 07 L t 3 \ t, t, 3 3 t, t, L 3 3 l t, ? 2 2 7 3 l7 17 17 16 085 085 0.85 0.80 'll 0.55 2 7 I 3 3 3 2 2 t, 3 3 7 3 3 3 3 l 7 s s l1 13 13 I 0.56 0.56 0.69 081 081 056 (1)Austenitic shaft(2)Torque measuring shaft must bemoved Irigure-5.62. Economicevaluationof the remainingconceptvariants,seeFigure5.46 ttv 15t754 0.8 I 0.4 L t, L 2 Figure 5.63. Compari.cn of the technical and economic ratings of the concept variants in Figures 5.61and5.62 tl -M L t, (2) -lL 3 il) 3 (1)Totque chanqes withaxialdlsplacement ol pinion l r i g r r r c5 , ( r l . ' l c t l t t r i c rtr' lr ' r r l r r i r t i t r lrtrh c r c n u r i r r i n g c o n c ctp' itr t i : r t t l s c. c I r i g t r r c5 , J ( r 0.4 06 10 t , , r r s i r n i l e rcr l a s s i f i c a t i o nt o t h e o n e m e n t i o n e da b o v e . T h e r e a r e t h r e e p o s s i b l e l rr rn t s o f r c p r e s c n t a t i o n ,n a m e l y : lcchnical rating with implicit economic aspects(seeFigures 5.77 and 5.96); or sr'prrratctechnical and economic ratings (see Figures 5.61 to 5.63); and r r t l t l iilo n i r l c o m p a r i s o r o . tf c c l n s u m e r sa' n d p r o d u c e r s 'c r i t e r i a . \\'hich onc is choscn depends on the problem and the amount of information lr;rilirtrlc. )rtrrtrt i rri tt g rtyc rul I yuI rtc t l c t c r r r r i n i r t i oonl t h e o v c r a l l v i r l r r ci s u n l i r t t c r< l f s i m p l e a d d i t i o n s ,o n c e s h r r v cb c c l r i r s s i l l n e dt o t h c c v i r l u i r t i o rct r i t c r i i ri r n r ll h c v i r r i i r n t s I. l ' . b c c a u s c l h c c v a l u i t l i o n u n c c r l i r i n t y .i t i s o r r l y p o s s i b l cl ( ) i l s s i l l ni r r i r n l l co l p o i t t t s t o 138 5 Conceptualdesign r39 - 5 . 9 E x a m p l e s o f c o n c e p t u a lr l c s i u n individual variants, or if tendency signsare used, one can additionally determine the possible minimum or maximum overall point number and so obtain the probable overall value range (see Figure 5.96). in the caseof assess the probabilityandmagnitudeof the possiblerisk, especially importantdecisions. Comparing concept v ariants 5.9 Examplesof conceptualdesign A relative value scale is generally more suitable for purposes of comparison. In particular, it makes it fairly simple to tell whether particular variants are relatively close to or far from the target. Concept variants that are some 60 per cent below the target are not worth further development. Variants with ratings above 80 per cent and a balanced value profile-that is, without extremely bad individual characteristics-can generally be moved on to the embodiment design phase without further improvement. Intermediate variants, too, may, after elimination of weak spots or an improved combination, be releasedfor embodiment design. If often happens that two or more variants are found to be practically equivalent. It is a very grave mistake, in that case,to base the final decisionon such slight differences. Instead, evaluation uncertainties,weak spots and the value profile should be looked at more closely. It may also be necessaryto firm up on suchvariants during a further step. Schedules,trends, company policy and so on must be assessedseparatelyand taken into account [5.28]. Estimating evaluation uncertainties This step is very important, especiallyduring the conceptualphase, and must not be omitted. Evaluation methods are mere aids, not automatic decisionmechan-. isms. Uncertainties must be determined by the proceduresexamined in 5.8.1.7. At this point, however, only such informational gaps need be closed as bear on the favourite concept variants (for example, variant D in Figure 5.96). Searchingfor weak spots During the conceptualphase, the value profile plays an important role. Variants with a high rating but definite weak spots (unbalanced value profile) may provc extremely troublesome during subsequent development. If, because of an unidentified evaluation uncertainty, which is more likely to occur in the conceptualthan in the embodiment phase, the weak spot should make itself fclt later, then the whole concept may be put in doubt and all the developmentwork may prove to have been rn varn. In such casesit is very much less risky to selecta variant with a slightly lr rating but a more balanced value profile. Weak spots in favourite variants can often be eliminated by the transfcr better sub-solutionsfrom other variants. Moreover. with better information. it 'Ihus possibleto searchfor a replacementof the unsatisfactorysub-solution. c r i t c r i aw e h a v e l i s t e c o l l a v e da n e s s e n t i arl o l e i n t h e s c l c c t i o no f t h e b c s t c r ). hcn csti v a r i a n ti n t h c p r o b l e r nc l i s c u s s ebdc l o w i n . 5 . t ) . 2( l i i g u r c . 5 . ( ) 6W c v a l u i t t i o t tu n c c r ' l r i l l l i c sa n d i r l s oi n t h c s c : a r c ht i r r w c t t k s p o t s i l i s a c l v i s i r b l c In this section we shall be examining examplesof conceptual design involving the flow of energy, material and signals,and providing clear illustrationsof the procedure and applications. 5.9.1 Rig for testing impulse torque loading of keyed connections Step 1: Clarifying the task and elaborating the specification The problem was the development in a researchinstitute of a test rig capable of loading keyed shaft-hub connections impulsively with definite torques. Before drawing up a specification,a number of questionshad to be answered: - What is meant by impulse loading? -What impulse torque loadings actually occur in rotating machines? - What load measurements are possible and useful in the case of keyed connections? While the first question led through physical considerations to the impulse concept and to the possibleeffects of the collision of two or severalmasses,and rvhile fundamental studies,subsequentlypublished [5.32], helped to answer the third question, the answerto the secondquestioninvolved considerableresearch u,ork. Since the test rig was meant to apply and measureimpulse torque loadings ,rl defined magnitude and rate of increase (Figure 5.64), it was important to torque Steady Decaying torque Iru,adlustable sleeplall olf afterImax adjustable rateol increase Figure 5.64. Adjustable settings for impulse torque loading: rise time, maximum value and duration ,rrrrlvscthc torque loadings that occurred in practice in terms of the maximum t.rt('sof incrcusc cllldt. To that end, the torque/time characteristicsof lathes, (tirnc transt.nissions,agricultural machines and rolling mills were considered 'l'he rnirxirnuntrilte of increasefor the caseof non-steadyoperationswas l\ \ f l. f t r r r r rttol b c r l 7 7 c l l - l 2 - 5 x l 0 r N m / s a n d t h i s w a s u s e d a s t h e b a s i s f o r t h e t e s t r i g ; h ' rl r r r r n i r r r c c A f t c r l l t c s c l t l c l i r n i t t i r r ys t u d i c s . i t w i r s p o s s i b l c l o c l n r w u p t r . s p a < ' i f i c u t i o n ( l ; t g t r r c . 5 . ( r . 5 ) . ' l 'rlcr tcg r r i r c r r r c nl rtrsc i r r r i l n F c ti ln i r c c o r t l i r n cwci t h t h c g u i c l c l i n c s d i r c u s s c di r r 4 , l . l , 5 C o n c e p t t r a ld e s i g n 140 5 . 9 E x a m p l e s o f c o r r c c p t u a lc l c s i g n r41 it lssue 10.6197 TUBerlin SPECIFICATION TUBerlin lnpulse-loading lesl tig D )hanges W < 100mm t0betested: ofshaft Drameter onDIN6885) based (Keydimensions direction in longitudinal to bevariable loadtake-ofl Hubside Measurements: Torque before andaltertestconnection Surface stress overlength ol c0nnecti0n andkeyface D W Measurements to berecorded oPtional direction Loading oPtional Torque ' inPut lromthehubtotheshalt tothehub ortromtheshaft D D 3s toratleast t0 bemaintalned torque Maximum small loading: of Frequency (reason: measurement PrinciPle) aslaraspossible system in shaft-hub-key vibrations Suppress to theloadupto 15000Nmcorresponding adlustable torque Maximum 0fa shattol 100mmdiameter capacity carrying mustbepossible torque altermaximum lallolfoftorque Steep adjustable be mu$ dl/dl in torque, Rateof increase dTldt:125x 103Nm/s Maximum W qurck (e.9. 0peration ol testrigassimple aspossible andsimple reassembly ofrig) W principle (littlen0se,dirt,vibration) Working ol rigenvironmentally sound Production andqualitycontrol D D Forces t0rsi0n bypure ofshatl-nub Loadinq moments) andbending byshear (i.e.nolinfluenced points Measurement easily access ble Saletyandergononics withshattstatlonary to beapplied Loading onlY in onedtrection Loading D W Resp. D Kinenatics D D Requirements mustbeheldin positi0n Testconnection D Page2 lmpllseloadln0 Ieslrlg Signals Geonetry W D SPECIFICA'IION lor lndividual manufacture 0fallparts Quality ofkeyed shalthub to DIN6885 andofshalt endsingear boxes, electric motors, etc:to DIN748,Sheets 2 and3 W W Manufacture ol testrigin ownworkshops parts Bought-out possible andstandard wherever W Testrig:smalldimensions lowweight W Nospecial foundations w Fewandsimple pans working W Minimum maintenance Assembly andtransport andmatntenance 0peration mustbereproducible curve Torque{ime mustbe olc0nnecti0n destruction andif necessary delormatron Plastic Possible Costs Manufacturing costs< 20,000 DM(seeresearch application) EnergY Schedules < 5 k[380V consumption Power D Materrals Shaltandhub:45C 2 86 . 7 3 Conclusion phase: ofconceptual Juiylg/3 phase: Conclusron ol conceptual 20July1973 Rcplaccs tc$lri8 Figure5,65.SJxcificnlionof irnpulsc-loldin8, l,(r5(colrlirructl) ssuo oi Herr Militzer t42 5 Conceptualdesign Step 2: Abstracting to identify the essentialproblems in Figure5.6-5 Table 5.4. Abstractionand problemformulationbasedon specification Resultsof first and second steps - Shaft diameter to be tested <100mm Hubside load take-off to be variable in longitudinal direction Loading to be applied with shaft stationary Adjustable pure iorque loading of test connection up to 15000 Nm (max.) Torque to be maintained for not lessthan 3s Torque must be able to fall off steeply Maximum possible rate of torque increase dTf dt : 125 x 103 Nm/s Torque-time curve reproducible Magnitudes of 16"1,,r",7r,,x",tlfld o to be measured and recorded Resultsof third step - Provide impulse torque loading for keyed shaft-hub connections adjustable in respect \ of magnitude, rate of increase,duration and fall off Torque test to be conducted with stationary test shaft Resttlts of fourth step - Provide adjustable dynamic torque loading for component testing Allow measurement of input loading and component stresses Results of Jifth step 'Apply dynamically variable torques with simultaneous measurement of loading itnd component stresses'. Step 3: Establishing function structures The drawing up of function structuresbegins with the formulation of thc ovcrall function, which results directly from the problem formulation (Figurc -5.66). E,ssentialsub-functions to fulfil this complex overall functitln chicfly invt t l t c f k r w o f c n e r g y i t n c l ,f < t rm e a s u r e m c n t st,h c f l t l w t l l s i g n i r l s : - - ( ' h t t r t 1 1i tct p t t l o t c r , q ' ,itn' t t l l t l a c l( t t l r t l t r c ) . ( ' l t u n g ai t t ; t t t lc l t c r g V i t r l o l r t r x i l i i r r y( t t ( r ( . v l o r C o n t r r t l t t t t t ' l i o r l . energy Detormation Energy loss Energy Mainenergy (mech., electr.)Auxiliary energy In accordancewith 5.2.2, the specificationmust be abstractedstep by step to identify the essential problems: Steps 1 and 2. Omit requirements (demands and wishes) that have no dirept bearing on the function and the essentialconstraints. Step 3. Transform quantitative data into qualitative data and reduce to essential statements. Step 4. Generalisethe results of the previous step. Step 5. Formulate the problem in solution-neutral terms. T a b l e 5 . 4 s h o w s t h e r e s u l t o f t h e s es t e p s - 143 5 . 9 E x a r n p l e so [ c o n c e p t u a l t l c s i g n lVlaterial Slgnals Shaft Hub Key Startof control sequence Drive on-ofl 1 L0a0 shalthub connection torque withdynamic loading andmeasure andstresses .t Shalt Hub Key shaflhub I(4 belore connection shatt-hub I(0 after connecti0n o(x,y,z,t)on lhe surfaces c0ntact testrtg Figure5.66.Overallfunctionof impulse-loading - Store energy. Control load energy or load magnitudes. Vary load magnitude. Channel load energy. Apply load to specimen (shaft-hub connection). - Measure load - Measure stress (strain). In a step by step elaboration, the combination, change of sequence and addition or omission of these sub-functionsresult in several function structure variants.Figure 5.67 recordsthese in the order in which they were developed. In this particular problem, the measurementtasksdo not determine the concept so the function structure is developed for the flows of energy and material alone. liunction structure variant 4 is followed up in the search for solution principles b c c a u s e i t i n c l u d e s t h e s u b - f u n c t i o n so f v a r i a n t 5 , w h i c h s e e m s e q u a l l y 'change promising. A finer breakdown of individual sub-functions-for instance 'change 'channel 'change force back force' and ellcrgy' into torque into force', irrto torque'-will not prove helpful before the searchfor solution principles has lrccnstarted. Step 4: Searching for solution principles to fulfil the sub-functions. lrr thc search for solution principles, the following of the methods listed in \r'ction 5.4 are the most commonly used: Arnclng conventional aids: l . i t c r a t u r es e u r c h e s Analysis clf an existing universal test rig A r n o n g r n c t h o d sw i t h a n i n t u i t i v e b i a s : llririnstorming z \ r n o n g n r c t h o t l sw i t h a d i s c u r s i v eb i a s : schemes S v s t c r n l r t i cs c a r c hw i t h t h e h el p o f c l a s s i f i c a t i o n s l trfaces l t n c l t l r t l t i t l t l s o f c r r c r g y , V r r l r i r l i o t to l l h c t v l . l c s l s c o l l i r r c c t r i t t t s l i l r t l t l t t i on o l i r r l c s i g r rc l t l i t l o g t t c I llrilrcilllcs 'Ilre c o t t t l l i t t c tbl v t t l c l t t trsl l ' i tc l i r s s i l i c a i t r c t l i s c o v t e e t l l h t t s s o l t t l i o t lt l r i r t c i l l l c s l t o t r s e l r c r r r c( l r i g r r r c . 5 . ( r t { ) .l r r r r c { t r i ( t n lril l r p t t c c . o t t l v t l t c t l t t t s t i t t t p o r t l t l t t -5 ConcePtual design 44 145 5.9 Examples of conccptualdesign lttl ll tttl otttl llrl =E s6 structure Function Comments rn \-|/ --r- --i--- S, | gyllow ll with Energy rols signals. control 'control' r i f - - l r 'Change' l e o n t.r_ql ft nge'and Tl L;nanse E!n#l,lX.t ffi il beinterchanged can 0e .- Mi- ffi -"-== !=E o \\:2_-19 = [1*6-lit1t'oss) specimenl-._ __[-----l = o IU': il - ' L _specimen _ [iri-] F* tl| Y- gad= >- 3 o9p ns* a I fA{bl O /';\ a./ Eo EE O.= -,-- i | -_l_ --)-y __rEl-.t r __L_ l C o n t rlo' t li '= '- 9^a i:, 6 ,/ d9 \ ( ./ ,O tv s.i lstorel subdivided j"i' i - - - - - p ; ' s ; Sub-lunction'change' 'load'to belore inserted thecontrolled change into'torque energy : ----r f I ;lu-m.q.ffi-mmffii' I G', I- [ ernto r al--1 nse -],,[5 L0a0 f--------;.- I ii llorquel+ ,1, I i it = tr \-\ - ''f E E J I ttl l:l fr Iog .- c o{ L E- o I l_ | __F-_ t . l l n c r e a s el ul n a n Q e l m a o l l i -al control unanse r corp L0a0 n ll f-] ,JlJi,, l*l [* thesystem S, i _loulside ,-,.-' Mt 5 . ( r 7 ,l i r n c t i o n s t r u c t u r ev i t r i i t t t t st t t r i l tt r t rs t c l ) - b y ' s l c p ]f- t I E o >!: Ia= U; 8-_-8 oFi lli I g60l/\l oEor(:-): ;- llo ul I V = E e E = == 't -t >g ru o c E WE _ r----rl I I ^nN \!-/ :,j lt h\ @ ll E =€lN O L *f { l I. B -l .9 \ \:.r / e=- ii (ftroua aOueq3 il t r" 'ENI I ttl at I ffi: o- o t i _= 1; (hE lc o^ o c E E lA- 9 6 E -a: J c l "l l+ -.1 :.NIN t-{ =.1 ' E \11d a: : i-tr{f OE : lntn a -SE E =b c t N ! , ' ! N € 3 o'_ l o r\'I Et Ll{.J ill ,^ t\ .!: o U T-'l r A lt lEN l I c f ( s) :>4 EOF: d I U b 9 EO-I F q 4T\ o =o | tM', o.-6 c E{ .9 o _l --.1\ s >o €+\ 5\9 - .E .1, E { mt + / o /,/> .s2 __Jif 5 /--4 './e A-' o t l Ir .-l lv\ il ;y ,/l--.r-l '{t\6 dTt o e<,ti / vl IY ll o c o O <o c {lC U .= Y _rl il oc =./ / - 12 32 UU 'Srore prosram I !] .z @ E/ a/ ll O> .9 t R ' i ' i l ] l l n"l::":l-*| t""*l raqn " 0 1L0a0I ,* nChanqe H "Store - " F-|'"::_:: ll-.-|, c o m p ll a r d t r f*l n eLl - l l t I lenersv i ,, !1"'-'" 1 l t g5- € -r-- urrryur' ! =o- !o I f^---___lr l-ioaJ_-|i t, urange $r------l:.'::: t--lProsrl electr Inputenergy :-tNl - c_- lM | -.-- $TN Fa-] GO energY itoreadditional Store rndIprogramme and qdd.'increase' Add. r, l 'switch', i.e. \dd. 'elease energy r i--__l 'lncrease'function E I E c .9 E g-{d (i U i 1l E !? s: \:,/ o titiI r---f g lrits | 6 +,xI L ..tl q, /tvl E/ ll>l il = T,rs g'fil,/l //?\=> (soecimen) | !4' I nputenergy ---t -T--l into int more ;hanged chanqed r r i l-----l lconr.or lablelntermedtate ;ontrollable control --l Chanoe Hrnaqn,tuoe n lnergy l'landlime r l l .----,-*-trir,r'*'qy ee* o (/) g 'l ..,I = tr^ :_, >E-l o =--o ii : E € H-f Yn. c:oil X Irl -rrl co l* l/\i EJ | 6 r; | S s F , : , , i .s 3 |J. 146 5 Conccptual design sub-functions and solution principles have been included. Solution principles that are useless from the start are omitted from, or crossed out in, the classificationscheme.Especiallywith mechanicalcontrol principles, it is possible to derive new ones by systematic variation of surfaces and motions. In the present example, energy provides an important classifyingcriterion with respect 'change energy' and all the to the solution principles fulfilling the sub-function other sub-functionsas well. 'release energy', the solution principles involve various For the sub-function 'load' they involved clutches and locking mechanisms; for the sub-function various shaft-hub connectionsand also rigid couplings;and for the sub-function 'measure', which is not a concept determinant, they involve extensometerstrips and inductive and capacitativetransducerswith electronic amplifiers. 5 . 9 E x a r n p l e so f c o n c c p l u u l t l c s i n r r principle Solution Subjunction 1 c o 3 t, Step 6: Selectingsuitable combinations \ It was pointed out in 5.6 that, with a large number of variants, it is advisable, even before further firming up, to make a preliminary selectionso that detailed design drawings and calculations are not made for other than promising combinationsof principles. Figure 5.70 gives an example of this type of.selection procedure: only four of the seven combinations are pursued after selection. c 3 2 \ electr.-mech Z 5 i"\ it. Nli t! \ I 6 Control energy in resp. of magn.andtime l varyenergy component 7 B I n ,/l I \ m e c nm . -e c n l o Storeenergy 6 l\\ ,..T\ electr.-hydr. m e c nh. -y d r . 5 L .4v, -\. .1 'rv Nsl V . l Step 5: Combining solution principles to fulfil the overall function The sub-solutionsobtained during the last step must now be combined into overall solutions. In the present example, this was done with the help of the same classification scheme. From the field of solution principles, various combinations of principles are derived by combining solutio\ to one subfunction with solutions to the neighbouring sub-functions. Here, function structure variants 4 and 5 serve as the basis for the combination, the sequenceof sub-functions being slightly varied. In this process, the compatibilities and technical possibilitiesare considered,not systematicallybut by discussion.In this context it is often helpful to draw up compatibility matrices (Figure 5.43). Figure 5.69 showsthe areasof feasiblecombinationsof principles. I4l ".,1 t2 \-l:._\ <\ b- "gz =::l: '"\ A i\\ Figure5 '69. Combinationschemeshowingsevencombinationsof solutionprinciplesin accordance with Figure5.68 Variantl:1.1-5.3-6.5-3.4-3.7 V a r i a n t 2: 1 . 1 - 7 . 4 , 5 . 1 - 7 . 4 - 6 . 2 - 3 . 7 V a r i a n t 3 :1 . 1 - 5 . 1 - 3 . I - 6 . 1 - 3 . 7 V a r i a n4 t : 2 . 1- 6 . 8 - 4 . 1- 3 . 2 Variant5:6.7_1.2_7.3_3.7 V a r i a n6t : 6 . 7 - I . 7 _ 7 . 3 _ 3 . j V a r i a n t T: 6 . 7 _ l . I _ 7 . 4 ('alculation steps: Time needed to reach the maximum torque at the required rate: o': 15 x 10: tx * tor: o ' 1 2S Force at the end of the loading lever: x F n r u * : T : 15 10:: 1 7 . 6 x 1 0 : N 0.85 Step 7: Firming up into concept variants Before the most promising concept variant can be discovered, the selcctccl combinations of principles must first be prepared for evaluation. To that cnd. the designer will make a number of layout drawings (Figures 5.71to 5.74). Often a line diagram will not, however, be sufficientfor judging the functional efficacy of a solution--calculations or model experiments may be needcclat well. As an example, take the cylindrical cam and the flywheel in variant V2 u s c d t o p r o v i c l et h e i m p u l s e t o r q u e l o a d i n g . ( ' a n t h c c y l i t t t l r i c i r cl u r n s h o w n i n F i g u r c 5 . 7 5 p r o c l u c cl l t c r c t l u i s i t c l t irrcrcirsc ol d ' l ' l t l t- - I 2 . 5x l 0 r N n r / s i r t r d t h c n l f l x a l l l u n l t ( ) r ( l u c 7 , u u . =l 5 x l ( l r N t r r ' l 'ift.- lhe l.ading lever is treated as a weak cantilever spring with the end moving l l t r t r t r g ha d i s t a n c eo f h : 3 0 m m w i t h a f o r c e o f F - o " l n s u c h a w a y that th; p t ' rr r r i s s i b l eb e n d i n g s t r e s si s n o t e x c e e d e d . l r r n g c n t i a lv e l o c i t y o f t h e c y l i n d r i c a l c a m : f*: V' lr ().l2 A n g u l i r i v c l o c i t y i r n c lr l t r n o l ' c y l i n c i r i c a cl i r r n : (11"' -. r()rcv/rrirr ,, :.1'.i,t, - 2.()r'tr/sr,, {}. ti) }r -5 Conceptualdesign TU Berlin r49 5.9 Examplcs of conccptualdcsigrr SELECTION CHART lmpulse-loading leslrlg (Sv)evaluated by variants Solution (Sv) lVarksolutjon variants S E T E C T ICORNI T E R I A : yes n0 Lackol inlormation Check specifrcalion Testconnection (+) Pursue solution ( - ) E l i m i n astoel u t i o n ( ? ) C o l l e ci nt f o r m a t i o n (re-evaluate solution) ! ) Check specification I lask withtheovera Compatible Fufilsdemands 0l lhespecilicati0n Linkage Figure5.71.Conceptvariant7, from [5.53] measures salety Inc0rp0rates direcl company Preferred by desiqneas Lol1uJ. og caalroVto.ule- Hvolraul^cs rct- No cxF€rae^ce-Poder Cle.^o,acl 2 I No experie*cs ye.f wLEL bral<es Cylindrical cam probleaaht appLled lvleoJ og u-cgne.| u'utt- Supply 3- hqristor li' rcForS l too grecC cnlrol Testconnection l:isure 5.72. Concept variant I/2 from [5.53] Pcriod of revolution: 2n Lf - - - 3 . 1 4s a) l jigrrrc 5 . 7 ( 1S, c l c e t i o t ot l c o n r b i r u t i o n so l s o l u t i o np r i r t c i p l c (r v u r i t t r t t s ) Sincc the switching time of electromagneticallyoperated clutchesfor connectrttgitncldisconnectingthe cam drive is somewherein the region of a few tenths of rr sccor]cl,thcrc should be no problems in applying this principle. The magnitude o t , r r r r rrl i l t c o l i n c r c a s ci n , t h e i m p u l s et o r q u e l o a d i n gc a n b e a l t e r e db y m e a n so f I n l c r c h i u r g c i r b l cea n r sa n c la l s o b y v a r y i n g t h e p e r i o d o f r e v o l u t i o n . S t c t r sI r l r c s t i n r l r t i n ul ' l v w h c c ll n ( ) l n c n to l ' i n c r t i i r : ( i s i i n t r r t co l ' r l r c c l r c r g yn c c d c r l l i r r t h c i r r r p u l s c( i r r r dI r c n c co f t h c c n c r g y t o b c s t o r c d ) o t l l l r c i l s s u n l l ' t l i ( ) nt h a l t l l l o u d c a r r y i l U l p i l r t s t r r c c l i r s t i c a l l y dclirrrncd. 150 5 C o n c e p t u a ld e s i g n 5 . 9 E x a m p l e s o f c o n c e p t u a ld c s i g n 151 Stored energy of flywheel: | J y a t 2 : 1 5 9x 1 0 z J \: -Rotational speed after the impulse: Eafrer: Er - En'o" : 15 640 J Testconnection l2E*,"" -- : @artcr : 125rad/s; flartcr: 1 190rev/min V Figure5.73.Conceptvariant/1 from [-5.531 The drop in rpm is thus seen to be very low. Hence all that is needed is a ,motor with a small output. Step 8: Evaluating the concept variants Rack andpinon Figure5.74.Conceptvariantlz. from [5.531 ^ mcx 9 Result lirrrther firming up by means of scalelayout drawings and also development and t onstruction of the actual test rig were all based on concept variant V2. Figure :.79 depicts the end result. -,b-3cm mox c=d.x _250,r_ Having thus firmed up on the combinationsof principles, we can now evaluate the concept variants with some confidence. In what follows we shall be considering the four variants selectedduring Step 6. Important wishes in the specificationprovide a seriesof evaluationcriteria of varying complexity. These are assessedand elaborated with the help of the checklist (Figure 5.60). Next, a hierarchical classification(objectives tree) is drawn up to facilitate closer identification and better co-ordination of the weighting factors and the parameters of the variants (Figure 5.76). In Figure 5.77 parameter magnitudesand values assignedto the variants have been set against the evaluation criteria. It appears that variant V2 has the highest overall value and the best overall rating. However, variant V3 follows close behind. For the detection of weak spots, it is advisableto draw a value profile (Figure -5.78).The figure shows that variant V2 is well balanced in respect of all the important evaluation criteria. With a weighted rating of 68 per cent, variant V2 thus representsa good starting concept for the embodiment design phase. Figure 5.75. Development of cylindricalcam Stored energy at maximum impulse torque loading: E.u* : iF.o*'h :260 J This amount of energy is needed in the time interval At : 0.12 s - F l y w h e e ld i m e n s i o n s : S c l c c t : M a x i l n t t m r p m , r l . r . r:r 1 2 ( Xr)e v / m i n ; r r r' l 2 ( rr i t t l / s I . i r r I ' l y w h c c ld i r t t c r t s i o n o s l r ' : 0 . 2 n t i r n t l r r ' - ( ) ,I t t t . t l t e l l t ' r v h c c lr t t i t s s . r r r l- l ( X l k g ,i r r r t lt l r c r r r o r r r c nolt i n c r l i i r .J 1 = l t n s r : - 2 k [ n r r . 5.9.2 One-handedhouseholdwater mixing tap ,\ onc-handed mixing tap is a device for regulating water temperature and tltroughflow with one hand. This task was sent to the design department by the p l r r n n i n gd e p a r t m e n t i n t h e f o r m s h o w n i n F i g u r e 5 . 8 0 . Slcp l: ()larilying the task and elaboratingthe specification N t ' w t l i r t i tt t t t l i t t i n g s , s t a n c l u r c l s ,a f c t y r c g u l i r t i o n sa n c le r g o n o m i cf a c t o r sl e d t o tltc rcplircctttctto t l ' t h c o r i g i n i r l s p c c i l ' i c l t i o nb y t h c r c v i s c d v c r s i < l ns h o w n i n l r i g u r c5 . l t l . , 52 5 Conceptualdesign 153 5 . 9 E x a m p l e s o f c o n c c p t u i r ld c s i g n O @ o @ @ N = il-4 >=< ?' 5 @ @ N N @ ei O d o 6- s >!.? d la E> od = = c E c E E E E o o O N o N d = s-- it@ N @ N @ =:5 o - s _ '.^ *46 o it srl E> = il|13n? 0.084 0.12 021 0.09 E !l .. = 3 E o O ti-^@ S-= Y O @ = O N N a d O =: t\ a - + S S E> ? ilY s!, d \ I ! Reliable and simple testing device = C N e", ;'s i = = O E o @ @ o = o O @ O = = O N N ll S-? >@ Go(t @ N tl -= o O c6 SI ts> ilo sil 6S c E = .E E = = c) J o rvl3-v. I = E 004 0.06 0ulckexchange ol test c0nnectt0ns O J E E L = 6 <- z .2 o a? c I E E9 E = I E c o c 3a >a OY >a >= x:j e_> oo El! oEi 8.9 o ah z cG- 66 E' <E ':-o i,j I^l @ = Good accessibil ol measuring systems rigurc5.76.()bjcctivcstrcc for impulse-loading tcstrig d d o O O d o a d O E !': -o q r9 )W= 1 a c; t' 9 € G o.E EF" 6 ,: ilr a >6 .!? E it €g }F AE c :g -3 *9r |.Ett 5 a -- ziz '64 E E= a;a E c t3 :jE tr.t so >c^ I}J rll hdg !l E (n 3a O* tr! q, LL i Conceptual design 5.9 Examples of conccptuulrlesitn 155 Variant V2,0WV2: 6.82| Variant \. )WVs: 6.45 ii IhN.N I| I a \\\l [r.a ,,','rl l\\.\\\\\\\\\) N\.\ \\\\\N i Y/,/./ ./,/,/././,/ r//j .,/,/ ./ ,/,///f\.\\\ \ .\\\\\\.].) for chansesl il/ = ]I tb :| t_-.s --_1-------.5S , i-, viz 10 9 I 7 6 5 4 3 ? 1 o r 2 3 4 5 6 7 B 9 ro Figure5.78.Valueprofilefor detectionof weakspots I D D D D W vie D D D D W D F i g u r e- 5 . 7 9F. i n i s h e di m p u l s e l o a d i n gt e s t r i g c o r r e s p o n d i n gt o concept variant Iz2,from [-5.-531 -handed watermixino tap D D D D D W D W W D D D characterlstics mtxing tapwlththel0 lowing r red:one-handed household water Throughput l\ilax.pressure 1 0l / m i n 6 bar pressure Normal 2 bal Hotwater temperature 60"c SPECIFICATION THDarmstadt --= v./ I =l= 10mm srze Connector product in twoyearslrm Finished t0 bemarketed ention t0 bepaidl0 appearance. Thefirm'strademarkl0 beprominently displayed. nufacturing DIV30 eachat a production rateol 3000tapsperrnonth coslsnott0 exceed ). ( )nc-ltirrttlcd rrtixi rrg 1it1.r[ : x a n r p l eo f a n i t s s i g t t t t t c tsl t l g g c s t c (bl y t h c I i gurc .5.13( l l r u r i n gr l c p i r r t r t t e r t t D D D jne hdnded nixing lap Page1 Requi renents Resp. 1 Throughput (mixed flow)max10/minat 2 bar pressure 2 Max. pressure 10bar(test perDIN240,1) 15baras 3 Temp. ol waterstandard 60.C,j00.C(short_time) 4 Temperature setting independent ofthroughput andpressure 5 Permissible +5"Cata pressure temp fluctuation diff.of +5 barbetw 3enhotandcold supply 6 Connection: 2 x Cupipes, l0 x I mm,I : 400mm p 351fmm,basin 7 Single-hole attachment thickness 0 jB mm (0bserve basin dimensions DINEN3j, DINEN32,DINj368) B Outflow above upper edge ofbasin: 50mm 9 Tofithousehold basin 10 Convertible intowallfittjng 11 Light (children) operation 12 Noexternal energy l3 Hard water (drinking supply warerl 14 Clear identification 0ftemperature setting .15 Trade prominenily mark displayed l6 Noconnection of thetwosupplies whenvalve shut 17 Noconnection whenwater drawn on 18 Handle nottoheat above 35.C l9 Noburns fromtouching thefinrngs 20 Provide protection scalding if extra costs small 21 0bvious operation, simpie andconvenient handling 22 Smooth, easily cleaned contours, nosharp edges 23 Noiseless operati0n (<20dBasperDtN52218) 24Service life10years at about300000 operations 25 Easy maintenance andsimple repairs. Usestandard parts spare 26 Max.manuf. costs DM30(3000unitspermonth) 2/ Scheduies frominception ofdeveropment c0nceptual embodrment designdetail /pe designprotot, des rgn alter2469 Repiaces 1stissue of 12.6.j923 - l l r r r t l ertrl r i x i r r ;gr p f'rgrn5 C. X 1S . p c c i l i c i r t i loi r ro r r e months l 156 5 Conceptualdcsign Step2: Abstractingto identifythe essentialproblems The basisof abstractionis the specification,from which it is possibleto arrive at that the Figure 5.82. Simple householdsolutionsfor mixing taps suggested chosensolution principle must be basedon meteringout the water throug\a diaphragmor valve. Suchalternativesasheatingand coolingby the introduction tormulation: Problem Flowof hotandcoldwater orsometered either stopped temperature thatthemixed toany canbeadiusted desired value regardless ofthethroughput. Functions @ @ @ FdFrl --------5 Flowof materials Flowof signals r57 5 . 9 E x a r n p l e so f c o n c e p t u a ld e s i g n unchanged-that is, the relationship V,lV,n must remain constantand independ e n t o f t h e s i g n a lp o s i t i o n s i . On changing the temperature 8n. V. must remain unchanged-that is, the sum of V, + Vh: Vn,must remain constant.To that end the componentflows /. and V6 must be changedlinearly and in the opposite senseto the signalsetting s,y. Step 3: Establishing function structures The function structures were derived from the sub-functions; - Stop-meter-mix - Adjust flow rate -Adjust temperature The physical principle being well known, the function structurescould be varied and developed to determine the best system and its behaviour (Figures -5.84to 5.86). From the results, the function structure shownin Figure 5.86 was chosen as the most satisfactorv. I ,,,i.. boundary System Figure5.i32.Problemformulationandoverallfunctionasper specification, Figure-5.81 V : volumeflow rate;p : pressure; dl: temperature lndex:c : cold,h : hot, m : mixed,o : atmosphere of external energy through heat exchangersetc could be dismissed:they were more expensive and involved a time lag. Whenever known solution principles can be applied such '4 priori determinations' are frequent and permissible. Next, the physical relationshipsfor the diaphragm or valve flow rate and the temperature of mixed flows of similar fluids were determined (Figure 5.83). Temperature and flow rate adjustments are based on the same physical principle-valve or diaphragm. On changing the flow rate V^, the flows must be changed linearly and in the same senseas the signal setting s9. The temperature rl.. however, must remain lo rl I I l c 100 20o 300 1,0o ' 500 Flowrate h - p z = E ( a / 2 )G ; V : v A . v = aC ,Vt Ti Y h - P z q" 0sh: 0sc : 2.5 Oar :Y dh Ym 8m : samc fluid l/h. Oh + 0.2 0.4 0.6 0.8 L0 : 2.5bar;p,, : 2 barjlPsrr I r.*trrc5.114.Furrctionstructure for a one-hanclccl water mixinq tap baseclon Figurc -5.132. \ t c l r ' r ' i r r gl ' l o w@ l r r r t rl r t l j u s t i n gl c r r r P c l l r t r r r c s c y r r l l ; 1 1 1l .'c1iy, r . i r n i x i n ! rI.n t l i e g r l r y r h s . @ I t t t t ' so l c ( ) l l s l i t l tttc n t p c r a t u r cl r n d c o r r s t i r r lr)tc l c c l r t i l A cl ' l o wn r t c h u v cb c c n p l o t t c c lf i l r f l t v c t lt c l l l l ) c r i l t u r cs c t t i n g s( , r u ) i r n rlll o r v r i r t cs c t t i r r ; l (s. r i . ) , ' l ' l r r o r r gnhr u t u a lc f f c c l so l l h c l t h c l c r r t l t c t ' l l t r r ci r r r t l k r u c l l r r i r c t c r i s t i cusr c n o t l i r r c i r r I l ( ' s \ t l r c so t l t l t c i r t l e t sl t @ r r r r t@ r\('cl)t li)r tlrc sctting,Ti,- ().fll5. ittrd hctrCC OtCUnrui|ttblc lirr rrrrirlll'Lrrvnrtcs. At rr llrr\sr!t'ctlillcrcrrcc[rctwccrrt]ic coldtnd hot wt||t b i t r )r h c l i r r c s h i l t . ' lh c s c t t i n ptrtic n o l o n a t * {l t{25(dilgrtm on riSht) rcltrrrlr,ri. il ( i n t h i rc i t \ r/ , . r , / ) . , ( ) , 5 of rrch olltcrcvert litt tlte 5 Conceptualdesign 158 IVIETER B 5 . 9 E x a m p l e s o f c o n c e p t u a ld e s i g n 1-59 N4ETER |7 IVETER zl I I i stJ t-. I Se f 10 di o oB 0 P s n: P s c = 2 . 5b a r 7n - +- 06 70 60 50 40 30 70 10 400 0.4 60r 50f 0.2 401_ 30F 0 55 02 04 06 0 ps6: 2.5bar;p,, : 2 53t Figure 5.85. Function structure basedon Figure 5.82 in which the temperature rs set before, and the flow metered after, mixing. With equal pressuresin the supply pipes, the flow and temperature settings are independent of each other due to equal pressule differences aiross each temperature-flow-metering valve. The behaviour is linear. Witil different supply pressures, however, the characteristic ceasesto be linear and is strongly displaced, especially with small quantities, when the pressurein the mixing chamber approximates the smaller supply pressure. If it is exceeded, then cold or (here) hot watcr only will run out regardless of the temperature settrng. Step 4: Searching for solution principles to fulfil the sub-functions 'vary two flt Brainstorming was used as a first attempt to solve the problem: areas,simultaneouslyor successively,in one senseby one movement and in opposite senseby a second, independent, movement'. This resulted frolrl sub-function: 'meter flow rate and temperature' in accordancewith the fu structureshownin Figure5.86.The resultsare shownin Figure5.tt7. Analysisof brainstormingresults 'fhc s0f B0t- s o l u t i o n s s u g g e s t e d . d u r i n gt h e b r a i n s t < l r m i n gs c s s i o nw e r c c h c c k c d c s t i r b l i s h c rwl l r c t h e r t h c I / i r r r c il 9 s e t t i n g sw c r c i n ( l e l ) c t t t l c t t tA. r t i r t t a l y s i tst f l c l i r l l o w i n g $ ( l l u t i ( ) tpl r i l l c i l l l c s : c o n r h i l t c dn l ( ) v c l n c l l l ss t t g g c s l c ct h 20l10.10.2 04 06 0B Psn: Psc: 2.5bar itt- 10 - 0 . 2 0 4 0 6 OB 9v: 2.5bar;Pr,: 2 5ut Figure 5.86. Function structure basedon Fig. 5.82, in which the temperatureand flow at each inlet is metered out independentlyand then mixed. Linear temperatureand flow characteristics.No seriouschangeseven at different supply pressurei. Solutions with separatemovementsfor v and 0 tangential to the valve seat face '[-he independenceof the 7 and r?setting is only guaranteedif each of the flow rurcas of the valves is bounded by two edges running parallel to the correspondingmovements. This implies that the movements must proceed at irn angle to each other and in a straight line. Every valve setting thus has two pairs of straight and parallel bounding edges(Figure 5.88). This ensuresthat when one setting is adjusted the other setting is not simultaneouslyadjusted. l)istribution of bounding edges.Each of the componentsproducing the valve tlow ureas must have at least two edges that face each other and lie in the t l i r c c t i o r ro f t h c m o v c m e n t . willr llrc 7 sctting. both valve areas must approachzero simultaneously. W i t l r t h c r , ls c t t i r r g .o r r c l l r c a l r u s t i r p p r o a c hz c r o i r s t h e o t h e r a p p r o a c h e si t s t t t i x i t t t t t l t t . ' l ' h i si r t r l t l i c sw . i t h 7 s c ( t i n g s (. h l r t t h c l t o u n c l i n ge c l g c so n b c l t h v i { l v c i t r c ; t st l l t t s t t l l ( ) v c t o w i r r c l sc i r c l t o l l t c r o r l r w i t v l r o n t c a c h o t h c r i n t h c \ i l l t l c s c l l s c . W i t h r ?s c t t i r t g s t. l t c l r r t t t t t t l i t tcgt l g c r r l r r t l r t , t $ ' ( )v i l l v c l t r c i t sl ) l g s t n r ( ) v ci n l h c i r p p o s i l cs c l l s c l ( ) c t c h o t h c r , 160 dcsign 5 conccptual pipe cylindrical #* A x i aml o v e m:eDn t J A / ent R o t a r y m o v e: tm 4 A x )a -Beam principre - Inverse otcylpioe -The seat face may be plane, cylindricalor spherical. - Solutions of this type can be effectedwith a singlevalve element,and seem simple to design. t,, #] r rn 'il---E- principre Inverse olbeam vl p lc. .{YA' Flowarea Nir a_|! r----- v, Beam withptugs - Ooposing valves {,,*( operated bysc,ssor orincip'e n fl *.nr= andrackandpinon H frf ilH -Sliding plates + sliding wedges plates (asabove) lnverse ofsliding f F+l to & P ^.r f-Tl' v n 0 + W Lr a Ballsin pipesactivated by conical cam - Rotatrng valveplate withaxialmovement -' --' (sharp edges t0 ensure correct 4i, lll , a+P2P'&t -F mrxrns)a-m _fl| UU(ln rwo wedqes VA pump (notpursued) lnjection Throttle flap - Twothrottle flaps - Three-way mixer Mffi ffi r#ry # r'Tt - Chamfered cylinder -Pivotandswivel - controllever - ball central 00re eccentric bore wTv (tZ -;- | ,------ GE// -..- s /=, fl Movementsand Figure5'88. bounding edgesof valve positions 2. Solutions with separatemovements for V and B normal to the valve seatface - This group includes all movements which involve lifting a valve from its seat face. However, only a movement at right anglesto the seatface is possiblein practice. -The independent setting of V and rl can only be achieved with additional control elements (coupling mechanism). -The design seemsto require greater effort. 3. Solutions with one type of movement for V and I tangential to the seat face -To guarantee the indepenclenceof the V and I settings, additional coupling elements are needed. - The solutions are similar to those listed under 2. They only differ in the shape of the seat face and the resulting movement. 1. Solutions with one movement fnr i normal to, and one movement for I tungential to. the seatface and vice versa These solutions do not, even with the help of coupling mechanisms,satisfy thc demand for indepenclent 7 and r1 settinss. The overall function is not achieved. ._ *-.N\s..tr'::."Yt ffi '16\ v Twotlexiblelubes (squeezewilhovat wto Camorwedoe) f _TW fi w N Movewedge lwoapertures belween Jl t,., lrl - Membrane #.ut-ij=l.- @ -t) _al .4\ /,q \aVqJ r" A' trr V a\1,*' rrV.U tv Twobasicpossibilities: rigidcoupling/via mechanisms - lns SPhincter Vortex t-----"--\ lffi:3) (|\ 'ftJ Twoptates r67 5.9 Examples of conceptual design .\-" \-\$, \$.v* F i g u r e5 . t 3 7R . c s u l to f u b r a i n s t o r m i n gs c s s i o nt o d i s c o v c rs o l u t i o np r i n c i p l c sl i r r t h c I r s s i g n n r c n t ' v i t rt vw o l ' l o wi r r c l s . s i n r r r l t l n c o u s lov r s t t c c c s s i v e l ivt ,t o t t c s c r t s cl r v o t t c tt'tcrtl' n l ( ) v c n rnel l r r t l i r r t l r c o p P o s i t cs c r t s cb v i r s c c r t t t r li.t t d c ; r c n d c t t lt l ( ) v e l)iscursive searches for solutions could be conducted with the. help of t llssification schemesfor an independent setting of the flow rate 7 and the I t ' n r p c r e t t u r re) . l}irsed on the identified solution principles, the classificationcriteria were l i s t c c la s s h o w n i n F i g u r e 5 . 8 9 . 'I'hc combination of the solutions thus discovered into a slngle scheme was torrnclto be inexpedient for the following reasons: S o l r r t i o r r sw i t h r n o v e m e n t sf o r I / a n d d t a n g e n t i a l t o t h e s e a t f a c e c a n b e irtlvirnlagcouslyvaried and classifiedby type of movement and form of valve. ( ( ' l r r s s i l i c a t i osnc h c r n eI s h o w n i n F i g u r e5 . 9 0 . ) W i t l r s o l r r t i o n si n v o l v i n g a s i n g l c I n o v c r r c n t f o r 7 a n d r ) a t t h e s e a t f a c e ( i i r r r g c r r t i i rrll r r r o r r n l r l ) t, h c i r r r a n g c r r r c rot ll l l t c s c i r l l ' i r c c si t t r col f t h c c o u p l i n g r r r c c l r i t r r i s rl r irr si r r l c c i s i v cb c i r r i n go n l l t c s o l u l i o t t . ( ( ' l a s s i f ' i c i r t i osnc h c n r cI I s h o w l t i n | i g u r c 5 . ( ) 1). r62 5 Conceptualdesign icatlon Classif criteria parameters Associaled Direction of movement Tangential toseatface ilhN\ Normatosealface Coupiing of movemenl 7 -lI Twomovements al ananglet0 eachotherlor t/andf ilovement notcoupled 0pp0srte K In oneplane L ln Oned rect0nfor l/ andr? f,Iovemenl lilovement couped Typeof movement r63 5 . 9 E x a m p l e so f c o n c e p t u a d l csign 1r Trans aton nn lrv Cylnder i aranangle Cone M Sphere Special shapes Elastic bodles Arrangement of seatfaces .^- vt.,+-J il ,lv frc '^V,;l TU' p ate,wedge Plane Formof valve 0pposite I n o n ep a n e A1anange Figure 5.89. Classificationcriteria for a one-handedmixing tap a[7,^. 2\r{' L _l T, A[Jrk\]F -tlv Rotation -.ii lvF JL_ 3 (v/ .4.:r' sh J .".'n\^ v!l j\% s K3 "V\ I Figure 5.91. Classification Scheme II for solutions to the one-handed mixing tap problem Movement tangential or normal to the seat face Counled movement in one direition for I/ and r1 Step 6: Firming up into concept variants Step5: Selectingsuitable solution principles Ihe solution principles fulfilling the demands of the specificationand promising ,o be most economic were those found in the classificationscheme shownin Figure 5.90, and were therefore preferred. With the help of further research into possible setting or operating elements which we have not discussedhere, the solution principles could then be firmed up into concept variants (Figures 5.92 to 5.95). Step 7: Evaluating the concept variants il*q' planeplate A cylinder B c0ne C sphere 0 trons./tro ns. trons./rot. rot. /rot. 3 7 i+rl"E ru BJ' t+t d'i" {ili' o &r,, F i g u r c 5 . 9 0 .C l a s s i f i c a t i o S n c h e r n eI f o r s o l u t i o n so f t h c onc-hantlerlrnixing tirp proltlcm M o v c n r er t t t a n g c n t i a tl o t l r c s c i t tf a c c 'l'wo irrtlcpctttlctttntovcnrcnlsitt iul irrrglclirr V nttd rl Irr accordancewith VDI2225, this step was taken with the help of an evaluation ehart. In addition, evaluation uncertainties and weak spots were examined ( F i g u r e5 . 9 6 ) . -fhanks to the balanced profile and the discernable improvement possibilities, Solution B (Figure 5.93) was found to be preferable to all others. 'l'he ball solution D (Figure 5.95) would only have been consideredif further strrrlicsinto production and assemblyproblems had been undertaken and led to p o s i t i v cr e s u l t s . ll Rcsult l ) r l r w i r r g so l ' S o l r . r t i o nI l w i t h i m p r o v e m e n t st o t h e o p e r a t i n gl e v e r i n r e s p e c to f r l l i r ( c r c r l t t i r c n l c n t s ,c i r s i c r c l c a n i n g a n c l r t u r n b e ro f p a r t s w e r e p r o d u c e d .T h e l c v c l o l i n l o n n l l i o r t l i l r S o l u t i o r tI ) w l s i t t t p r o v c t lw i t l t i r v i c w t r l r c - c x l r n r i n i n gi t f o r l ' i t t i r lc v i r l u i l l i o r t , 5 Conceptualdcsign u $ 5 . 9 E x a m p l e s o f c o n c e p t u a ld e s i g n I. THDarmstadt EVALUATION CHART lor One-handed mixin0 tap ln lheorderof varianl A I P:presenl lhechecklisl l-eadings| (P):possib'e aflerrnproverenl critenon P lNoI Evaluation Page7 B (P) P E (P) P (P) P (P) FunclI Reliabi ityofstopping llow without drips 1 1 3 3 Work Princ Reliable, reproducible setting (calcium-resistant parts) fewwearinq 1 2 3 2 I mbo 2 Lowspace requiremenl 1 3l 2 2 + 4 Fewparts 7 I 2 I 1 + 5 S i m p lm e anufacture I 7 3 2 /? y Easy assemb I 2 2 2l 7 Convenient operation, sensitive setting I I 8 Easyupkeep (easy to clean) I 4t 2 I 3 ) (wlthstandard S mpe maintenance t0os, littings neednotbedismanfled) 1 1 3 2 7? J 16 24 (%, 7 (1) P) 1 I Prod ASSY 6 ( F 3 J J .9 Figure5.92. Figure5.93. E O vlainr9 2 J il l4 joint or solder Screw ? Evaluation uncertain better I Tendency: worse I Tendency: 4 2 1 2 3 20 Rr a+51 Ranking 4 (J),Weakspot(W),lmprovement (l) of variant/criter Justificati0n 0n '26 756 2 (3) 2) levermechanism 8 1 Simplity 9t Figure5.94. Figure 5.94. Figure 5.95. 'plate solution with eccentric Figure 5.92. One-handed mixing tap, solution variant A: and pull-and-turn grip' 'cylinder solution with lcvcr' I ' i g u r c 5 . 9 3 .O n e - h a n d e dm i x i n g t a p , s o l u t i o nv a r i a n tB : ' c y l i n c i c sr o l u t i o r rw i t h c n t l I ; i g u r c5 . 9 4 .O n e- h a n d e dm i x i n g t a p . s o l u t i o nv a r i a n t( ' : v i r l v c si r n t li r r l t l i t i o r t i sr lc i r l i n g ' 'ltlll solrrtion' t l t P ,s o l t t l i o t tv : l r i i l t l ll ) l l r i g u r c 5 . 9 5 ,( ) t t c - l t i t n t l c trl r r i x i r r g position Indeterminate y of ballduringassemb lmprove with84 ) 9 1 A l t ,irrrn n not l e v e r t , , ,,. u , , l P t u r i o s0 0 r u l | 0Bnw i l ni m p r 0 v e r e0nl (l 0 n l r 0e e m e r l s i erse s ernet s tus i n 2 m o n t h s | 5 0 l u0l n u L x a m i nper o d u c t ipo0ns s i b i l i t p lj,lr: n ro tJ, I litltlrc 5,t)6. ( )ttc'-h;uttlcttnlixilUttl;l livrlultir)n (rl conccl)t vlliiults A. Ii. (', I) P (P) 6 , I S t e p so f e m b o d i m e n td c s i g n Embodiment design Embodiment design is that part of the design process in which, starting from the concept of a technical product, the design is developed, in accordance with technical and economic criteria and in the light of further information, to the point where subsequentdetail design can lead directly to production (see 3.2). 6.1 Stepsof embodimentdesign Having elaborated the solution concept during the conceptual phase, the designer can now firm up on the underlying ideas. During the embodiment phase, at the latest, he must determine the overall layout design (general arrangement and spatial compatibility), the preliminary form designs(component shapesand materials) and the production procedure, and provide solutions for any auxiliary functions. In all this, technological and economic considerations are of paramount importance. The design is developed with the help of scale drawings, critically reviewed, and subjected to a technical and economic evaluation. In many cases several embodiment designs are needed before a definitivc design appropriate to the desired solution can emerge. In other words, the definitive layout must be developed to the point where a clear check of function, durability, production, assembly, operation and costs can be carried out. Only when this has been done is it possibleto prepare the final production documents. Unlike conceptual design, embodiment design involves a large number of corrective steps in which analysis and synthesisconstantly alternate and complement each other. This explains why the familiar methods underlying the seorch Jor solutions and evaluatior?must be complemented with methods facilitating thc identification of errors (design faults) and optimisation. The collection o.l inlrtrmution on materials, manufacturing processes,repeat parts and standards involvcs a considerableeffort. 'l'ltc c m b o d i m c n t p r o c e s si s c o m p l e x i n t h a t : - n l a n y i r c t i o n sh a v c t o b e p c r f o r m c c ls i n r u l t i r n c o u s l y ; - ' s o n l c s t c p sh a v c t < lb c r c p c i r t c ci lr t t r l r i g h c rl c v c l o l ' i n l i l r r n i r t i o ni;r n c l - - i t t l t l i t i o t t si r t t di r l t c r i r t i o n si n o n c i r r c i rl r i r v cl c p c r c u s s i o n o s n l h c c x i s t i n gr l c s i g r r itt rllhcrlrrcirs. l f c c i r t t s cr l l t l t i s . i t i s r r o t r r l u ' ; r v s; x r r r r b l g t r l d r n w u p i l s l r i c t p l i r r r l i l r t l r t . r61 embodiment design phase. Instead, the designer may have to adopt a general approach. Particular problems may demand deviations and subsidiarysteps,and these can rarely be predicted in detail. It is always advisableto proceed from the qualitative to the quantitative, from the abstract to the concrete, and from rough to detailed designs,and to make provision for checks and, if necessary,for corrections (see Figure 6'1). 1. Using the specification,the first step is to identify those requirements that have a crucial bearing on the embodiment design: -Size-determining requirements such as output. throughput, size of connectors etcl arrangement-determining requirements such as direction of flow, motion, oosition etc: and - material-determining requirements such as resistanceto corrosion, service life, specified materials etc. Requirements based on safety, ergonomics, production and assembly involve special design considerationswhich may affect the size, arrangement (see 6.2) and selection of materials. 2. Npxt, scale drawings of the spatial constraints determining or restricting the e6bodiment design must be produced (for instance drawings showing clearances,axle positions, installation requirements etc). 3. Once the embodiment-determining requirements and spatial constraints have been established, a rough layout, derived from the concept, is used to identify the embodiment-determiningmain function carriers-that is the assemblies and components fulfilling the main functions. The following subsidiary problems must be settled, due regard being paid to the principles of embodiment design (see 6.4): Which main functions and function carriers determine the size, arrangement and component shapesof the overall layout? (For instance,the blade profiles in turbo-machines or the flow area of valves.) What main functions must be fulfilled by which function carriers jointly or separately?(For instance, transmitting torque and allowing for radial movement by means of a flexible shaft or by means of a stiff shaft plus a special coupling.) 4. Preliminary layouts and form designs for the embodiment-determining main function carriers must be developed; that is, the general arrangement, component shapesand materials must be determined provisionally. To that end, it is advisable to work systematicallythrough the first three headings of the checklist (Figure 6.2). The result must meet the overall spatial constraints and thcn bc completed so that all the relevant main functions are fulfilled (for instancc by spccifying the minimum diameters of drive shafts, provisional gear r a t i o s , m i r t i r r r u r nw a l l t h i c k n c s sc t c ) . K n o w n s o l u t i o n so r e x i s t i n gc o m p o n e n t s ( r c p c a t p a r t s . s t i r n c l a r dp a r t s c t c ) n r u s l l r c s h o w n i n s i m p l i f i e df o r m . I t m a y b e usclul to stirrt working orr sclcctct[lrcits rlttly, conrhiningthese later into - p r c l i t r r i r t i t rlvl t y o t t l s . 5 . ( ) t t c ( t 1 t l l ( l r cs r t i l l r l r l cP t c l i t t t i l t ; r r r ' l ; t v r l tltl ll rt l $ ll ) c s c l c c t c t il t t l t c c r l r t l i t t t c c 6 Embodiment dcsign 168 II t 6.1 Steps of embodiment design Headings Examples Function ls thestipulated function lulfilled? Whatauxiliary functions areneeded? principle Working principles produce Dothechosen working thedesired effects andadvantages? Whatdisturbing factors maybeexpected? lnlormation I I of spatial constraints Produce scaledrawings provide: Layout andformdesign Dothechosen overall layout, comp0nent shapes, materials anddimensions (strength) adequate durability permissible (stiffness) deformation adequate stability freedom fromresonance unimpeded expansion acceptable corrosi0n andwearwiththe stipulated service lifeandloads? Definition carriers mainluncti0n embodiment-determining ldentifv E; preliminary layouts andt0rmdesigns Develop carriers maintunction fortheembodiment-determininq L - - Safety Haveall thefactors alfecling lhe safety of thec0mponents, ol thelunction, ol theoperation andol theenvironment been takenintoaccount? Erfnomics Have themanmachine relationships been takenintoaccount? Have unnecessary human stress or injur;ous lactors beenavoided? Hasattention beenpaidto goodphysical layout? Production Hastherebeen a technological and procedure? economic analysis ol theproduction control Quality Canthenecessary checks beapplied during andafterproduction 0r at anyother required time, andhave theybeen specif ed? Assembly Canall theinternal andexternal processes assembly beperformed simply and in thecorrect order? Transport Havethernternal andexternal transp0rt conditions andrisksbeen examined andtaken intoaccount? Onprrlinn Have all thefactors influencing the operation, suchasnoise, vibration, handling, etcbeenconsrdered? Maintenance Canmaintenance, inspection andoverhaul beeasilyperformed andchecked? Costs Have thestipulated costlimitsbeen observed? Willadditional operational or subsidrary costs arise? Schedu les Canthedelivery dates bemet? Arethere design modifications that mightimprove thedelverysituation? preliminary layouts suitable Select Creatlon preliminary andformdesigns layouts Develop carriers mainfunction fortheremaining functions to auxiliary forsolutions Search c o E torthematnfunction layouts andformdesigns Develop detailed carrlers luncti0n withtheauxiliary compatibility ensuring carriers E E q E E lunction lortheauxillary layouts andlormdesigns detailed Devel0p layouts the0verall and complete carriers E layouts theoverall andrefine Check - criteria andecon0mic technical aqainst Evaluate Evaluation cnecK II I Decision + Creation c factors anddisturbing lorerr0rs Check E c 6 E I l r i g r r r c( r L S t c p so l c r r t b o r l i t t t r , t' rl ct tr t g n + Eval uation cnecK II I + Decision + r69 l " i g u r c6 2 . ( ' h c c k l i s tf o r e m b t t d i m c n tc l c s i g r r 170 6 Embodiment desisn and with the with the proceduresdescribed in 5.6 (modified if necessary) checklist(see6.2). for 6. PreliminarY laYouts and form designs must now be developed considered the remaining main function carriers thit have not yet been until becauseknown solutionsexist or they are not embodiment-determining this stage. as support' 7. Next, determine what essential auxiliary functions (such known exploit retention, sealingand cooling) are neededand, where possible' If this solutions)' solutions (such as repeat piitr, stundardparts, catalogue already procedures the proves impossible,s"ar.h ior specialsolutions,using described(ChaPter5). carriersmust now 8. Detailed liyouts and form designsfor the main function and guidelines rules design be developed 'andin accordancewith thi embodiment detailed regulations, standards, to 6.5), with due attention i*" o: compatibility of problem the to also calculationsand experimentalfindings,and If necessary,divide with those auxiliary functions that have now been solved. individually' or areasthat can be elaborated into assemblies for the auxiliary 9. Proceedto developthe detailedlayoutsand form designs refine the necessary' If parts. fungtion carriers,addin! standardand bought-out overall into carriers function all designof the main funcii,oncarriersandcombine layouts. compatibility, 10. check the overall layoutsfor mistakesin function' spatial (see 6'2) and the the-.checklist to p.rfor-un"e, durability eic by referencestep. by step refine Then iault_eliminationmethod outlined in 6.6. (see6'7)' 11. Evaluatethe layoutsagainsttechnicaland economiccriteria concrete more put in be to If a particular project requires several.concepts be course, of not, must process embodiment form piior to evaiuaiion,then the final the that so demands' variants the of prr.rrr"d beyond what the evaluation economicallyas iayout designcan be determinedand elaboratedas quickly and is thus possible,in somecases,to take a on carriershavereachedthe preliminary :cisionwill haveto be deferreduntil after '::'#ii ffi:'*:*::il:-:l;"H,Tffi 6 2 Checklist for embodiment desisn NI 15. Conclude the embodiment design phase by preparing a preliminary parts list and preliminary production documents. 16. Fix the definitive layout design and pass on to the detail design phase. In the embodiment phase, unlike the conceptual phase, it is not necessaryto lay down special methods for every individual step. The representation of the layout and form designs may be based on standard drawing conventions or, if necessary,on simplified scale drawings, as suggested by Liipertz 16.122). The searchfor solutions for auxiliary functions and other subsidiary problems is based either on the procedure described in Chapter 5, but simplified as far as possible, or else directly on catalogues.Requirements, functions and solutions with appropriate classifying criteria have already been elaborated. The embodiment (layout and form designs) of the function carriers is based on the checklist and involves reference to the principles of mechanics and to structures and materials technology. It calls for calculations ranging from the simplest to complex differential equations or the method of finite elements applied with the help of computers. For these calculations, the reader is referred to tfe literature listed in 6.5.1, and for even more complex calculationsto the sp/cial literature. In addition, fixed rules and principles, to be elaborated later, must be followed. Because of the fundamental importance of the identification of errors (design faults) in several of the steps, the reader is especiallyreferred to 6.6. In the elaboration of embodiment designs,many details have to be clarified, confirmed or optimised. The more closely they are examined, the more obvious it becomes whether the right solution concept has been chosen. It may appear that this or that requirement cannot be met, or that certain characteristicsof the chosen concept are unsuitable. If this is discovered during the embodiment phase, it is advisable to re-examine the procedure adopted in the conceptual phase, for no embodiment design, however perfect, can hope to correct a poor solution concept. This is equally true of the solution principles applicable to the various sub-functions. However, even the most promising solution concept can cause difficulties in detail .design. This often happens because various features were originally treated as subordinate or as not in need of further clarification. Attempts to solve these sub-problems compel the designer to reiterate the appropriate steps while retaining the chosen solution concept. evaluationis Possible. 72. Fix the PreliminarYlaYout' 13.optimiseand"o-pt"t"theformdesignsfortheselectedlayoutby in the courseof the 6.2 Checklist for embodiment design elimination of the weak points that have been identified repeatthe previousstepsand adopt evaluation.If it shouldprove advantageous, suitablesub-solutionsfrom lessfavouredvariants' Embodiment design is characterisedby repeateddeliberationand verification 1 4 . C h e c k t h i s l a y o u t d e s i g n f o r e r r o r s ( d e s i g n f a u l t s ) i n f u n c t i o n , s p a t i a(ls e e6 . l ) . factors' Make what compatibility etc (sel 6.\ anJ for the effectsof disturbing Evcry embodimentdesign is an attempt to fulfil a given function with feasibilitymust bc improvementsmay be needed.The technicaland economic appropriutelayout, componentshapcsand materials.Thc proccssstartswith at this point at the latest' established 172 6 Embodiment design preliminary scale layout drawings based on spatial requirements and a rough analysis, and proceeds to consider safety, ergonomics, production, assembly, operation, maintenance and costs. In dealing with these factors, the designer will discover a large number of interrelationships, so that his approach must be progressive as well as reiterative (verification and correction). Notwithstanding this double character, however, his approach must always be such as to allow the speedy identification of those problems that must be solved first. Though individual factors may be closely interrelated, the designer can derive important checklist headingsfrom the general objectives and constraints(2.1.6) which, moreover, provide him with a useful procedural order and a systematic check of each step in turn. The checklist thus not only provides a strong mental impetus, but also ensures that nothing essential is forgotten in the embodiment phase (see Figure 6.2). All in all, reference to the headings will help the designer to develop and test his progress in a systematic and time-saving way. Each heading should be examined in turn, regardless of its interrelationship with the rest. The actual sequence is no indication of the relative importance of the various headings,but ensuresa systematicapproach. For instance,it would be futile to deal with assembly problems before ascertaining if the required performance or minimum durability is ensured. The checklistthus provides a consistentscrutiny of embodiment design and one that is easily memorised. 6.3 Basic rules of embodimentdesign The following basic rules apply to all embodiment designs. If they are ignored, breakdowns or accidentsmay occur. They underlie nearly all the steps listed in 6.1. When used in conjunction with the checklist (Figure 6.2) and with the design fault identification method (see 6.6) they also help with selection and evaluation. The basic rules of clarity, simplicity and safety are derived from the general objectives, that is: - fulfilment of the technical function; - economic feasibility; and - individual and environmental safety. The literature contains numerous rules of, and guidelines for, embodimenl d e s i g n 1 6 . 1 2 7 , 6 . 1 2 8 , 6 . 1 4 5 , 6 . 1 5 2 , 6 . 1 1 0 1 .O n c l o s e r a n a l y s i si t a p p e a r s t h a t clarity, simplicity and safety are fundamental to all of them. Clarity, that is clarity of function or the lack of ambiguity of a dcsign. facilitatesreliable prediction of the performanceof the end product and in mitny casessavestime and costly analyses. Simplic:ity generalty guarantees economic l'casibility. A smallcr nutrrbcr tll' c ( ) r n p o n c n t si t n c ls i n t p l c s h i t p c si t r c p r o c l t t c c dn t < l r cq u i c k l y a n d c a s i l y . 6 3 Basic rules ol embodiment desisn n3 Safety imposes a consistent approach to the problems of strength, reliability, accident prevention and the protection of the environment. In short, by observing the three basic rules, the designer can increase his chancesof successbecausethey focus his attention on, and help him to combine, functional efficiency, economy and safety. Without this combination no satisfactory solution is likely to emerge. 6.3.1 Clarity In what follows we shall be applying the basic rule of clarity to the various headings of the checklist (Figure 6.2): Function Within a given function structure, an unambiguous interrelationship between the various sub-functions and the appropriate inputs and outputs must be guaranteed. W/rking principle The chosen working principle must, in respect of the physical effects: - reveal a clear relationship between cause and effect, thus ensuring an appropriate and economical layout; - guarantee an orderly flow of energy, material and signals. If it does not, undesirable and unpredictable effects such as excessiveforces, deformations and wear may ensue.For this reason alone, it is advisableto avoid the so-called'double restraints', the more so as they can causefurther difficulties during production and assembly. By paying attention to the deformations associatedwith a given loading, and also to thermal expansion, the designer can make the necessaryallowancesfor possible expansion in a given direction. The widely used bearing pairs, with a locating and a non-locating bearing Figure 6.3a), avoid'double restraints'and have a clearly defined behaviour. stepped bearing pair (Figure 6.3b), on the other hand, should be specified ly when the expected changes in length are negligible or when the resulting is permissible. By contrast, a spring-loaded arrangement, in which which : operating axial force Fu must not exceed the pre-load Fo, will permit a clear finition of the force transmissionpath (see Figure 6.3c). Combined bearing arrangements often present problems. The combination in Figure 6.4a consistsof a needle roller bearing which is intended to nsmit the radial forces and a ball bearins which is meant to transmit the axial forces. Hclwever,this particular arrangementdoes not clearly define the pathfor the radialforces,because the innerandouterracesof both transmission bearingsarc rcstraincdradially.As a rcsultthc scrvicelife cannotbe predicted shownin F'igurc6.4/r.rrnthc othcr hand,satisfies Thc arrangcmcnt accuratcly. providcdthc dcsigncrcnsurcs,during rulc with thc sanrcclcnrcnts. thc <'/arity t74 Figure6.3 6 Embodiment design Figure6.4 r75 6 3 Basic rules of embodiment design If these data are not available, the implementation must be based on reasonable assumptionsand the expected service life specified accordingly. In any case, the embodiment must be such that the loads can be defined and calculated under all operating conditions. No impairment of the function or the durability of a component must be allowed to arise. Similarly, behaviour in respect of stability, resonance,wear and corrosion must be clearly established. 'for safetl"s sake'. Thus a Very often one comes across double restraints shaft-hub connection designed as a shrink fit will not have a better load-carrying capacity if it is also provided with a key as in Figure 6.5. The extra element merely ensurescorrect positioning in the circumferential sense,but becauseof the reduction in the area at A,the resulting stressconcentration at B and the presence of complicated and almost incalculable stressesaI C, it decreasesthe strength in a drastic and fairly unpredictablemanner. Schmid [6.185] has shown that an axially pre-loaded taper joint for the transmissionof torque requires a spiralling motion when the hub is assembledon the shaft in order to ensure a religble shrink fit, and the use of a key would prevent this. Figure 6.6 shows a housing adapter for a centrifugal pump which can be used to provide various annulus profiles to fit different blade shapes so that new housings need not be constructed for each case. Unless the intermediate pressure in the gap between the adapter and the housing can be clearly Figure6.4.Combinedrolling-element bearing: (a) Transmission pathof radialforcesnot clear;(b) combinedrollingbearingwith the sameelementsasin (a), but clearidentification of the transmission pathsof theradialancl axialforces assembly,that the right-hand race has enough radial play in the housing, thus making certain that the ball bearing transmits axial forces only. Layout and form design T h e l a y o u t ( g e n e r a l a r r a n g e m e n t ) a n d f o r m d e s i g n ( s h a p e sa n d r n a t c r i a l s ) r c q u i r e a c l e a r d e f i n i t i o r ro l ' t h c m a g n i t u d c . t y p c , f r c q u e n c y a n d c l u r a t i o n o l klads. l.l7 rA pressure balancing t0 ensure Passage, llal P;= Po Figure 6.6 F i g u r c 6 5 . ( ' o n r b i n c csl h i r l t - h u t r c o l r l t c c t i o r t b y n t c a n s o f s h r i n k f i t a n d kAenye x a m p l e o l n o t r r p p l y i n gt h c l l l i r r c i p l co l c l i r r i t y . \ l r i g u r c( r , ( r ,I l o u s i t t gl t l u p l c r i t t i t c o l t l - w i t l c ll ) t l t t t l ) 176 6 Embodiment design regulated, or some other means of attachment is used, the adapter might travel upwards and damage the blades by rubbing against them' This is particularly true when similar fits (H7-j6) are chosen for the two locating diameters which are approximately the same size. This is because, depending on manufacturing tolerances and working temperature, gaps may unknown upp"u., tile relative size of which is unpredictable and which produce The housing. and the adapter the intermediate pressuresin the space between designed specially of the means by solution shown in Figure 6.6 (detail) ensures, connecting passageA (which must have a flow area roughly four to five times greater thin the-maximum gap area that might appear at the upper locating Iiameter), a clearly definable intermediate pressure corresponding to the lower inlet preisure of the pump. As a result the housing adapter is always pressed downwards when the pump is in operation, and attachments are only needed as locating aids for assemblyand to prevent any tendency of the adapter to rotate' SeriJus damagehas been reported in gate valveswhose operational or loading conditions were not clearly defined [6.80,6.81]. When closed, gate valves separate, say, two steam pipes and at the same time close off the inside of the vaive housing. The result is a small, self-containedpressurechamber as in Figure 6.7. If condensatehas collected in the lower part of the valve housing, and steam appearson the inlet side with the valve closed so that the valve is heated, 6 3 B a s i c r u l e so f e m b o d i m e n t d e s i g n n7 then the enclosed condensate may evaporate and produce an unpredictable increase in pressure inside the valve housing. The result is either a ruptured housing or serious damage to the housing cover connection. If the latter is self-sealing,serious accidentsmay ensuesince, in contrastto what happenswith overloaded bolted flange connections,there is no preliminary leakageand hence no warnlng. The danger lies in the failure to specify clear operational and loading conditions. Possibleremedies are as follows: - Connect the inner chamber of the gate valve housing to an appropriate steam pipe, operational conditions permitting (purru": ppip.). -Protect the valve housing againstexcesspressure (puu1u. restricted). - Drain the valve housing, thus avoiding collection of condensate (puulu"= - Pe*t..nol). Design valves in such a way as to minimise the housing volume (collection of condensatekept low). Similar phenomena in welded membrane seals are discussedin [6.154]. Safetv S e 6 b a s i cr u l e i n 6 . 3 . 3 . Ergonomics In the man-machine relationship, the correct operation must be ensured by the logical layout of the equipment and of the controls. Production and Quality Control These must be facilitated by: - clear and comprehensive data in the form of drawings,parts lists and instructions; - the designer's insistence on adherence to the prescribedproductionand organisationalprocedures. bly and Transport uch the same is true of assemblyand transport. A planned assemblysequence nting mistakes should be incorporated in the design (see 6.5.7). O p eration and Mainte nsnce ar installation intructions and the appropriate embodiment design must ensure that: the performance is easily checked; and maintenance involves the smallestpossible variety of tools and equipment. 6.3.2 Simplicity l j i g u l c ( r , 7 ,( i i r t c v l l v c w i t l t r c l i t t i v c l yl i r r g c krwcr collc('linl nlcil 'sirnlllc' rncilns 'not For tccluticll applicationsthc wrlrcl complcx', 'casily u t t r l c r s t r x l tilu' t d ' c i r s i l yd t l r t c ' . n8 (r Embodimcnt design A solution seemssimpler if it can be effected with fewer components, because the probability of lower production costs, less wear and lower maintenance is then greater. However, this is only true if the layout and shapes of the components are kept simple. Hence the designer should always aim at the minimum number of components with the simplest shapes[6.118, 6.144,6.154]. As a rule, however, a compromise has to be made: fulfilment of the function always demands a certain number of components. Cost-efficiency often imposes a decision between numerous components with simple shapesbut with greater overall costs, and a single and cheaper cast component with the greater uncertainty it may entail in delivery. Returning to the checklist: Function In principle, only a minimum number and a clear and consistentcombination of sub-functionswill be pursued during considerationof the function structure. Working Principle In selectingworking principles, only those involving a small number of processes and components; having obvious validity; and involving low costsare taken into consideration. 'simpler' in individual casesdependson the problem and the What constitutes constralnts. In the development of the one-handedmixing tap (see 5.9.2) several solution principles were proposed. One group (Figure 5.90) involved two independent adjustments in directions tangential to the valve seat face (types of motion: translation and rotation). The other group (Figure 5.91), though involving only movements in one direction (normal or tangential to the seat face) required an additional coupling mechanism to convert the two single adjustments into one direction of movement. Quite apart from the fact that, in the secondgroup, the pre-set temperature is often lost when the tap is shut off, all solutions represented in Figure 5.91 involve a greater design effort than does the first group. Hence the designer should always begin with a group such as that depicted in Figure 5.90. Layout and form design Here the simplicity rule requires: - geometrical shapes which can be analysed simply for strength and stiffness; - symmetrical shapes which provide clearer identification of deformations during production and under mechanicalor thermal loads. In many cases,the designer can reduce the work of calculation and experimentation significantly if he tries, by means of a simple design, to facilitate thc a p p l i c a t i o no f b a s i cm a t h e m a t i c a lp r i n c i p l e s . Safety S c c u r r r l c r( r . 3 . 3 . 6 3 Basic rules ol cmboclitncnl dcsrgn t79 Ergonomics The man-machine relationship should also be simple and can be significantly improved by means of: - sensibleoperating procedures; - c l e a r p h y s i c a ll a y o u t ; a n d - easily comprehensiblesignals. Production and Quality Control Production and quality control can be simplified, that is speeded up and improved, if: - geometrical shapespermit the use of well-established,time-savingmethods; - the manufacturing methods involve short setting up and waiting times; and -shapes are chosen to facilitate the inspection process. Leyer, discussingchangesin manufacturingmethods [6. 119],usesthe example of a sliding control valve approximately 100mm long to demonstrate how the replacementof a complicated castingby a brazed product made of geometrically simple turned parts helped to overcome difficulties and paved the way for more ecafromicalproduction. Pursuing his line of approach, we discover that further simplifications are possible (Figure 6.8). Step 3 helps to simplify the geometrical shape of the central, tubular part. Step 4 (fewer parts) can be taken when the surfaceareasat r i g h t a n g l e st o t h e v a l v e a x i s n e e d n o t b e r e t a i n e d . r e 6 . 8 . S i m p l i f i c a t i o no f a s l i d i n gc o n r r o lv a l v e ( a p p r o x 1 0 0m m l o n g ) [6.l l9]. rleted by steps3 and 4 Castin^gdifficult and expensive;2 Improvement by splitting into simple, brazed, parts; simplification of central tubular part; 4 Further simpiificati*on possibility A f u r t h c r c x i r r r r p l ei s p n l v i c l c cbl y t h e o n e - h a n d e dm i x i n g t a p d i s c u s s e de a r l i e r c l c s i g no l ' t h c l c v c r i r r r a n g c n r c nst h o w n i n F i g u r c 6 . 9 i s e x p e n s i v et o m a k e 'l'hc c l i l ' l i c u lt to c l c i r n( s l i t s ,o l l c n r c c c s s c s ) . o n c s l l r w n i n F i g u r e6 . l 0 i s m u c h r l c r i t l t t l i t l s r tr l l t t r cs t t i t i r l l l cl i r r l r l r t g c rp r r l t l r r c t i r l lrtt r r t s . ' l ' h cl c v c r . w h o s cc n c l i r rl ' ( ) o v cr,' c t p r i r eisr s r r r i r l l crrr r r r r r b cort ' C t t t ts l i t l c i l t l ( l r ( t l i t l ci r t : t c i t t ' t t r t t l c r c r l l i E 6 Embodiment design 180 Figure6.9 Figure6.10 Figure 6.9. Proposedlever arrangementfor a one-handedmixing tap with translational and rotational movements Figure 6.10. Simpler solution with improved embodiment (basedon Schulte) parts and avoids wear in areas that are difficult to readjust. All in all, therefore, this solution is by far the better. 6 3 Basic rules ol ernbodiment design 181 F i g u r e6 . 1 1 .A d j u s t a b l e sealing r i n go f a n i n d u s t r i aslt e a mt u r b i n e a; d j u s t m e n o t sn A i n the samesenseproduceverticalmovement,adjustments at A in the oppositesense producea rotationaboutB that approximates to a horizontalmouement Operation and maintenance With respect to operation and maintenance, the simplicity rule means: - the operation must be possible without special or complicated instructions; -the sequenceof operations must be clear and simple, and any deviations or faults easily identified; and -maintenance must not be clumsy, laborious and time-consuming. Assembly and transport Assembly is simplified, that is facilitated, speeded up, and rendered more reliable, if: - the components to be assembledcan be identified easily; -the assemblyinstructions can be followed easily and quickly; -no adjustment has to be repeated; and - reassemblyof previously assembledcomponents is avoided (see 6.5.7). During assembiy,the sealing ring of a small steam turbine has to be adjusted vertically and horizontally with the turbine shaft already assembled,to ensure uniform clearance around the labyrinth seal. Doing this without having to remove the shaft several times for adjustment poses a problem which may bc solved by the design shown in Figure 6.11: the adjustment can be made at thc joint, roiation of the adjustment screws(A) in the same senseproducing vertical movement only, and rotation in opposite sensesproducing a tilting movemen( about pivot (B) that approximates to horizontal movement. The pivot itsell must, however, allow for vertical movement during the adjustment and also for radial heat expansionwhen the turbine is operating. l'his is achicvcclwith a fcw of simplc shal'rc. A suititlllc ilrrilngclncnt tlf lltc easily produ."d "1"-"nts s u r f a c e s ,m o r e o v e r , g l ' r v i a t c st h c n c c t l t ( ) s c c u r c l h c p i v o t p i n w i t h i t c l c l i t i o n i t l l o c k i n g c l c r r r c n l s :i t i s l o c i r t c r li n s r r c l ti r w r t y l l t t t l i t c t l n t t o tI ' i t l lr l t l t . 6.3.3 Safety I Type and scope of safety techniques fety considerationsaffect both the reliable fulfilment of technical functions also the protection of man and the environment. The designerhas recourse safetytechniquesthat, following16.43],may be classified as: direct; indirect;or warning. In general, the designer should try to guarantee safety by the direct method, t is by choosinga solutionthat precludesdangerfrom the outset.only when proves impossible should he have recourse to indirect safety methods, in er words,constructspecialprotectivesystems(see6.3.3.3)[6.44].Warning y systems, which merely point out dangers and indicate danger areas, ld be avoided by the designer and should only be used as a last resort and vcr ils a shortcut, I n t h c s o l u t i o n o f r r t c c h n i c a lp r < l b l c r r t ,t h c c n g i n c c r i s l ' a c c d w i t h s e v e r a l c o n s t r i l i n t sn. o t i r l l o l ' w h i c h c i r r rI t c l t o p c l ( ) ( ) v c r c o l l l cs i r n u l t i r n c < l u s l y . I l c r t r u s t r r c v c r t h c l c s ss t r i v c t o p r o v i d c I t o l u l i o n l l t i r l c o r n c s l l c i l r c s t l o t82 6 Embodiment design satisfying all the requirements. A single constraint may, under certain circumstances,put the realisation of the whole project in doubt. Thus a high demand for safety can greatly complicate a design and, by reducing the clarity, may even lower the inherent safety of the product. Moreover, safety provisions may also render a product uneconomic and lead to its abandonment. Such cases,however, are exceptional becausesafety and economy generally go hand in hand. This is particularly true of expensive and complex plant and machines. Only smooth, accident-freeand safe operation can ensure long-term Protection againstaccidentsor damage,moreover' goeshand economic Success. in hand with reliability [6.48, 6.226] which makes it possible to operate a machine to full capacity, even though lack of reliability does not necessarilylead to accidentsor damage. A11in all it is therefore advisableto achieve safety by treating direct methods as an integral part of the system. There are four special categoriesof safety technology, namely: l. Component safety, which concerns the protection of components against fracture, inadmissible deformation, instability etc. The durability of a component under given loads over a fixed time must be consideredin connection with the materials and manufacturing method used. 2. Functional safety, which concerns the safe operation and reliability of plant or machines designed for specific tasks by the appropriate combination of assemblies and components. It involves the avoidance of dangerous and economically undesirable operating conditions. 3. Operator safety concerns the safety of human beings, that is the protection of their physical and mental health while operating the plant or machine. 4. Environmental safety concerns the safety of human beings not directly involved in operating the plant or machine and the protection of the environment against harmful effects. Becauseof the growth of technology and populations, this category has become one of acute importance. For the designer, all four categoriesare intimately connected.Thus, component safety influences functional safety and operator safety. Functional safety may, under certain circumstances,affect component safety, and, in the case of damage, all can endangerman and his environment in many ways. Therefore the designer should pay equal heed to all four categories[6'158]. 2 Direct safety PrinciPles Direct safety methods aim at achieving safety through systems or components directly involved in the perforrlance of a particular task' To ensure the safe functioning and durability of components, designerscan a d o p t o n e o f s e v e r a ls a f e t y p r i n c i p l e s[ 6 . 1 - 5 8 1B. a s i c a l l y ,t h e r e a r e t h r e e s u c h principles, namely: - t h e s a f e - l i f ep r i n c i p l c ; - t h c f i t i l - s a l cp r i n c i p l c :i r r l c l - - t h c r c t l r t r t r l i t r t lcl vr i t t c i p l c . 6 3 Basic rules of ernbodimelrtclesign 183 The safe-life principle demands that all components and their connections be constructed in such a way as to allow them to operate without breakdown or malfunction throughout their anticipated life. This is ensured by: - the clear specification of the operating conditions and envilonmental factors, such as the anticipated loads, service life etc; \ -adequately safe embodiment based on proven principles and calculations; - numerous and thorough inspectionsduring production and assembly; - the analysisof components or systemsto determine their durability when they are overloaded (load levels and/or running time) or subjected to adverse environmental influences; and -the determination of the limits of safe operation, due regard being paid to possible breakdowns. It is characteristicof this principle that it basessafety exclusivelyon accurate qualitative and quantitative knowledge of all the influences at work or on the determination of the limits of failure-free operation. The application of this principle calls for a great deal of experience, or for costly and time-consumingpreliminary investigations,and for continuousmonitoring of the state of components. If a failure should neverthelessoccur, and if a safe life is essential.then as a rule there will be a seriousaccident. for instance the fracture of an aeroplane wing or the collapse of a bridge. The fail-safe principle allows for the failure of a system function or for a component fracture during the service life by ensuring that no grave consequences ensue. To that end: -a function or capacity, however small, must be preserved to prevent dangerousconditions; - a restricted function must be fulfilled by the failing component or by some other component until such time as the plant or machine can be put out of operation without danger; -the failure or breakdown must be identifiable; and - the effect of the failing component on the overall safety of the system must be assessable. In essence,the impairment of a main function must be signalled. The signal can take various forms-increasing vibrations, loss of sealing, loss of power, slowing down-each without causing immediate danger. In addition, special monitoring systems may be provided to indicate the incipient failure to the operator. Their layout should be governed by the general principles of protective systems. The fail-safe principle presupposesknowledge of the consequencesof failure and provides a means for taking over the impaired function. By way of example, let us consider a spherical rubber element in an elastic c o u p l i n g ( F i g u r c 6 . l 2 ) . T h e f i r s t v i s i b l cc r a c k a p p e a r so n t h e o u t e r l a y e r b u t t h e f u r t c t i r l ni s n o t y c t i r n p a i r e c l( S t i r t c l ) . O n l y w h c n t h c n u m b e r o f r e v o l u t i o n s u t r d c r l o i r c li s i n c r c i r s c r cl k l c st h c s t i l l r t c s sl r c F , i rtto r l c c r c i r s cw i t h a c i l n s c c l u c n l l l t l r c c o r r p l i r r gw. l r i e hn l r n i l ' c s l si t s c l l ' ,l o r i n s t i r n c cb. y i r c l u r n g ci n t h c b c h i r v i o r r o (r Ernbodinreut design 184 _ 1st crack -l ^' @40mm 185 without loss of essential information. Redundancy is often used delitrerately to allow for transmissionlosses,and hence to safeguardthe system [6.69]. Redundant safety arrangements lead to an increase in safety, provided that the breakdown of a particular element of the system is not dangerous in itself and that other elements, arranged in parallel or in series, can take over its function fully or at least in part. to : 30l/s - 6 3 Basic rulesof ernbodiment design rqo l State @ 1stcrack layer at edgeof rubber r 50 0 0 N ^ I lTlfi'l - 1 00 0 0 Figure6.13.Fastening of components: coveringof the boltedconnectionmaintains functionandpreventsbrokenpartsmigratingin the eventof bolt failure C 6 s000 Noofrevsunder Load against andstiffness Figure6.12.Fail-safe behaviourof an elasticcoupling:crack-state n u m b e ro f r e v o l u t i o n s lowering of the critical speed (State 2). With further operation, the crack grows larger and causesthe stiffnessto decreasestill further (State 3), but even if the crack went right through, there would not be complete failure of the coupling. No sudden effect with serious consequencesneed therefore be feared. Another example is the behaviour of flange bolts made of tough material which, on overloading, relax. Their impaired function is indicated by the resulting loss in flange sealing but does not give rise to sudden failure. Figure 6.13 illustratestwo safe methods of fastening components.The means of attachment should be so designed that, even if the bolts begin to fail, the mountings remain in place, no broken parts can migrate and the equipment continues to function to some extent [6.154]. T h e r e d u n d a n c yp r i n c i p l e p r o v i d e s a n o t h c r n l c i r n s o l ' i n c r c a s i r t gb o t h t h c s a f e t y a n d t h e r e l i a b i l i t yo f s y s t c n r s . I n c o m m o n u s a g c , r c c l u n c l a n c yn l c i l n s s t r p c r l l u i l y( ) r c x c c s s .l t t i t t l i r r n t i t t i o t t t h c o r y , r c c l u n r l i r n cryc l er s t o t h i r l l ' r i r c l i o rot l u l n c r t u g c w h i c l t t t t i r yb c c l i l t t i n i t t c t l The provision of several engines in aeroplanes, of multistrand cable for a high-voltage transmission line, and of parallel supply lines or generators, all ensure that, should a particular element break down, the function is not completefy impaired. In that case, we speak of active redundancy, becauseall the components are actively involved. Partial breakdownslead to a corresponding reduction in performance. If reserve elements (for instance alternative boiler feed pumps)-usually of the same type and size-are provided and put into operation during breakdowns, then we speak of passive redundancy. If a multiple arrangement is to be equal in function but different in working principle, then we have principle redundancy. Depending on the situation, safety-enhancingelements can be arranged in parallef (parallel redundancy), for instance emergency oil pumps, or in series (seriesredundancy), for instance filter installations. In many cases,layouts in parallel or serieswill not suffice and crossoverlinks will have to be introduced to guaranteetransmissiondespitethe breakdown of severalelements(Figure 6.14). In a number of monitoring systems, signals are collected in parallel and c<rmparcd with one another. Selectiveredundancy (two out of three) and utmpurative rcdundancy arrangementsare shown in Figure 6.14. l i , c c l t t n c l i t n cl iyt y o u t sc a n n o t , h o w c v c r , r c p l a c c t h c s a f e - l i f eo r f a i l - s a f ep r i n c i p l c s , ' l ' w oc i t b l cc i t r so l t c r i r l i n gi n l l i r r i r l l cwl i l l . l d r r r i t t c c l l yi n , c r c a s ct h c r c l i a b i l i t y l t c o r t t r i l ) u l cn r l l h i r t gt o l h c l ) i l s s c n g c r s ' s i r l c t y . ' l ' h c t t l p i r s s c l t g c t ' t r i t t t s p.r lbr u 6 Embodiment dcsign 186 6 3 B a s i c r u l e s o f e m b o d i n . r e n td e s i s n r87 Noredundancy Paral lel-redundancy Series-redundancy Plant redundancY Selective (2 outof 3) 0uartet-redundancY -f-l U-I 0uartet-crossredundancy redundancy Principle in parallel: prtnciple working of A differslromB l ollil signals Switch or dilterent signat Yjusione indicates critical i condltion Plant redundancY Comparative t --ii.^ Fr Figure6.14.Redundantarrangements redundant layout of aeroplane engines will not increase safety if any of the engines has a tendency to explode and hence to endanger the system. In short, an increasein safety can only be guaranteedif the redundant element satisfies the safe-life or the fail-safe principles. Adherence to all the principles we have mentioned-that is the attainment of safety in general-is greatly facilitated by the principle of the division of tasks (see6.4.2) and by the two basic rules of clarity and simplicity, as we shall now try to show with the help of an example. The principle of the division of tasks and the clarity rule have been applied with great consistencyto the construction of a helicopter rotor head (Figure 6.15), and help the designer to come up with a particularly safe construction based on the safe-life principle. All four rotor blades exert a radial force on the rotor head due to the centrifugal inertia forces, and a bending moment due to the aerodynamic loading. The rotor blades must also be able to swivel so that their angles of incidence can be changed.A high safety level is achieved by thc following measures: - A completely symmetrical layout so that the external bending moments and the radial forces at the rotor head cancel out. - The radial forces are transmitted exclusively by the tr:rsionally flexiblc member Z to the main central component. - T h e b e n d i n g m o m e n t i s o n l y t r a n s m i t t c ctl h r o u g h p a r t / l a n c li s t i t k c n b y t h c r o l l e r b e a r i n g si n t h c r o t o r h c a c l , As a rcsult. cvcry corrrp()ncnlciln bc optimally dcsigncdin itccortlittrccwitlt ils gure 6 15. Rotor-btade attachment of a helicopter basedon the principte of the division tasks (Messerschmitt-Bolkowsystem) k. Complicatedjoints and shapes are avoided and a high safety level is tained. Indirect safety principles irect safety methods involve the use of special protective systems and ive equipment. They are applied whenever direct safety methods prove adequate. In what follows, we shall be consideringprotective systemsonlya discussionof protective equipment, that is equipment to shield danger (for instance,machineguards),the readeris referredto $.aa]. ve Systems tective systems serve either to render endangered plant or machinery ically safe, which generally means putting it out of action or restricting flow of energy or material; or to prevent any plant or machinery in a ngerous state from being put into operation. Very often, direct and indirect safety methods are inseparable.Thus, control regulation systemsare not protective systemsas such, but often make :llent first monitors(primaryprotection)in a safetychain.In that case,the trol and regulation systemsmust embody certain propertiesof protective ms. The properties of protective systems should, in principle, be used in all ms, provided that they can be incorporated without unacceptableextra cost t2'7\. In the layout of protective systcms,thc ftlllowing requirementsmust be taken to consiclcrrtti<ltt: 'aning rrc it pr()lcclivcsystcrnpttxlttccschllnfcf in the wtlrking conditionsol' a E 188 6 Embodiment design plant, a warning must be given so that the operating or supervisingstaff can, if not eliminate the source of the danger, at least take the necessarycountermeasures. Surprise effects should be avoided as far as possible. If a protective system stops or prevents the operation of the plant, it ought to indicate the reason. Self-Monitoring A protective system must be self-monitoring, that is, it must not only be triggered when the system breaks down, but also by faults of its own. This requirement is best satisfied by the closed circuit principle because, with it, the energy needed for activating the safety device is stored and any disturbance of, or fault in, the protective system will releasethat energy and switch off the plant or machinery. The closed circuit principle can be used not only in electronic protective systemsbut also in systemsusing other types of,energy. In a hydraulic protective system based on the closed circuit principle (Figure 6.16), a pump .1 with a pressure-regulatingvalve 2 ensures a constant prepressure pp. The protective system with the pressure p, is connected to the pre-pressure system by means of an orifice 3. Under normal conditions, all outlets are closed, so that the quick-action stop valve 4 is held open by the pressurep, to admit the energy supply of the machine. In case of a faulty axial shaft position, the piston valve 5 at the end of the shaft opens, the pressurep, drops, and further energy supplies are cut off by the quick-action stop valve 4. The same effect is produced by damageto the pre-pressureor protective system, F i g u r e6 . 1 6 .H y d r a u l i c p r o t e c t i o n systemto prevent incorrect axial shaft positionsbasedon the closed circuit principle Figure 6.i7. Safetyfence contact layout: (a) not self-rnonitoringif s p r i n gf a i l s l ( b ) e v c n i n t l ' r cc a s co 1 ' s p r i n gl ' : r i l r r rtch c c o r t t i r c its b r o k c nb y o w n w c i g l r t l( c ) c i r c t r i t c l o s c t bl v l r r rr r rl i t 6 3 Basic rulcs of ernbodiment dcsign I fi() for example by pipe fracture, lack of oil or pump failure. The system is self-monitoring. Figure 6.17 shows layouts of safety fence contacts,for example for a machine guard. closed contacts signal safety fence in position. In layout a, failure of the spring would close the contacts and give an incorrect signal. Layouts b and c solve this problem in two different ways. Multiple, multi-principle and independentprotective systems If human life may be endangeredor if large-scaledamagehas to be averted, then multiple (at least double) protective systemsbased on different principles and independent of each other (primary and secondaryprotection) must be used. Because a single protective system may break down, its mere doubling or replication ensures greater safety: it is unlikely that all the systemswill fail at once. This is, however, only true provided that the replicatedprotective systems do not all fail due to a common fault. Safety is considerably increasedif the double or multiple systems work independently of one another and are, moreover, based on different working principles. In that case?common faults, for instance due to corrosion, will not have catastrophic consequences:the simultaneousbreakdown of all such systemsis highly improbable. This requirement is met in the control of steam turbines for example (Figure 6.18). In the case of overspeeding,the energy supply is cut off by two systems differing in principle. Increases in speed first bring in the regulating system whose speed measurement and regulating valve are independent of, and different in principle from, the quick-action shut-off system. (The simultaneous hydraulic supply on the closed circuit principle is permissiblebecauseit involves a common self-monitoring effect in that pump failure closesboth valves.) The triggering values are staggered so that the regulating system is brought into operation first and only if it should fail is the quick-action system activated. Figure 6.19 depicts two methods of guarding againstexcesspressurebuild-up in a pressure vessel. Mere doubling of the protective system would not help against common faults, for instance against corrosion, use of wrong materials etc. The use of two different working principles makes simultaneous breakdowns more unlikely. It ought to be remembered that, in protective systems, the simultaneous occurrence of different types of redundancy is possible or even necessary(for example in parallel and principle redundancy: Figures 6.18 and 6.19). 4 Designing for safety Here, too, the checklist (Figure 6.2) can prove a great help. Safety criteria must be scrutinisedin respect of all the headings listed. Func:tiotrand wnrking principle l)it(l 0 I t i s i r r t p o r t a n tt o c s t a b l i s hw h e t h e r o r n o t t h e f u n c t i o n i s f u l f i l l e d s a f e l y a n d rcliirlrly lry thc choscttsoltttion. Likcly fatlts irntldisntrbin14 I'actorsmust be taken i n i r l r r c c o u t t itt s w c l l . l t i s t t o t i r l w i r y sc l c r r r . l t o w c v c r , t ( ) w h i r t c x t c n t i r l l < l w a n c c ( r E r n b o c l i m e n td c s i g n 190 6 3 Basic rules of ernbodimcnt design 191 case of new techniques and their application. It has been argued that technical risks must be no greater than the risks man must expect from natural causes should, t6.921.However, this must be a matter for discretion. The final decision in any case, reflect a responsible attitude towards mankind. F i g u r e6 . 1 8 D o u b l e - P r i n c i P l e closed-circuitprotectivesystem , againstoverspending I e Doubling 0 Figure 6.19 Protectivesystetnagainst excessivepressurebuild up in pressure working vessels:(a) two safetyvalves(not safe different p n n c r p l e s againstcommon faults); (b) safetyvalve and shearplate, double 0 principle protective system must be made for exceptional, purely hypothetical, circumstancesthat could affect the function. The correct estimation of the scope and likelihood of a risk should be based on the successivenegation of each of the functions to be fulfilled and on the analysis of the likely consequences(see 6.6). Sabotageneed not necessarilybe considered in this context, although general safety methods are likely to decreaseits effects. What we have to consider first and foremost are failures due to possible disturbancesof the structure, operation and environment of a machine, and what preventive stepsshould be taken. Harmful effectsnot due to technological factors (such as operator ignorance) cannot be eliminated by technical systems but must be considered and if possible limited. A further question is whether the direct safety technique we have been discussingis adequate, or whether safety should be increased by additional protective systems.Finally, we might also ask whether, should it be impossiblc t o m a k e a c l c q u a t es a f e t yp r o v i s i o n si n a p a r t i c u l a rc a s c ,t h e w h o l e p r o j e c t s h o u l t l bc al'rancklnccl. 'f'ltc:tnswcr l u r sb < ' < ' rutt t u i n e r l ,t l n t h c c l c p c n c l so n t h c d e g r c ct t l " ; u . l t ' t . tlln(l ' t l t u ()ll t lhc rttrrgrrilrrdctt.l tlrt pntbultilitl,tll'tttrltrevarttultlatlttrtrttliattr u<"<'iilt'tt, 1 r o . r l ; i l t<l a' ( n t . \ ( ( l u ( , t t ' t(, )r l.l j c c t i v c s t i u l ( l n f ( l rt r c ( l l : l c nl n c k i n g .p r r r t i c t t l i r r l ivr t t l t c Layout and form design External loads produce stressesin components. By analysiswe determine their magnitude and frequency (steadyand/or alternating loads). The various types of stressproduced can be determined by calculation or experiment. Materials technology provides the designer with limiting values of stressfor particular conditions (tension, compression, bending, shear and torsion), beyond which the material will fail. Since these valuesare usually obtained from tests on specimensand not from tests on the component itself, the stressesto which the latter is subjected should be kept well within the limits if adequate durability is to be guaranteed. The ratio of the limiting stresso1 of the material to the acceptableworking stress oyu'in the component is called the safety factor, SF : oyloy. The value of the safety factor dependson uncertaintiesin the determinationof the limiting stress,on uncertaintiesin the load assumptions,on the calculation method, on the manufacturing processesused, on the (uncertain) influence of shape, size and environment, and also on the probability and importance of possible failures. The determination of safety factors still lacks generally valid criteria. An investigation by the authors has shown that published recommended safety factors cannot be classifiedby type of product, branch of engineering or such other criteria as toughnessof material, size of component, probability of failure etc. Tradition, figures based on one-off and often inadequately explained failures, hunches and experiencesare often the basis of numerical data from which no generally valid statementscan be derived. What figures are given in the literature must therefore be treated with circumspection.Their application usually calls for knowledge of the individual circumstances and of the special practices or regulations of the branch of engineering in question. Toughness, that is the ability to undergo plastic deformation before failure and thus relieve stress concentrationscaused by unevenly distributed loads, is one of the most important safety features any material can have. The usual overspeedspinning tests of rotors with the correspondinglyhigh stressesthey set up, and also the required overpressuretests of pressurevessels-provided that they are built of tough materials-are good examples of the direct safety technique aimed at reducing stressconcentrationsin finished components. Becausetoughnessis a crucial safety-enhancingproperty of materials, it is not c n o u g h s i r n p l yt o a i m a t g r e a t e ry i e l d s t r e n g t h .S i n c e ,i n g e n e r a l ,t h e t o u g h n e s s g l ' r r r l l c r i i r l s c l c c r c a s e sw i t h i n c r e a s i n gy i e l d s t r e n g t h ,i t i s e s s e n t i a lt o e n s u r e , l t h c r w i s ct h c b c t t c l i t s< l t ' p l t r s t iccl c f o r m t t t i o nt l r c n o l t l n g c r n r i n i r r r r r rtror u g h n c s s < S u i r r i u l i c c (. l t92 6 Ernbodiment desrgn Dangerous too are those casesin which the material turns brittle with time or for other reasons (for instance, due to radiation, corrosion, heat, or surface coatings). This is particularly true of synthetic materials. If the safety of a component is calculated merely by the difference between the computed stressand the maximum permissiblestress,a vital point is missed. Of the utmost importance is the loading condition and the effect on the properties of the material due to aging, heat, radiation, weathering, operating conditions and manufacturing processes,for instance welding and heat treatment. Residual stresses must not be underestimated either: brittle (fast) fractures without plastic deformation can occur suddenly and without warning. The avoidance of a build-up of additive stresses,of brittle materials, and of manufacturing processesthat encourage brittle fractures, is therefore an essential requirement of the direct safety technique. If plastic deformation is monitored at a critical point, or can be used to impede the function in such a way that the danger can be noticed before man or machines are endangered,it becomes a fail-safe device [6.154]. Elastic deformationr must not be allowed to disturb the smooth functioning of a machine, for instance through loss of clearance. If this happens the force transmissionpaths or the expansionscan no longer be determined with certainty and overloading or fracture may ensue. This is true of stationaryno less than of m o v i n g p a r t s ( s e e6 . 4 . 1 ) . By stability we refer not only to the basic stability of a machine but also to its stable operation. Disturbancesshould be counteractedby stabilisingeffects,that is by automatic return to the initial or normal position. The designermust ensure that neutral equilibrium or potentially unstable statesdo not lead to a build-up of disturbancesthat might get out of control (see 6.4.4). Resonancesproduce increasedstressesthat cannot be accuratelydetermined. They must be avoided unless the vibrations can be sufficiently damped. This applies not only to the stability problem, but also to such associatedphenomena as noise and vibration which impair the operator's efficiency and health. Thermal expansionmust be taken into account under all operating conditions if overloading and impairment of the function are to be avoided (see 6.5.2). Inefficient seals are a common causeof breakdown or trouble. Careful choice of seals, provision for pressure relief at critical sealing points and careful attention to fluid dynamics help to overcome these problems. Wear and the resulting particles can also impede operational safety, and must therefore be kept within tolerable limits. In particular, the designer should ensure that such particles do not damage or interfere with other components. They should be removed as near as possibleto their point of origin. Uniform coruosion reduces the designed thickness of components and local c o r r o s i o n ,p a r t i c u l a r l yo f c o m p o n e n t ss u b j e c tt o c l y n a m i cl o a d i n g ,m a y a p p r e c i a b l y i n c r c a s c t h c s t r c s s c o n c c n t r i r t i < l ri r n t l l c i r d t < l l ' i r s t l l ' a c t u r e sw i t h l i t t l c 'l'ltcrc clclilrttrirti<lrt. i s n o s u c h t l t i r t gi r sl ) c r t n i l l l c l l is l i r l ) i l i t yu n c l c rc < l r r o s i o n - t h c l < l r r t lc i r l ' l i t c i t yo l c o r r r p o n c n t st l c c r c l s c s w i l l t l i n t c , l f o t l r r u r i l ' < l r r ri r r r r l k l c a l c r l r r o s i t l n( c i r r r s c se. l l e c t s t r r r t l l c r r r c t l i c r )t r r c d c n l t w i t h i l r 6 . 5 , 4 . l i i n i r l l y . 6 3 Basic rules ol embodiment design 193 corrosion products can impede the functioning of machines, for instance by jamming valve spindles, control mechanismsetc. Ergonomics The application of ergonomic principles to industrial safety involves the careful scrutiny of safety at work and of the man-machine relationship. A great many b o o k s a n d p a p e r s h a v e b e e n d e v o t e dt o t h i s s u b j e c t[ 6 . 2 6 , 6 . 4 6 , 6 . 7 3 4 , 6 . 1 6 3 , 6.195]. In addition [6.43] specifies the basic requirements of general safety design, and [6.44] deals with protective equipment. Regulations by various professionalbodies, factory inspectoratesetc must be scrupulouslyobservedin all branchesof engineering, and so must a great deal of speciallegislation [6.68]. In a book of this kind it is impossible to examine every aspect of industrial safety, but operator ignorance and fatigue are two factors that should alwaysbe taken into consideration. For that reason alone, machinesmust be designedon ergonomic principles (see 6.5.1). Tables 6.1 and 6.2 list the minimum requirements for safe industrial desisn. with varioustypesof energy. Table6.1. Harmfuleffectsassociated Mechanical Acoustic Hydraulic Pneumatic Electrical Optical Thermal Chemical Radioactive Relative movement of man and machine, mechanicalvibrations. dust Noise Jets of liquid Jets of gas, pressure waves Passageof current through body, electrostatic discharges Dazzle, ultra-violet radiation, arcs Hot/cold parts, radiation, inflammation Acids, alkalis, poisons, gases,vapours Nuclear radiation, X-rays Table 6.2. Minimum industrial safety requirements in mechanicaldevices Protectiveequipmentregardlessof the operational speedis required: - for gear, belt, chain and rope drives - for all rotating parts longer than 50 mm, even if they are completely smooth - for all couplings - in case of danger from flying parts - for potential traps (slides coming up against stops; components pushing, or rotating against, each other); descendingcomponents (weights, counter-weights) - for slots. for example, at material inputs. The gap between parts must not exceed [i mm; in the case of rollers, the geometrical relationship must be examined and, if ncccssary,special guards must be installed. Iilectricul inttullution nrust always be planned in collaboration with electrical experts. In thc cirsc ttl uctttt,;ti< , cltctttiutl itntl nttliout li lc rlitttgct, cxllcrt itclviccntust bc sought for t h c r c t l u i s i t cl ) l o t c c t i o n . 194 6 Embodiment design Production and quality control Components must be designed in such a way that their qualities are maintained during production. To that end special quality controls must be instituted, if necessaryby specialregulations. The designermust help to avoid the emergence of dangerous weak spots in the course of manufacturing processes(see 6.3.1, 6.3.2 and6.5.6). Assembly and transport The loads to which the product will be subjected during assemblyand transport must be taken into consideration at the embodiment design stage. Welds curiied out during assembly must be tested and, where necessary,heat treated. All major assemblyprocessesshould, whenever possible,tre concludedby functional checks. Firm basesand support points should alwaysbe provided and marked clearly. The weights of parts heavier than 100kg should be marked where they can be seen easily. If frequent dismantling is called for, the appropriate lifting points must be incorporated. Suitable handling points must be provided for transport and marked clearly. Operation operation and handling must be safe [6.43,6.44]. The failure of any automatic device must be indicated at once so that the requisite actions can be taken. Maintenance Maintenance and repair work must only be undertaken when the machine is shut down. Particular care is needed with assemblyor adjusting tools. Safety switches must ensure that the machinery is not started unintentionally. Centrally placed, easily accessibleand simple service and adjustment points should be provided. During inspection or repair, safe accessshould be possiblethrough the provision of handrails, steps, non-slip surfacesetc. Costs and schedules Cost and schedulerequirements must not affect safety. Cost limits and delivery dates are ensured by careful planning, and by implementing the correcr concepts and methods, not by cutting corners. The consequencesof accidentsand failuies are generally much greater and graver than the effort needed to prevent them. 6.4 Principles of embodimentdesign The generalprinciplesof embodimentdesignhave been discussedat some lengthin the literature.Kesselring[6.9tt]hasset out the principlesof minimum manufacturing costs,minimumspacerequiremenltr minimumweight,minimum losses,and optimumhandling(t,2.1). Lcycr 16.l20ldiccusses thc principleof { r 4 P r i n c i p l e so I e m b o d i m e n td e s i g n 195 lightweight construction. It is obviously neither possible nor desirable that all these principles should be implemented in every technical solution-one of them might be crucial, the rest merely desirable. Which principle must predominate in a given case can only be deduced from the task and the company's general objectives. By proceeding systematically, elaborating a specification, abstracting to identify the crux of the problem, and also by following the checklist given in 4.2.2, the designer transforms the principles into a concrete proposal that enables him to determine the manufacturing costs, spacerequirements, weights, etc, and to compare these with the requirements in the specification. The systematic approach also highlights the question of how, with a given problem and a fixed solution principle, a function can be best fulfilled and by what type of function carrier. Embodiment design principles facilitate this part of the design process.In particular, they help with Steps3 and 4, but also with Steps 7 to 9 as listed in 6.1 above. For the relatively common task of transmitting forces or moments, it seems 'principles of force transmission'. advisable to establishspecial Tasks requiring changes in the type or variations in the magnitude of a force are primarily fulfilled by the appropriate physical effects, but the designer must also apply the 'principle of minimum losses' [6.98] for energy-conservationor economic reasons,which he does by adopting a small number of highly efficient steps. This principle also applies to the efficient conversion of one type of energy into another, whenever this should be required. In that casethe designproblem, in terms of generally valid functions, reduces essentially to one of channelling, connecting and storing. Energy storage problems involve the accumulation of potential and kinetic energy, be it directly or indirectly through the collection of material. The storage of energy, however, raisesthe question of the stability of the system, and the consequent application of the 'principles of stability and planned instability'. Often, several functions have to be fulfilled by one or several function carriers. Here the 'principle of the division of tasks' may be useful to the designer. Its application involves a careful analysis of the functions and their assignment to function carriers. This analysis of functions is also helpful for the application of the'principle of self-help', when supplementaryeffects must be identified and exploited. In applying embodiment design principles, the designer may find that they run counter to certain requirements. Thus the principle of uniform strength may conflict with the demand for minimum costs. Again, the principle of self-help may conflict with fail-safebehaviour (6.3.3), the principle of equal wall thickness chosen for the purpose of simplifying the manufacturing process [6.118] may conflict with the demand for lightweight construction, and so on. These principles represent many strategies that are only applicable under certain conditions. In using them, the designer must strike a balance between compcting demands.To that end, the present authors have developedwhat they c o n s i d c r t o b e i m p o r t a n t e m b o d i m e n t d c s i g n p r i n c i p l e s ,w h i c h w i l l n o w b e t96 6 Embodiment design presented. Most are based on energy-flow considerationsand, by analogy, they apply equally well to the flow of material and of signals. 6.4.1 Principlesof force transmission I Flowlines of force and the principle of uniform strength The problems solved in mechanical engineering generally involve forces and/or motions and their connection, change,variation or channelling,and involve the conversion of energy, material and signals. The generally applicable function 'channel forces' includes the application of loads to, the transfer of forces between, and the transmissionof forces through componentsand devices [6.18, 6.27, 6.118). In general, the designerwill try to avoid all sudden changesof direction in the flowlines of force, that is force transmission path, caused by sharp deflections 'flowlines of force' aids the and abrupt changes of cross section. The idea of visualisationof the force-transmissionpath (load path) through componentsand devices, and is analogousto flowlines in fluid mechanics. Leyer [6.118,6.I20) has dealt with the transmissionof forces at some length, so that we can dispensewith a detailed discussionof the problem. The designer is advised to consult these important texts. Leyer, moreover, emphasisesthe complex interaction between the functional, technological and production aspects. 'Force transmission' must be understood in a broad sense, that is, it must include the application, transfer and transmission of bending and twisting moments. The external loads applied to a component produce axial and transverse forces plus bending and twisting movements at every section. These set up stresses, direct and shear, and produce longitudinal, lateral (Poisson) and shear strains (elastic or plastic deformations). 'mental The section dimensions transmitting the forces are obtained by dissection' of the components at the point under consideration. The sum of the stresses over these sections produces internal forces and moments which must be in equilibrium with the external loads. The stresses, determined from the section dimensions, are then compared with the material properties of tensile strength, yield strength, fatigue strength, creep strength etc, due regard being paid to streJJconcentrations,surface finish and size effects. The principle of uniform strength 16.7,6.2051aims, with the help of appropriate materials and shapes, to achieve uniform strength throughout the device over its anticipated operational life. Like the principle of lightweight construction [6.120], it should be applied whenever economic circumstancesallow. This important consideration often misleadsthe designer into neglecting thc deformations (strains) associatedwith the strcsscs.It is, however, thcse very deformations that often throw light tln thc bchaviour of comprlncntsand tcll us what wc ncccl to know about thcir functionul cfflciency. 6 . l P r i n c i p l e so f e m b o d i r n e n t d e s i g n r97 2 Principle of direct and short force transmission path In agreement with Leyer [6.11g] we consider the following principle of great importance: If a force or moment is to be transmitted from one place to another with the minimum possible deformation, then the shortest and most direct torcetransmission path is the best. This principle ensures: -minimum use of materials (volume, weight); and - minimum deformation. This is.particularly.true if it is possible to solve a problem using tensire or compressivestressesalone, becausethese stresses, unlike bending and torsional stresses,produce smalrer deformations. when a component is in compression, however, speciarattention must be paid to the danger of buckling. If, on the other hand, we require a flexibre component capableof consid.erabre elastic deformation, then a design using bending o, torrionir ,rr"rirris generally the more economical. The principte is illustrated in Figure 6.2o-the mounting of a machine frame on a concrete foundation-where different requirements Jemand supports with different stiffnesses-This, in turn, has repercussionson the operationar be- 6.20.,Supportinga machine frame on a concrerefbundation: ,I\ry1:: t]cid supporr due to short force transmissron path and low stresson the l1l^i:ly Daseplates: (b) longer force transmissionpath, but still a rigid support with tubes or box sections undercompression: supportwirh pronounced bendingdeformati ron(a lI:]..1:..s n v o t v e:l,tU r g r c a t e ru s eo f m a t e r i a l s ) ; supnort underbendins srresscs: 11] :1:l*lj,.l,iblc s tiffer i constructionwould rhcroacr intorsion. rhiscan be [,:l,l;il,ll;:;l]:;:ry::1,:::.'.* l,:i:l;l,ilr,i.li''r'',,,'*n'its uscrl lilr irltcringthc ics.rrrancc char.ircic,risliris 198 6 Ernbodiment design haviour of the machine: different natural and resonant frequencies, modified response to additional loads etc. The more rigid solutions are obtained with minimum material and space requirements by means of a short support under compression; the most flexible solution by means of a spring, which transmits the force in torsion. If we look at other design solutions, we shall find many examples of the same principle: for example, in the torsion bar springs of motor cars or in flexible pipes that rely on bending or torsional deformations. The choice of means thus depends primarily on the nature of the task-that is on whether the force transmission path must be designed for stability with maximum stiffness, or whether certain force-deformation relationships must be satisfied first and stability can be treated as a subsidiary problem. If the yietd point is exceeded, then the following facts have to be taken into consideration (see Figure 6.21): 1. When a component is loaded by a force, it is invariably subjected to deformation. If the yield point is exceeded,then the linear-elasticrelationship 199 6 4 P r i n c i p l e s o f e r n b o d i m en t d e s i g n 3 Principle of matched deformations Designs matched to the flowlines of force avoid sharp deflections of the transmission path and sudden changes in cross section, thus preventing the uneven distribution of stresseswith high stressconcentrations.A visualisationof the flowlines of force, though very graphic, does not always reveal the decisive factors involved. Here, too, the key is the deformation of the affected components. The principle of matched deformations states that related components must be designed in such a way that, under load, they will deform in the same sense and, if possible, by the same amount. As an example let us take solderedor glued connectionsin which the solder or adhesive layer has a different modulus of elasticity from that of the material to be joined. Figure 6.22a lllustrates the resulting deformation [6.129]. The deformations and the thickness of the solder or adhesive lavers have been Tmax: Tmean c Figure 6.21 Force deformation diagram of tough materials. Arrows indicate causeeffect relationship c o C F c between the force and the deformation no longer holds. Relatively small Figure6.22 Figure6.23 re 6.22. Overlapping adhesiveor solder joint with strongly exaggerateddeformation [6.12e1; I F u r t 6 c r r c ( l u i r c m c n t sa r c t h c u s c r l l t o u g l t t t t a t c r i i t l si t n c l t h c a v o i c l a n c co l ' i t 5 g i l l - u p 9 l ' i l u l t i - a x i a l s t r c s s c si l t l l t c s n t t t c$ c n l i c .U x l l t l t p l c si t r c h i g h l y - c l i s t o r t c d s l t r i t t k t ' i t s .p r c l o i r t l c t l r o l t s i t l t d c l l r t t t ; l r , ) ) Parts 1 and 2 deformed in the samesense Parts I and 2 deformed in the oppositesense Figurc 6.23. Distribution of forces and shear stressesin overlappingjoints with laycr ot a d h c s i v co r s o k l c r ,f r o r n [ 6 . 1 2 5 ] ; ( t t ) o v c r l a p p c t lo r t o r t c s i c l e( b e n d i n gs t r e s sI r c g l e c t c c l ) ( h ) s l l l i c c t lw i t l t l i n c a rl y c l c c r c a s i n tgh i c k n c s s (c) prortouncctl'tlcl'lcctionol thc flowlincs ol lirrcc' with tlclirrnrittionsirt thc oppositc tcnsc (bcntlinF strcssncglcctctl) 6 E , m b o d i m e n td e s i g n strongly exaggerated.The load F, which is transmitted acrossthe junction of Parts 1 and 2, produces distinct deformations in the overlapping parts, the adhesive layer being subjected to particularly marked deformation near the edges due to differences in the relative deformation of Parts I and2. While Part 1 bears the full load F at the upper edge of the adhesivelayer and is therefore stretched, Part 2 does not yet bear a load. The relative shift in the adhesivelayer sets up a local shear stressthat exceedsthe mean calculatedvalue. A particularly unsatisfactory result is shown in Figure 6.22b where, as a result of opposite and unmatched deformations of Parts 1 and 2, the deformation in the adhesive layer is considerably increased.This example makes it clear why provision should be made for deformations to take place in the same senseand, if possible, to be equal in magnitude. Magyar 16.125)has made a mathematical study of the relationships between load and shear stress:the result is shown qualitatively in Figure 6.23. The same phenomenon also occurs between nuts and bolts in bolted joints 16.2441.The nut (Figure 6.24a) is in compressionand the bolt is in tension, that is they are deformed in the opposite sense.In the modified nut (Figure 6.24b) a deformation in the same senseis set up in the leading threads,which gives rise to a smaller relative deformation and hence a more even distribution of the load borne by individual threads. Wiegand 16.2441hasbeen able to demonstrate this effect by showing that such nuts have a longer service life. Paland [6.160] has shown more recently that standard nuts are not as unsatisfactoryas Maduschka 6 4 P r i n c i p l c so l e n r b o d i r n e n tc l e s i g n [6.I23] has suggested,becausethe moment F'h producesadclitionirloutwatd deformations of the nut and thus relieves the leading threads of their loirtl. 'l'h! load-relievingdeformation of the nut due to this moment and also to thc bcrr4ing of the threads can be considerably increased by using material with l lrwcr modulus of elasticity, for instance titanium [6.102]. If, on the other hancl, rlrc load-relieving deformations are resistedby a very stiff nut or a very small lcvcl arm h, then the type of load distribution describedby Maduschka would ensuc. As a further example, let us take a shaft-hub connection formed by a shrink fit. In essence,this, too, involves the deformation of two components (scc [6 '82)) . In transmitting the torque, the shaft experiencesa torsional deformation that decreasesas the torque is transferred to the hub. The hub, for its part, is deformed in accordancewith the transmitted toroue. Figure 6.25 shows that the maximum relative deformation occurs at A. In the case of alternating torques, this may lead to fretting corrosion; moreover the right-hand end, to all intents and purposes,contributesnothing to the transferof the torque. Lifiit,fisA3i"drn* PALAM i after Figure6.25 403020100 9o Share of load borneby individual threads ryl Figure6.26 Figure 6 25 Shaft-hub connectionwith strong 'force flowline deflection' l'6rsional deformationsof shaft and hub in oppositesen-se (yr : angle of twist) Figure 6.26. Shaft-hub connectionwith graduat 'force flowline def'lection' Torsional deformationsof shaft and hub in the same sense Figure6.24. Nut shapesand load distribution irfter | 6.2441 ( a ) S t a n c l a r cnlu t : l i n r i t i r i gc a s ca f t c r M i r r l u s c l r k1i r6 . 1 2 . 3I 1) i;r l i r n 1d 6 .l 6 0 l a l l o w i n g lirr tlclirrrttirtirlrttlrrc to rrrorrrent/, . /r ( h ) M r x l i l ' i c dt t u l w i t l t r t t i r t c l t c t l t l c k r r r t t n l i o n irn l l t c t c r r s i o rltt r r r t The solution shown in Figure 6.26 is much better because the resulting formations are in the same sense. The best solution appears when the nal stitlness of the hub is matched to that of the shaft. The transfer of rque then takes place along the whole length of the connection and hish stress ncentrations are avoided. Even if the shrink fit were replaced with a keyed connection, the layout pictccl in Figure 6.25 would, becausethe torsional deformations are in the ) s l t cs c n s c ,s c t u p v e r y h i g h c o n t a c ts t r e s s e isn t h e n e i g h b o u r h o o do f A . T h e . r tr l c p i c t c ciln F i g u r c 6 . 2 6 ,o n t h e o t h c r h i r n c.l w i l l , b c c a u s ct h e d c f o r m a t i o n s a r c i l r t h c s i u n cs c n s c ,c n s u r c i u ' lc v c n s t r c s st l i s t r i l l u t i o n [(r.1331. 6 Embodirnent dcsign The principle of matched deformations can also be applied to bearings as in Figlure6.27. Mention must also be made of welded joints. Here the residual stresseswhich occur on cooling and the stress concentrations caused by deflections of the force transmissionpath can be reduced by careful design (see [6.8, 6.12]). 6 4 P r i n c i p l e so f e m b o d i m e n t d e s i g n transmission path, and the right side a relatively low torsional stiffness because of its greater path length. when the torque is first applied, the left wheel will be set in motion while the right wheel remains stationary until the right hand part of the shaft has twisted sufficiently to transmit the torque. The drive assemblyhas a t e n d e n c yt o r u n s k e w . It is essential to provide the same torsional stiffness to both parts of the shaft so as to ensure an appropriate division of the initial torque. This can be achieved in two distinct ways if the input torque is taken in one position only: either by symmetrical layout (Figure 6.28b); or by adaptation of the torsional stiffness of the appropriate parts of the shaft (Figure 6.28c). 4 Principle of balanced forces in bearings: Figure6.27.Forcetransmission (a) edgecompression because of insufficient adaptationof the bearingto the deformed shaft (b) moreevenbearingpressurebecause of matcheddeformations (c) lackingadjustment to shaftdeformation (d) moreevenbearingpressurebecause of adaptability of bearingbush The principle of matched deformationsmust be taken into account not only in the transfer of forces from one component to another, but also in the division or combination of forces or moments. A well known problem is the simultaneous propulsion of wheels that have to be placed at a considerabledistancefrom onc another, for instance in crane drive assemblies.In the layout shown in Figurc 6.28a, the left side has a relatively high torsional stiffnessdue to the short force Figure 6.28. Application of the principle t=11+!7 o f m a t c h e d ,h e r e e q u a l ,d e f o r m a t i o n s in crane drives: (a) unequal torsional deformation of l e n g t h s1 , a n d 1 , ( b ) s y m m c t r i c a l a y o u tc n s u r c sc c l u a l t o r s i o n a lt l c l i r r r n a t i r l n ( c ) i r s v r t t n t c t r i c l li r y o u tw i t l r c r l u a l t o r r i o t t t l d c f o r t t t i t t i t t lttl t l c t < t: t t l l t l l l i t t i o n ol loniottul rlill ttcrrc;' Those forces and moments that serve the function directly, such as the drivins rque, the tangential tooth force and the load torque in a gearbox, can, in rdance with the definition of a main function, be described as functionally ermined main forces. In addition, there are many forces or moments that do not serve the function irectly but cannot be ignored, for instance: the axial force produced by a helical gear; the force resulting from a pressure difference, for instance across the blades of a turbine or across a control valve; tensile forces for producing a friction connection; inertia forces due to linear accelerationor rotation of components; and fluid flow forces inasmuch as they are not the main forces. Such forces and moments accompanying the main ones are called associated and may either produce an auxiliary effect (see auxiliary function) or else r merely as invariable concomitants. Associated forces place additronal loads on the components and require an ropriate layout or must be taken up by further surfacesand elements such as iffening members, collars, bearingsetc. As a result, weights are increasedand rther frictional lossesmay be incurred. For that reason, the associatedforces ust, whenever possible,be balancedout at their place of origin, thus obviating e need for a heavier construction or for reinforced bearins and transfer lements. As has been shown in [6.151], this balance of forces is essentiallyensured by typesof solution: balancing elements; or symmetrical layout. Figure 6.29 shows how the associatedforces can be balanced in a turbine, l i c a l g c a r sa n d a c o n e c l u t c h , w i t h t h e h e l p o f t h e p r i n c i p l eo f d i r e c t a n d s h o r t : c t r a n s r . r ' r r s s rpoant h . A s a r e s u l t . n o b e a r i n g p o s i t i o n i s a d d i t i o n a l l yl o a d e d n c l t h c c l c s i g r risr r c h i g h l y c c o n o m i c a l . W h c n i t c o l l l c st o l h c l ' l i r l a n c i n g of incrtia lilrccs,wc find that a rotationally n y n t r t t c t r i c il ti tl y o t t li s i r t h c r c n t l yb i r l a n c c c l . ' l ' hsci r n r cs o l u t i o np r i n c i p l ci s i r p p l i c d 204 6 Embodiment design for reciprocating masses, as we know from automobile engineering. If the number of cylinders is too small to ensure a perfect balance, either special balancing elements, weights or shafts [6.168] are introduced, or cylinders are arranged symmetrically, as for instancein opposed cylinder engines. balance without (small lorces) element balancing (med !m forces) symmray0ul (large lorces) C € 6.4 Principles of embodiment design XE The concept of the flowlines of force should be consideredin conjunctiorrrvith the following principles: The principle of uniform strength which ensures, through the careful selccriol of materials and shapes, that each component is of uniform strength irrrtl contributes equally to the overall strength of a device throughout its servicelilcl The principle of direct and short force transmission path which ensures nrinimum volume, weight and deformation, and which should be applied particularly if a rigid component is needed; The principle of matched deformatlons which ensures the matching of deformations of related components so that stressconcentrationsare avoided and the function can be reliably fulfilled. The principle of balanced forces which ensures, with the help of balancing elements or a symmetrical layout, that the associatedforces accompanying the main ones are reacted as close as possible to their place of origin, so that material quantities and lossescan be kept to a minimum. 6.4.2 Principle of the division of tasks I Assignment of sub-functions forcesillustratedby meansol Figure6.29. Fundamentalsolutionsfor balancingassociated a turbine,helicalgearsanda coneclutch Even during the setting up and variation of the function structure, it is important to determine to what extent severalfunctions can be replacedbv a sineleone. or hether one function can be subdivided into several sub-functions(5.3). These questions reappear in the embodiment phase, when the problem is to lfil the requisite functions with the choice and assignment of suitable function rriers. We ask: what sub-functionscan be fulfilled with one function carrier only? what sub-functionsmust be fulfilled with the help of several.distinct function carriers? So far as the number of components and the spaceand weight requirements concerned, a single function carrier fulfilling several functions would, of rse, be the best. In respect of the manufacturing and assembly processes, wever, this may prove disadvantageous,if only becauseof the complicated ape of the resulting component. Nevertheless, for economic reasons, the tempt should always be made to fulfil several functions with a single function As a general rule (which, however, can be ignored if there are overriding reasonsfor doing so) balancing elementsshould be chosenfor relatively small or medium forces, and a symmetrical layout for relatively large forces. All in all, we can say of the transmissionof forces (in the discussionof whiclr the physically undefinable but descriptiveidea of the flowlines of force is mosl helpful) that: - t h e f l o w l i n e s o f f o r c c m u s t a l w i r y sb c c k r s c d i i r n r l - s l r i r r p c l c l ' l c c t i o nosl t h c t ' k l w l i n c so t ' l i r r c ci r t t r lc l t i r r t g cisr t t l t c ' r l c n s i t y ' < l l ' t l r t ' l i n c s r c s r r l t i n gl ' r o r r rs t r t l t l e rcr ' l r ; r n r l t i' tst c t o s s s c c l i r l t tr t t t t s tb c r r v o i t l c r l . Numerous assemblies and components can fulfil several functions simuleouslv or successivelv. Thus a shaft on which a gearwheelhas been mounted transfersthe torque and rotating motion simultaneouslyand at the same time takes up the bending ents and shear forces resulting from the normal tooth force. It also locates g thc cars irxially ancl, in the case of helical gears, carries the axial force c o n l p o n c l l t sl r o n r t h c t c c t h . I n c o n j u n c t i o n w i t h t h c b o d y o f t h e g e a r w h e e l ,i t p n l v i r l c ss u l l i c i c n ts t i fl n c s st o c n s u r cc t l r r c c tr r r i r t i n o gl'thc tccth. A 1 t t 1 l cI ' l i t t t g c r l t t t t c c t i o trtt t i r k c sl l o s s i l r l cl l r c c o r r r r c c t i o inr n r ls c p i r r i r t i o no l t h c I c c O 206 6 Embodiment design pipes, ensuresthe sealingof the joint and transmitsall forcesand momentsin ini pipe resulting from residual tension, from thermal expansionor from unbalancedpipe loads. A turbine casingprovidesthe appropriateinlet and outlet flow areasfor the fluid, provides a mounting for the stationary blades, transmitsthe reaction forcesto the foundation, and ensuresa tight seal. The wall of a pressuretank in a chemicalplant must combinea retainingwith a sealingfunction and staveoff corrosion,while not interferingwith the chemical process. A deep groove ball bearing, apart from its centeringtask, transmits both radial and axial forcesand occupiesa relativelysmallvolume,for which reasons it is a popularmachineelement. The combinationof severalfunctionsin a singlefunction carrier may often prove economicallyadvantageous,but may have certain drawbacks.These do not usuallyappearunless: - the capacityof the function carrier hasto be increasedto the limit in respect of one or severalfunctions;or - the behaviourof the function carriermust be kept absolutelyconstantin one important resPect. As a rule, it is impossible to optimise the carrier of several combined functions. Instead,the designerhas recourseto the principle of the division of by which a specialfunction carrier is assignedto every function' tasks16.1551, Moreover, in borderline cases?it may even be useful to distribute a single function over severalfunction carriers. The principleof the divisionof tasks: - allows very much better exploitationof the componentconcerned; -provides for greaterload capacity;and unambiguousbehaviour, and hence fosters the basic clarity rlule ("trr1ttet 6.3.1) This is becausethe separationof tasksfacilitatesoptimum designin respectof everysub-functionand leadsto more accuratecalculations.In general,however, the constructionaleffort becomescorrespondinglygreater. To determinewhether the principle of the division of taskscan be usefully applied, the functions must be analysedwith a view to determining if the simultaneousfulfilment of several functions in one carrier introduces constraintsor mutual interference.If it does,then it is best to settlefor individual function carriers. 2 Divisionof tasksfor distinctfunctions of the divisionof tasksfor Examplesfrom variousfieldsillustratethe advantage distinctfuncti<lns. asfound for instancebetweena turbineand a gcnerator,it In largcgearhoxes. of the foundationsand bcarings,antl of thermalexpunsion is advisablc,hecause to usea radiallynnd torsiontrllyflcxiblc ulsebccauscof thc tonional oscillations. 6 4 P r i n c i p l e so f e m b o d i m e u t d c s i g t l 207 shaft whilst maintaining the shortestpossibleaxial length on the output side [6.150].However,becauseof the forcei betweenthe geaiteeth,the transmission shaftmust be asrigid aspossible.Here the principleof th" diuisionof tasksleads to the following arrangement:the gearwheelis fitted to a stiff hollow outer shaft with the shortestpossibledistancebetweenthe bearings,while the radially and torsionallyflexible componenttakes the form of an inner torsion shaft (Figure 6.30). igure 6.30. Large gearbox with an output torsion shaft; the bearingforces are ansmitted over a stiff hollow shaft; the inner torsion shaft is radiaiv and torsionallv ible, from [6.150](Siemens-Maag) Modern pressure-fedboilers are built with a membranewall, as shown in igure6.31.The furnacemustbe gas-tight.Moreover,optimumheattransferto water demands thin walls with large surface areas. Beyond that, thermal pansion and pressure differences between the furnace and its environment ust also be taken into consideration, and so must the weight of the walls. This mplexproblemis solvedwith the help of the principleof the divisionof tasks. tubularwallswith their weldedlipsconstitutethe sealedfurnace.The forces Iting from the pressuredifferencesare transferredto the specialsupports outsidcthe heatedareawhichalsocarrythe weightof the, usuallysuspended, walls. Articulatedarms betweenthc tubularwull and the supportsallow for 208 6 Embodirnent design P r i n c i p l e so f e m b o d i r n e n t d e s i g n 209 l-carrying function is performed by the clamp which, in its turn, is designed the principle of the division of tasks. The clamp is made up of segments, ich transmit forces and bending moments by means of a close tolerance fit, nd shrink rings hold the clamp segments together by friction in a simple and ffective manner. Every part can be optimally designed for its particular task d is easily analysed. The casingsof turbines must ensure a tight seal under all operational and al conditions if they are to conduct the working fluid with minimum loss turbulence. They must also provide an annular area and a support for the ionary blades. During temperature changes,sectionedcasingswith an axial nge have a particular tendency to distort and to lose sealing power due to rrked changesin shape at the inlet and outlet [6.166]. This effect can be offset by a separate blade carrier, that is by a division of The annular area and stationary blade attachment can be desisned rdlessof the larger casingwith its inlet and outlet sections.The outer casins n then be designedexclusivelyfor durability and sealingpower (Figure 6.33). Figure 6.31. Sectionof boiler with membrane walls and separatesupports (Babcock) unimpeded thermal expansion. Thus every part can be designedin accordance with its special task. The clamp connection in a superheatedsteam pipe shown in Figure 6.32has also been designedon the principle of the division of tasks.The sealingand load carrying functions are assignedto different function carriers: the sealing function is performed by the welded membrane seal, which is axially loaded by the tension in the clamp. Tensile forces or bending moments should not be carried by the seal, whose function and durability would thereby be destroyed, so the l ; i g r r r c6 , 3 2 ,( ' l i t r t t Pc r t t t t l c c t i t t ti t l i t iupctlrcnlc(l rlclrttt pipc ( Zikcsclt ) r e 6 . 3 3 A x i a l l y d i v i d e d t u r b i n e h o u s i n sf r o m [ 6 . 1 6 ( r l l;o w e r h a l f c o n v e n t i o n a lu; p p e r w i t h s e p a r a t eb l a d e c l r r i e r A further example is provided by the synthesisof ammonia, which involves ing nitrogen and hydrogen into a container under high pressures and atures. If the hydrogen were allowed to come into direct contact with a ritic steel container, it_ would penetrate into and decarbonise the latter. ucing decomposition at the grain boundarieswith the formation of methane ,72). The solution is again basedon the division of tasks.The sealingfunction provided by an inner casing of austeniticsteel which is resistantto hydrogen, 3 support and strength are provided by a surrounding pressure chamber tructecl of high-tensile ferritic steel, not resistantto hydrosen. l n t h c c l c c t r i c a lc i r c u i t - b r e a k e ri l l u s t r a t e di n F i g u r e 6 . 3 4 , t w o o r e v e n t h r e e t i r c t s y s t c n l sa r c p r < l v i d e d .T h e b r e a k c r c o n t a c t sl t a k e t h e a r c i n s c u r r e n t r i l t g t l r c c k l s i n g o r o p c n i n g o f t h e s w i t c h , i r n c lt h c m a i n c o n t a c t s3 c a r r y t h e r r c l t l t t l t t l c t ' t t o r t t t ci ttll l t t l i t i o t t s . ' l ' hl rcr c i r k c rc ( ) l t l i r c t/s i r r cs u bj c c l t r l b u r n i n q , 2r0 6 Embodiment dcsiln that is to wear and tear, and must be designedaccordingly,while the marrr contactsmust be designedto carry the full working current. The division of tasks is also illustratedin Figure 6.35: the Ringfedcr connectorscarry the torque while the correspondingcylindricalsurfacesensur(' 2II P r i n c i p l e so f e m b o d i m e n t d e s i g n the deep-groove ball bearing is not supported radially and hence transmits ial forces only, while the roller bearing transmits radial forces only. The principle of the division of tasks has been applied consistently to the struction of composite flat belts. They are made up, on the one hand, of a thetic material capable of carrying high tensile loads and, on the other hand, a chrome leather layer on the contact surface which provides a high coefficient friction for the transfer of the load. Yet another example is provided by the rotor blade attachment in a helicopter 6.1s). Division of tasks for identical functions increasesin load or size reach a limit, a single function can be assigned to ral, identical function carriers. In other words, the load can be divided and recombined later. There are numerous examples. The load capacity of a V-belt cannot be increased at will by increases ln lts section (number of load-carrying strands per belt) because, for a glven ley diameter, an increasein the belt height ft (Figure 6.37) leads to an Figure6.34 Figure 6.34. Arrangement of contactsin circuit breaker (AEG) 1 breaker contacts;2 intermediate contacts;J main contacts Figure 6.35 Ringfeder connector plus centralisingsurfaces the central location and seating of the pulley, something the Ringfeder connector could not provide by itself. A further example is provided by the design of rolling element bearings irr which the servicelife of the locating bearing is increasedby the clear separatiorr of the transmissionpaths of radial and axial forces (Figure 6.36). The outer racc load-carryi ng strands fabriccoating Figure6.37.Crosssectionof V-belt rubber filling ase in the bending stress. As a result of the ensuing deformation, the r, which has hysteresisproperties and is also a poor conductor of heat, overheated and this reducesits life. A disproportionally wide belt, on other hand, loses the stiffness needed to take up the normal forces acting on wedge-shapedsurfacesof the pulley. An increasein load-carrying capacity , however, be obtained by dividing the overall load into part loads each ropriate to the load limit and normal life of the individual belt (multiple ment of parallelV-belts). The coefficient of thermal expansion of superheated steam pipes made of nitic steel is approximately 50 per cent higher than that of pipes made of the I ferriticsteel.Suchpipes,moreover,areparticularlystiff.At constantinner res and fixed materialpropertylimits, the ratio of outer to inner pipe ter remainsconstantif the inner diameteris changed.However,whilethe roughput at constant flow velocitiesvaries as the squareof the inner diameter, F i g u r c6 . . l t rl,. o c i t t i t rbgc l r i n gw i l h pittltslor rcpglolclrttnstrtissiott rldld und ttxinllirrccs bcndingand torsionalstiffnesses vary asits fourth power.The substitution of pipc lincsfor a singlclargepipewouldadmittedlyleadto increased pressure h c i t l f r t s s c sl i r r t h c s a l n c f l o w a r c a , b u t w o u l d r c d u c c b v l l z t h c s t i f f n e s s 212 6 Embodiment tltsr resistingthermal expansion. With four or eight pipelines the individual reaclrrrtl forces would then be no more than I or * of that present in a single pipe [6.3{1, 6.2061. In addition, the reduction in wall thickness leads to a reduction in thermal stresses. Gearboxes, and epicyclic gearboxesin particular, make use of the principlc ,rf the division of tasks, or rather of forces, in the form of multiple meshing, whit lt will increasethe transmissioncapacityof the gearbox, provided that the therrrrrrl effects can be kept within reasonable limits. In the symmetrical layout ,tl epicyclic gearboxesbased on the principle of balancedforces ( 6.4.I.4) even t lrc bending moment in the shaft is eliminated because the forces produced by llrc gearscancel out. However, the torsional deformation is increasedbecauseo1'Ilrc greater load capacity (Figure 6.38). In large gearboxes,this principle is applietl a1t of embodiment design ztJ the caseof the multiple pipeline discussedabove,the individual pipe loss ts, the relationships between inflow and outflow, and also the geo- of the pipe layouts must be kept similar, or else the individualloss ients must be small and not very greatly affected by the flow speeds. the case of multiple gears, either a strictly symmetrical arrangement must equal stiffnessesand temperaturedistributionsthroughout the gearbox Double Triple 0uadruple 6.39 Figure6.40 6.39.Basicarrangements of multiplegears,from [6.150] re 6.40. Balanced forces in multiple gearsby meansof flexible torsion shafts,from forces,from [6 2] Figure6.38.Epicyclicgearboxwith balanced to advantage in the form of multiple drives equipped with spur gears, which havc external teeth only and hence are more easily manufactured. As Ehrlensprcl [6.57] has shown, it is possible to increasethe load capacitywith the number of force transmission paths, though not in direct proportion because each stcp introduces a different flank geometry with a slightly greater flank loading. Basic arrangementsare depicted in Figure 6.39. One problem of the principle of the division of tasks is the uniforrn participation of all the elements in the fulfilment of the function, that is thc provision of a uniform distribution of forces or loads. In general, this can only bc achieved if: -the participating elements adjust themselvesautomaticallyto balance out thc forces; or - appropriateflexibility is speciallyprovidedin the forcetransmission paths. In the caseof multipleV-belt drivcs,thc tangcntialforccsproducesliglrt crrorsin thc lcngtlts nnydimensional of the beltswhichhclp to ol'l'sct extensions in thc shaft.and thtt\ of thc bcltsor in thc pullcysor uny lnck of parallelism cnsurccqull loirdshirring. I flexibleelements[6.58]must ensurethe equalparticipationof all the nts. re 6.40 illustrates a flexible arrangement.Further balancingcomponents, as elasticand articulatedjoints, are describedin [6.58]. ll in all, the principle of the division of tasks provides for increases in the imum load capacity or for wider applications. By spreading tasks over al function carriers, we also gain a clearer picture of the relationship n forces and their effects, and, what is more, can increase the output, only that a balanced division of forces is maintained by adjustable or -regulatingelements. In general, the application of the principle of the division of tasks calls for ter effort on the part of the designer, which must be offset by greater overall or safety. .3 Principle of self-help Concepts and definitions thc lastsectionwc discussed the principleof the divisionof tasksand showed it couldhclp to incrcasckradcapacityand to provitJea clearerclefinitionof 6 E m b o d i m e n t t l er r l r r 214 the behaviour of the components. To that end, we analysed the variorrr sub-functions and assigned them to such function carriers as neither influerrt.' nor interfere with one another. The same analysis can also be used in conjunction with the principle t,l self-helpto achieve, through the appropriate choice of systemelementsand the ir arrangement, a mutual supportive interaction that improves the fulfilment of tlrr' 'self help' provides 1or function. Under normal conditions (normal loading), greater effect or relief; in emergency situations (overloading), it provides 1t,r greater safety. In a self-helpingdesign,the overall effectis made up of an initial effect ancl rr 2t5 P r i n c i p l e so f e m b o d i m e n t d e s i g n ral technical characteristics:efficiencv.servicelife. use of materials,techl i m i t , e t c . I t is defined as technical characteristic with self-help technical characteristic without self-help never the application of the self-help principle calls for a greater effort on designer'spart, then it must bring clear technical or economic advantages. Identical design approaches may turn out to be self-helping or self-damaging, nding on the layout. Take the case of an inspection cover (Figure 6.42). So supplementaryeffect. The initial effect setsoff the physicalprocessrequired by the solution but rs insufficienton its own. The supplementaryeffect is obtained from the functionally determined mairr forces (gearbox torque, sealing force etc) andior from the associated forct': (axial force produced by helical gears, centrifugal force, force due to thermirl expansion etc), provided, of course, that the two sets of forces are clearl\ correlated. A supplementary effect may also be obtained from appropriat, changes in force transmission paths. The idea of formulating the self-help principle was first suggested by the Bredtschneider-Uhde self-sealingcover, particularly suitable for pressure vess e l s [ 6 . 1 1 3 , 6 . 1 5 4 , 6 . 1 8 1 ] . F i g u r e 6 . 4 1 s h o w sh o w i t w o r k s . A r e l a t i v e l ys m a l l force provided by the central bolt 2, suffices to press the cover 1 against thc metal seal 5. The initial effect of this force ensures that the parts make thc proper contact. With increasing operational pressure a supplementary effect is produced, thanks to whrch the sealing force between cover and tank is increaserl appropriately. The internal pressure thus provides the required sealing forcc automatically. t-s lD I I ng sell-helpi selt-damaging re 6.42 zlt ping sell-hel 4-/ ll sell-damag ing Figure6.43 6.42. Layout of an inspectioncover i n i t i a le f f e c t ; S : s u p p l e m e n t aer yf f e c t O ; : overalleffect;p: internalpressure re 6.43. Force diagram to Figure 6.42 t : tensionscrew;subscript changein tength; subscript forces; Fo : preload; ll: flangeiseal 2 3 4 5 1 Figure 6.41. Self-sealingcover 1 cover; 2 central bolt;3 crossmember; 4 element with saw tooth thread; 5 metal sealing ring; p : internal pressure; d: temoerature contribution of the supplementary It may be usefulto specifythe quantitativc prodttcittg O in thc dcgrcc of self-hclp cl'fcctS to thc ovcrall effect x=SlO-(f..,1 'l'hc gnin l'rom scll-lrclpsolutionrcan ll| OrPrcmcdin tcrms ol' thc onc or -r&,- as the pressureinside the tank is greaterthan the pressureoutside,the becausethe pressureon the cover (supplementary ut shownis self-helping, ) increases the sealing effect (overall effect) of the initial tension-screw (initial effect). The layout shown on the right, by contrast, is self-damagingbecause the ssure on the cover dereasesthe sealingeffect (O) of the initial tenslon-screw (1). If, however, the tank were kept at below atmosphericpressure,the left t would be self-damaging, the right layout self-helping (see also Figure 43). This example shows that the degree of self-help depends on the resultant :t: in thc presentcasethe effect on the sealingforce resultingfrom the elastic , irncln<lton thc simple addition <lf thc forcc exerted by the screw and the :c ilcling 0n thc c()vcr. 276 6 Embodiment desigl Figure 6.43 can also be considered as a force-deformation diagram of a bolterl connection with a preload and a working load. The conventional bolted flangt' connection may be called self-damaging inasmuch as, under operational condi tions, the overall effect-that is the flange sealing-becomes smaller than thr' preload. Also the loading of the bolts is increased at same time. If possiblc. therefore, only such self-reinforcing arrangements should be chosen as increase the overall effect, while reducing the loading of the bolts (Figures 6.44 a-l illustrate such arrangements.) 217 P r i n c i p l e so f e m b o d i m e n t d e s i g n This group of self-helpingsolutionsis the most common. Under part-load itions, it ensures grealer service life, l e s s w e a r . h i g h e r e f f i c i e n c y e t c , use the components are only loaded to an extent needed to fulfil the tion at any particularmoment. As a first example, let us consider a continuously adjustable friction drive 6.4s). 12 x Adlustment / I I I 0,2 boe re 6.45 (a) multiple disc clutch with adjustment ring; (b) force acting on the adjustment ring; (c) adjustable disc of two-disc friction clutch; (d) crown wheel attachment,symmetrical take-up of forces For practical purposes, it is useful to classify self-helping solutions accordance with Table 6.3. Table 6.3. Summary of self-helping solutions Normal load Type of self-help self-reinforcing Supplementary effect due to main and associated associatedforces forces Important features main or associated forces act in the same senseas other main forces self-balancing associatedforces act in the opposite senseto main forces Overload self-protecting altered force transmissionpath force transmission path altered by elasticdeforniation; limitation of function permissible 7 /,'. s 7 Lz c0 Figure6.44.Self-reinforcing boltedconnections l,=oz 7 0 8t I 0,6x 0./0,2 F.l 0.4 r/r 1,0 0,6 0,8 0 Figure6.46 frictiondrive adjustabte re 6.45.Continuously load spring; b: drive shaft; c: cup wheel; d: conewheel; e.'roller follower;f cam on the cup wheel; r. radius on which F, and F,,act 6.46. Degree of self-help (x) and i n i t i a l ( / ) , s u p p l e m e n t a r y( S ) a n d o v e r a t l( O ) against the relative torque Z/T.u* for the friction drive (Figure 6.45) The preload spring d pressesthe freely movable cup wheel c on the drive shaft against the cone wheel d, thus providing the initial effect. Once a torque is ied, the roller follower e attached to shaft b is pressed against the cam / d on the cup wheel c where it produced a normal force F" that can be lved into a tangential force F, and an axial force F, which, for its part, s the contact force F. applied to the cone wheel in a fixed proportion to appliedtorgue T: Fn: Tl(r'tan a). The force Fu representslhe supplementaryeffect gained from the torque. The rall effect is obtained from the spring preload force Fo, plus the axial force F,, ich varies as the torque Z (c/Figure 6.46). The tangential driving force F6 on cone, which determines the transmittable torque, is therefore Fa:(Fp+F^).tt 2 Self-reinforcing solutions I n s e l f - r e i n f o r c i n gs o l u t i o n s .the supplcmcntilry effcctisubtaincddirectlyfrom ir main or associatedforce and addsto thc initiuleffectto proclucc a grcatcrovcrlll cffect. t h c c l e g r e eo f s e l f - h e l px : S I O : F ^ l ( F p + F ^ ) . It is obvious that the contact pressure between the wheels, which helps to r r n i n c t h e w c a r a n d t h e s e r v i c el i f e < l f t h e d r i v e . m u s t n o t e x c e e dw h a t i s i c l l v r r c c c s s i r r vA. c < l n v c n t i o n asl o l u t i o l t ( n o s c l l - r c i n f < l r c c m c n tw) o u l d h a v e 278 6 Embodiment desigrr demanded an axial force produced exclusively by the spring preload corresponding to the maximum torque, and therefore maximum pressure being applied to the contact area under all loads. As a result the bearings, too, would have had to carry a considerably greater load, which would have led to a reduced service lifc or demanded a much heavier construction. A rough calculation shows that if the actual loading is, say, 75 per cent of thc nominal maximum load, then the bearing load would be reduced by about 20 per cent which, because of the exponential relationship of service life to load, can lead to a doubling of the life of the bearings. In that case, the self-help gain in respect of the service life becomes: Life with self-help : TL: Life without self-helo /r-rn np\ n l-"'' | :L253:2 219 Principles of embodiment design trghtening before altertightening 0 \ctP I Figure 6.44 shows various self-reinforcing layouts of contact surfaces loadecl by bolts, in which the frictional forces are increased by the operational forces while the bolts themselves are off-loaded. The application of the principle of self-help in the design of self-reinforcing brakes has been describedby Kiihnpast [6.113] and Roth [6.176].Depending on the application, even self-damaging,and in this case self-weakening,solutions can prove interesting, inasmuch as they reduce the effect of variations of the coefficient of friction on the braking moment 16.62,6.1761. Self-reinforcing seals (Figure 6.47) provide us with further examples. In them. the operating pressure against which the seal has to be applied is used to produce the supplementary effect. Finally we must mention one case in which the supplementary effect is produced by an associated force. In hydrostatic axial bearings, the centrifugal inertia effect leads to an increase in oil pressure which, at high revolutions, will help to improve the load-carrying capacity, provided the heat can be removed (see Figure 6.48). The supplementary effect leads to an improvement in the load-carrying capacity due to the increased oil pressure resulting from the centrifugal effect alone, the overall effect is due to the load-carrying capacity of the combined static and dynamic pressures.According to Kiihnpast [6.113] it should be possibleat, say, 166 revls and x :0.38, to obtain a gain in self-helpof v:1.6. The supplementary effect of another associatedforce, namely that caused by the effect of temperature on the shrink-fitted rings of a turbine, is disiussed in [6.1s4]. 3 Self-balancing solutions In self-balancing solutions, the supplementary effect is obtained from an associatedforce, and offsets the initial cffcct to produce an improved overall effect. A simple examplc is providcd by turbo-machincr. A bludc uttachcd to a rotor Figure 6.47. Self-reinforcing seals; (a) setf-sealingwasher; (b) tubeless tyre; (c) radial-shaftseal; (d) sleeve seal; (e) sliding-ringseal 05 N mm2 ^s-/ \o/ 04 \l s/ I 03 I oo2 I /' 0,1 99' \ n-A 0 \ 510 \ 20 30mm38 12 13 from [6.113] axialbearings, Figure6.48.Self-helpeffectin hydrostatic iect to a bending stressdue to the tangential force acting upon it and also to axial tensile stressdue to the centrifugal inertia force. The two are additive , becausea certainstressmust not be exceeded,the transferable taneential is rcduccd(Figurc6.49).lf, howevcr,the bladeis attachedat an ansle,a 220 6 Embodiment clcsrll supplementary effect is produced: an additional bending stress due to tlrr. centrifugal force acting on the offset centre of gravity of the blade opposes tlrt. original bending stress and thus allows the application of a larger tangentirrl force-that is a greater overall effect. How far this balancing process can l)(, carried depends on the aerodynamic and mechanical conditions. 22r ples of embodiment design rcular, we try to protect componentsthat are frequently subject to slight . If special safety arrangements,for instanceto limit the load, are not ial, then a self-protecting solution may prove advantageous. It will timesbe simplicityitself. '-protecting solutions derive their supplementary effect from an additional transmissionpath that, in caseof excessloading, is generally created after :n elastic deformation has taken place. As a result, the distribution of the ines of force is altered and the load-carrying capacity increased. Admittedin that case,the functional properties associatedwith normal conditions may me altered, limited or suspended. : springsshown in Figure 6.51 have such self-protectingproperties. In case 'Es h o c k M c : F Bl : F c e. Fe:Fcsin(o-t) -f- ),,*oo a CG: Centre ofgravity Fr: Tangential rrl lorce; F,: Centrifugal force; o6r: Bending stress duetoF1;o. : Axial stress duetoF6;ouc: Bend component olF6 stress duetoF6;Fa: Axlal component ofF6;FsBending T1t o4 ],, F solutionfor turbineblades; Figure6 49 Self-balancing (a) conventional solution;(b) leaningof the bladeproducesa balancing supplementar,v effectdueto the additionalbendingstresses producedby the centrifugal force(os,) whitlr opposethe bendingstresses force(ou,); (c) diagramof forces causedby the tangential A self-balancing effect can also be produced by allowing thermally inducctl forces (stresses)to oppose other forces (stresses);for instance those resultirru from excessor other mechanical loads (Figure 6.50). All the examples we have given are intended to encourage the design ol technical systemsensuring that: - the forces and moments with their resultingloads cancelout as far as possible: or that - additional forces or moments are produced in a clearly defined way so that rt is possible to balance them out. 4 Self-protecting solutions I n g c n c r a l , i n t h c c v c n t o f i r n o v c r l o l t l . w c ( l ( t l t ( t t w i l r l t c o r n l ) ( ) r l c l l l st o l r r ' d c s l r o y c r.l u n l c s s r, r l c o r r r s ct.h c v h l r v r ' l t ' c t t l c l i l r c r i t tl ev t l t ' s i g t t c;rrl sw e ; r kl i r t k t b I I tr5 0 re6.50 a Figure6.51 :6.50. t{oop stressesin a thick-walledcylinder due to the internal pressureonnand :rllure differcncesat nearly steadyheat flow o6,n; n o t t - l t i t l i t t r c l tst go l t t t i o l t ,t h er r n a ls t r e s si s a d d e dt o t h e m a x i m u mm e c h a n i c asl t r e s so n l l l n c l s u rl i t c c : t r c l l - h ; 1 1 ; 1 1 1 t ' l l l g s o l t l l i o t t . lsl t rccr sr ns uo lp p o s c s m a x i m u m r n e c h a nsi tcraels s o n t h e s t t rl r r c c c { r ' 5 1 S c l l ' 1 l l ( ) t c ( ' t ist togl t t l i o titt rs l l t i t t g s(;l r )t r r ( t l ) l o r c ' ct r ' : r r r s r n i s s i o r r llrrth i c r l ,t l l t ' t l r r t t t t ;fttlt t t t ' t t r litrt r r t s l t t ' t t t l ro' rt ll i r r r i t c ti ln r ' l r s o c l c r c c s sl o l r t l r r i g 222 6 Embodiment dcsilrr of excessloading, the spring elementswhich are normally subject to torsional trr bending stresseswill transmit the additional force directly. The same effect mrrv also be produced if the springs are shock-loaded(Figure 6.51b). Figure 6.52 shows the layout of elastic couplings in which restriction of thc spring movements provides additional force transmission paths with consequcnt loss of flexibility but with increasedload-carrying capacity. In Figure 6.52a, the load-carrying capacity of the bar springs is altered inasmuch as, besides th,' normal bending, a powerful shear force between the two halves of the couplirrl appears with overloads. 223 Principles of embodiment design .4 Principles of stability and planned instability mechanics, we know the concepts of stable, neutral and unstable ilibrium, as illustratedin Fieure6.53 In elaborating solutions, the designer must always consider the effect of urbances and try to keep the system stable by devising means whereby disturbancescan be made to cancel out, or at least to mitigate one another. stable r e6 . 5 3 racteristics of Afterdisturbance, the system returns automatically to itsold position andequilibrium state Upondeflection, the potential energy ol the deflected bodyincreases andimposes a return tothe position original Afterdisturbance, the syslem adopts a new position withunchanged equilibrium state Upondeflection, the potential energy remains constant After disturbance, the system adopts a new position andequilibrium Upondeflection, the potential energy ol the dellected bodydecreases andimposes a newposition SIAIC uilibrium states isturbances are self-reinforcing, we have unstable behaviour. This effect 1 S irablein certainsolutions, in which case we speak of planned instability. I [ ilp 0b solutionin couplings;changeof forcetransmissionpathswith Figure6.52.Self-protecting (a) bar springcoupling;(b) elastic lossof elasticpropertiesin caseof overloading; couplingwith coilspringsandspecialbuffersto takeup the forcesin caseof overloadinr Figure 6.52b shows a coupling that, strictly speaking, may be considereclit borderline case between a division of tasks and a self-protectingsolution. Tlrc buffers will only take up forces in case of overloading. The characteristic of thc spring elements remains unchanged. However, the force transmissionpath is altered after a given elastic deformation has taken place. Kiihnpast [6.113] alse mentions cases in which there is an uneven strcsri distribution over a cross-sectionand where plastic deformation can then be usctl for purposes of self-protection. In such cases, however, sufficiently touglt materials and adequate dimensional stability are needed. I t i s h o p e d t h a t t h e p r i n c i p l c o f s c l l ' h c l p b a s c d o n s c l f - r c i n f o r c i n g ,s c l l ' b a l a n c i n ga n d s e l f - p r o t e c t i n gs o l u t i ( ) n sw i l l c n c o u r a g ct h c c l c s i g n c rt o c x i t n t i t t c every conccivahlc arrangcnrcnt in un cffort lo urrive ttt an cffcctivc itntl c c o n o r n i c i t sl o l t t t i t l n . of stability applyingthis principle, the designertries either to ensurethat disturbances I out or else to reducetheir particular effects. uler 16.167lhas discussed this subjectat lengthand we shall now look at of his examples. the designof pistonsfor pumpsor regulatingdevices,the main objectiveis achieve stable behaviour and minimum friction. igure 6.54a shows the layout of a piston with unstable characteristics. rbances due, say, to inaccuraciesin the cylinder bore can tilt the piston tly and produce pressure distributions over the piston that encourage tilting (unstable behaviour). Stable behaviour is ensured by the layout in Figure 6.54b, which, however, has a disadvantage:the piston rod inlet to be sealed off on the pressureside. According to [6.167], the layout shown in Figure 6.54a canbe stabilisedby the res shown in Figure 6.55 a-d. They ensure that a disturbance will itself iatc suchprcssurcdistributions as tend to correctthe misalignment. Anothcr exilmplcis thc wcll known cirscof hydrostatic bcaringswith oil s distributcrlirrounclthc pcriphcry.whcn thc bqrring is loaclccl. rhc 114 6 Embodimenttl.'r leakage path below the load is reduced with the result that pressurebuilds up in the affected oil pocket and decreases in the opposite one. Thanks to tlro combined effect, the bearing can take up the load with very small shrrfl displacement. The stuffing boxes and sealsof turbo-machinery must always be designecll,rl thermo-stable behaviour 16.1671.The seal of a turbo-charger shown in Figrrrr.. 6.56 is a casein point. In the thermo-unstablelayout (Figure 6.56a) most of llrc frictional heat generated by contact forces will flow into the rotor which will ht'rrt up further, expand, and hence increase the contact forces. In the stalrlc arrangement (Figure 6.56b), by contrast, the frictional heat will cause tlrc contact forces to be reduced. A disturbance thus produces a self-limiting eflcr t Figurg6 54, Pistonin cylinder, tilteddue to a disturbance. from 16.1,671; (a) resultingpressuredistribution producesan effectthat increases the disturbance(unstable behaviour); (b) resultingpressure distribution producesan effectthat opposesthe disturbance(stablebehaviour) of embodiment design 22s similar approach is used in the design of taper roller bearings.Thus, in the shown in Figure 6.57a, heating of the shaft, by excessiveloading for :, will tend to rincrease rrwrv4rv the L r r w rload vau even L vLrl l u t L r t g l because further uE;LduSE of u I the t r r c texpansion s)xpalrslult of o l shaft due to the increasedfrictional heat. The arrangementshown in Figure rb, by contrast, will lead to a load reduction. In the caseunder consideration. \ 6.57.Taper roller bearings in which the shaft heats up more than the housing; thermal expansionleads to increasedloading and hence to unstablebehaviour thermal expansionleads to reduced loading and hence to stablebehaviour reduction must not, however, be allowed to reach the point where one of the ings becomes unloaded, because the shaft at that point would then not be radially and the bearings easily damaged. nother interesting example of thermo-stable behaviour is provided bv the elical gearsused in marine gearboxes16.2391. ple of planned some cases, unstable or bistable behaviour is positively welcome. This when,on reachinga limit, a clearlydistinctstateor positionis required no intermediate state is acceptable. The requisite instability is initiated when ected physical quantity reaches a limiting value and then introduces for improvingthe resultingpressure Figure6.55.Measures distribution,from [6.167]; (a) unstablebehaviourmitigatedby pressure-equalising grooves; (b) stablebehaviourthroughconicalpiston; (c) throughpressure pockets; (d) throughjoint fittedabovecentreof gravityof the piston inforcing effects. well known application is in the design of safety and alarm valves [6.l6jl , on reaching a limiting pressure, will spring from a completely closed to a letely open position. This avoids undesirable settings with a low flow rate tter and wear of the valve seat (see Figure 6.126). ure 6.58 illustrates the solution principle. Up to the limitingpressure p: p.r,the valveremainsclosedunderthe preload the spring. If this pressureis exceeded,then the valve head will lift off verv ly. The result is an intermediate pressurepi,rhe valve head throttling the t. This intermediate pressureacts on the additional surface,4uof the valve and produces a supplementary opening force that offsets the elastic force of springF, to suchan extentthat the valveheadlifts rapidly.In the openstate, nt intermediate pressurep' is set up and keeps the valve open. To close valve,the pressuremustbe reducedconsiderably belowthe limitingopening becausc,in the openstate,the pressureis appliedto a greatersurface Figurc6.56.Scitlin turtr<tchsrgcr.fronr16.l67l One applic:rtionis thc prcssureswitchfor monitoringbearingoil pressure in l"'igurc(r,59,ll'thc bctringoil prcssurc dropsl'rcklwl ccrtainvalue,the 226 6 Embodimcntdcsisrr h =h, h=0 Figure6.58 Figure6.59 Figure 6.58 Solution principle for a valve with an unstable opening mechanrsm d: precompression of spring; s : stiffness of spring; { : spring force; /z : lift of valve head;p : pressureon valve;p, : limiting pressurejust sufficientto open the valve; pr : intermediate pressure on opening of valve ; p' : pressure after opening of valve; pe: atmosphericpressure;Au: valve-openingsurfacearea;Au: additional surfacearc;r Valve closed: F,: s' d> p' A,. h--0 V a l v ej u s t o p e n : F,: s -d { A. A,, h:0 Valve opening fully: F,= s(d + h) < p . A, + pi. At. h-,->h, Valve fully open: F,: s(d + hr) : p'(A, + A^), h : fu (newequilibrium positiorrI Figure 6 59. Diagrammatic sketch of a pressureswitch to monitor bearing oil pressure. from [6.167] 1 main oil system pressure;2 orifice,3 safety system activating rapid shut-off valles ; 4 drainage(no pressure);5 bearing oil pressure s +$/' $7 6.5 Guidelines for embodirnent design 227 piston jumps open and the pressure inside the safety system is reduced with consequent shut-off of the endangeredmachinery. T h e - p r . i n c i p t eo f p l a n n e c li n s r i b i l t t y , s h u t - o f f d e v i c e si n w h i c h a s t r i k e r p i n gravity slightly offset from the centie o angular speed is reached, the striker p r e l o a d . T h e r e s u l t i n gi n c r e a s ei n t h e e to an increase in the centrifugal force ar w i l h o r r t any e n r r further frrrrLo-:-^-^^^^ i .' without increase in the ang, the rate of increase in the centrifusai g r e a t e rt h a n t h a t o f t h e o p p o s i n g spiin, pin begins to move. The forces -uit Ui can be achieved provided that dF"/dr> once it has been displaced to th" outside, the pin strikes a catch which, in turn, activatesthe rapid shut_off mechanism. 6.5 Guidelines for embodiment design 5.1 General considerations addition to the three basic rures (clarity, simplicity and safety) derived from general objectives (g:1, the designer must also follow a number . rurtruvl ofl u guidelinesbased on ih" generarconstraints r v l out set v u r in \ L t 2.r.6 L'r'\) n dP:j]T:tt,q:sign the checklist in Figure 6.2. These guidelines cover the respective constraints and requirements and are in ping with the basic rules. In what follows we shall cover what we considerto t!9 most important of them lrrLru without wru_r(rLil_ rnaKlng any making any claims Clalms to comDreteness. completeneSs. n dirn y o.witrrvinenLv withm rrr,2>s, unstable behaviour startlng fromcu: ar, Figure6.60 Figure6.61 I l i g u r c ( r ( r 0R a p i d s h u t - o f f p i n l i n s h a f t - ] w i t h c c n t r cgor li r v i t y C G o f f s e t b y c a n c l s p r i r r l 2 I r o l t l i r r gt h c p i n i n t h c n o r n r a lp o s i t i o n ,f r o r t r1 6 ,l ( r 7 l l i i g t r r c( r . ( r l ( i r t r l l h o l s p r i n gl i r r c c i r r r t tl ' t ' r r t r i l r r g ilrol t r ' c ; r gr i r r r tt l t c t l i s l l l l r c c n t c rrr to l l l r t . c c t l t r co l g r i r v i t yo l t h c r a p i d s l t u t - o l lp i n ( l J i g u r c{ r , { { l ) ,e - c c c c r r t r i c i t oy l c c n t r c o f ' g r ; t v i t y :r / - s P r i n gJ r r c c o r n P r c s s i orl,rr;, l i n t r t t r t ln n l u l n t r r ; x c t lh c y o r r rw l h i c l rt h c l l i r r l i l t , oll u- ^ u.t* or special . which :i 1",:have f :::: l':l' l_1I:,0.: to "1 the reader will "., counts been published, be referred. Designing for durability (stressrequirements): the designeris ,li.r."o to tt. eraturecoveringmachineelements [6.6i, 6.7g,6.I0g,6..*5, O.iOl,6.2071. He shouldHqJ pay rpvLr.r specialdrrtrr'u' attentionto ro cnanges changes in loadrng ln loadingconditions conditionswith with rime :r time and the correct estimates of the level and type of the resulting stresses. amage-accumulationcriteria help to improve service-life predictions [6.79, 1 9 1 ,6 . 2 0 9 1 . In.determining stresses,the designer must take stressconcentrationsandior ultiaxial axiarsrress conditions inri account s.ii)",-i.2ts,6.240]. [6,.25,6.r.4r, sscssmcnts of durabilitymustbe basedon the materialprop".ir., and the l p p r . ; r r i i r rfci r i r u rccr i t e r i [a6 . 2 2 . 6 . 7 6 , 6 . g 6 , . 6 . r 3 g , 6e..2i r0i ,2o, . z z o , 6 . z 2 r l . : I)csignirrg t0 iril.w f<tr dij orrnatictn.,stahility anrJresonance:'the designer is ,:l]r]l,lll1i.rc c.tcur.1i.ns in nrcch.rrics ,,,,,i,,,u.ii,,"ctynamics: Ij:1,::1 l ' c c h l r . ::' i c s.ill: . i r ' st lr r c r r g t h . p r . b r c r r r s l o .r v . o l , t , 6 l l 7 . 6 .r 9 2' ,,6 , ,."2i 10 .r,l .lv, i, r1r r i r t i r l r r p r o h l 1 , l 1l (sr\, l ( ) 6 {, r l J i l : s t i r b i l i t p y r o l r l t , r rlr(rr l ( ) l l ; , , , , , 1 .l tirritc 228 6 Embodimeutrlcslr' elements [6.250].In 6.4.1 we have dealt briefly with the problems of designirrrr with due allowance for the deformation caused by the transmission of forces Designing to allow for expansion and creep, that is temperature phenometr;r, will be discussedin 6.5.2 and 6.5.3, and designingagainst corrosion in 6.5.4 Wear poses an extraordinarily complex problern that is currently beirl examined from many sides. The reader is referred to the literature [6.5, 6.lt t 6 . 1 0 56, . 7 t t ,6 . 1 6 26, . 2 3 4 1 . Safety problems are treated at some length in 6.3.3. General ergonomic problems are discussedin [6.96, 6.140, 6.764, 6.1]'' 6 . 1 8 7 ,6 . 1 8 8 ] .T h e e r g o n o m i cu s e o f e q u i p m e n ti s d e a l t w i t h i n 1 6 . 4 6 , 6 . 4 ,16 . e \ 6 . 1 I 2 , 6 . 1 1 7 , 6 . 7 7 3 , 6 . 7 7 4 , 6 . 1 8 4 ]M . o n i t o r i n g a n d c o n t r o l a r e c o v e r e di n [ 6 . 1 . 1i 6.I72, 6.I89]. The harmful effects of noise and their mitigation are examineclrrr [ 6 . 6 5 , 6 . 8 3 , 6 . 8 4, 6 . 1 1 4, 6 . 1 8 6 , 6 . 2 7 r, 6 . 2 1 7, 6 . 2 2 5 ] . The form design of technical products involves special rules which 31s ssf oLrl i n [ 6 . 5 6 ,6 . 1 0 3 ,6 . 2 1 . 8 ] . Design for production and assembly (including quality control and transport) s dealt with at some length in 6.5.6 and 6.5.7. Designing to standards(see 6.5 .t helps with this aspect and also in reducing costs and improving schedules. The problems involved in design for operation and maintenance depend vcrr m u c h o n t h e p r o d u c t a n d i t s u s e . T h e r e a d e ri s r e f e r r e d t o [ 6 . 4 5 , 6 . 9 3 , 6 . 1 ( ) s 6.I35,6.2461 and also to the literature listed under ergonomics above., 6.5.2 Designing to allow for expansion Materials used in technical systemstend to expand when they are heated. Tlrt' resulting problems must be taken into considerationnot only in the design ol thermal devices in which higher temperaturesmust be expected as a matter ol course, but also in high-performance engines and devices in which frictionrrl h e a t i n gc a n o c c u r a n d s p e c i a lc o o l i n g i s e m p l o y e d .A s a r e s u l t ,s e v e r a la r e a su i l l be affected by local heating. Moreover, deviceswhose environmental tempcr;r. ture fluctuates significantly will only work properly if the physical effects ol thermal expansion have been allowed for in the design [6.154]. Apart from the thermal effects of linear expansion, the designer must also consider the purely mechanical expansion of parts subjected to heavy loading G u i d e l i n e s f o r em b o d i m e n t d e s i g n nate axis only, while the coefficient of cubical expansion defines the tive change of volume per degree of temperature rise. For homogeneous its value is three times that of the coefficient of linear expansion. loefficients of expansion should be understood as mean values over the icular temperature range/0^; they depend not only on the material but also the temperature. At higher temperatures, the coefficient usually increases. Figure 6.62 gives the coefficients of linear expansion of distinct groups of ineering materials. It shows that with commonly used combinations of ls, for example of 35C carbon steel with austenitic (I\CI18% Cr-Ni-Nb) .-lnvar (64%Fe,36%Ni) ..- Tungsten Molybdenum - Chromium ....- Vanadium Greycastiron (10C/13% Cr) 13%Chromium Steel Titanium, 35CCarbon steel - Pureiron - Gold _ Austenitic stainless steel(10Cl1Bk Cr-Ni-Nb) Copper + Bronze 60 ._ Polycarbonate *- Brass Tin Aluminium alloy(cast), g0 .- (homopolymer) Polyacelal ..- Aluminium - 66 Nylon .- Magnesium Alloy =.- Nylon 6 (copolymer) Polyacetal - N y l o6n1 0 I Expansion E x p a n s i o nh a s b e e n t h e s u b j e c to f a h o s t o f s p e c i a ls t u d i e s .F o r s o l i d b o d i e st l r , ' cocfficicnt of linear expansion is defined as - Leao e: ,411U..40,,,) w l r c r c , , l / : c I l r r r g ci r r l c r r g t [ ( c x l l i r r r s i o r r ) t l rlrot .l r l ( , n l l ) c r i l t r t rrci s c < t 1t | 0 , , , . / - t h c l c n g t l rr l l t l r c c t l r r r p o n c nl ln r ( l c tr ' o t t s i r l c n r t i o rnr.r r t l . 1 1 / , ,- ,- n l c i r nt e n ' l l ) c r l r t r r r r . ' t l i l l r ' t cl no cwc l t i c l rt l r c l r o t l vi s s t r b j c c t c t l . 'l'ltc r ' o r ' l l i r ' i r ' r tot l l i t t c l r tc x l l t t t r i o s tt l c l t n c r l l t c c x p ; t t t s i o tot t l r s o l i r ll r l o r r Fr lr r r t ' D l V t ' i r ttt' o r ' l l i c i c rot ll l i n c l r c x p t r n s i olror r v l r ri o r r sr r r i r t ci ar l s l( a ) n r c t a l l i c 230 6 Embodimentti, steel, or of grey cast iron with bronze or aluminium, great care must be takcrr allow for relative expansions because of the sisnificant differences in r coefficients of thermal expansion. with larse dimensions. even the relatrrr. small differencesbetween, say, 35C carbon steel and 13 per cent chromium sr (10C11,3%Cr) can cause serious problems. Metals with a low melting point, such as aluminium and magnesium. greater coefficients of thermal expansion than metals with a high melting such as tungsten, molybdenum and chromium. Nickel alloys have different coefficients depending on their nickel conl(.nt, Very low values occur in the range of 32-40 per cent by weight, with .l()';r Ni-64% Fe (known as 'Invar') having the lowest coefficient. Synthetic materials have significantly higher coefficients of expansion metals. lines for embodiment design zJl re 6.63a shows a body clamped at one point with no degreesof freedom. al expansion it can expand freely from this point along the various Figure 6.63b shows a plate that can be rotated about the z axis and thus has ee of freedom. As shown in Figure 6.63c, this singledegree of freedom simply removed by means of a slide. were this plate to expand under temperature increases,it would have to rotate about the z axis. for the -) of treedom '/,/ Delees =0 r =0 . R 2 Expansion of components To calculate changes in length, /1, the designer must know the temperatrrrc distribution (position and time) in the component and hence the mean tempt'rr. ture change with respect to the initial value. If the temperature distribution does not change with time, we speak oi tg steady or fixed expansion. If the temperature distribution changeswith time. rvc speak of an unsteady or fluctuating expanslon. In the case of steady expansion, the physical ouantities on which tl expansion of the components depends are obtained from the basic equations: /l: a. l' /0n /o-:t7l'nour*, o* The change in length /l which concerns the designer is therefore dependent orr! - t h e c o e f f i c i e n to f l i n e a r e x p a n s i o nc : -the length / of the component; and -the mean temperature change A0^over this length, and can be determined accordinslv. The value thus determined has a direct bearing on the desisn: evt component must be clearly located and must only have as many degrees t freedom as are necessaryfor its proper functioning. In general, a point is fixcd and the requisite translational and rotational movements are set by approprirrrc guides, for example slides, bearings etc. A body in space (a satelritc or helicopter) has three translational degreesof freedom in the x, y and z directiorrr and three rotational degrees of freedom about the x, y and z axes. A sliclirrg p i v o t ( f o r e x a m p l e t h e n o n - l o c a t i n gb c a r i n go f a s h a f t ) p r o v i d e st w o d e g r c c s. [ f r c c d o m - o n e t r a n s l a t i o n a la n c l o n c r o l a t i o n i t l , A b o d y c l a m p c c la t o n c p o i r r l ( l i r r c x a n t p l c a b u i l t - i n h c a r n ) , o n t h c ( ) t h c r h u n d , h u s n o c l c g r c c so l ' l ' r c c c k l r r , l.ityouts lritscd tln thcsc coltsidctltaont rlonc do nol. howcvcr. allow lor cxpulrsiorrnutonrlticirlly. ils wc shttll now domonrtrute, S = .90.Z-9m Degrees of lreedom I = 0 ,R = 0 "A L"= r", "u cd 6.63 Expansiondueto steadyuniformtemperature distribution;continuous line: state;brokenline: highertemperature state; body attachedto a fixed poinl plate can rotate about the z axis; that is. one degree offreedom plate tn (b) (D) but Dut with wlth single slngledegreeof freedom removed by an additional sliding pivot Plate as in plate as in (b) but allowing for expansionwithout rotation. it would also be possible le simple slideswhich might equally well be arrangedalong the x axis as along a line the z axis inchned attan E = lyll, does not lie in the direction of the expansion that results from the change of h in the x and y directions..If the slide allowed only translational movement did not also act as a pivot, then jamming would occur. By fitting the slide in direction of one of the co-odinates(Figure 6.63d) it is possibleto avoid the i < l no f t h e c o m p o n e n t . fter deformationdue to thermalexpansiongeometricsimilaritywill only be ttaincd if the followingconditionsare met: Thc c<lcfficicnt of cxpansiono must be constantthroughoutthe component (isotrophy),whichcanbe takenfor grantedin practiceprovidedthat ontyone kirul of mrtcrial is uscd and that thc tcmpcraturcdiffcrcnccsare not too 8rcilt, 232 6 Embodiment tlLrr lines for embodiment design - T h e t h e r m a l s t r a i n s e a l o n g t h e xy,, z a x e s m u s t b e s u c h t h a t € * :€ y: e z : a . , l [6.131]. If a is constant throughout a component, then the mean temperature incrcr must be the same for all three axes, so that we have: symmetry line of the deformed state should, in the first instance. be tt both along the symmetry line of the component and also along that of the mposedtemperaturefield. However,asFigure6.646shows,tf,issymmetmay not be easilyidentifiablefrom the componentshapeand temperature bution, so that the ultimate state of deformation must also be taken into nt. Ihat state, as we said earlier, may also be causedby external loads. To extent' our remarks also apply to guides of components subject to large /l*: 1*'a'/0^ llr:lr.a./0^ Zlt: lr'a' /0^ nical deformations.An examplewill be found in 16.12l. following examplesmay serveas further illustrations: and for the x and y axes: al. /l* t a n c '0 : J : J 1., l* The component must not be subjected to additional thermal loads, which rirll not happen if, for instance, it is completely surrounded by a source of hr.:rl [6.131]. As a rule, however, different temperaturesare measured in a single comp.. nent. Even in the simplest case, with the temperature distribution changing linearly along the x axis (Figure 6.64a), a change in angle is produced whicli. Figure 6.64. Expansion under non-uniform temperature distribution, here decreasing linearly along the x axis: (a) Plate correspondingto Figure 6.63d; non-uniform temperaturedistribution product deformation shown by broken line; sliding pivot required (b) Guide placed on symmetry line of deformed state so that a simple slide can be userl again, can only be taken up by a guide with a sliding as well as a pivoti movement. A simple slide, which allows translational movement with r degree of freedom, can only be used if the guide lies along the line of symmct of the deformation (Figure 6.64b). If this conditionis not fulfilled, a furtherdegreeof freedommust be allowc{. Hencewe obtainthe rule that guidesthat take up thermalexpansionand hirvc one degreeof freedomonly must lie on a line throughthe fixed point, and tlrir line must be the symmetryline of thc dcformcdstate, The deformed state can be causcdby loud-depcndcntancl tempcraturc. depcndcntstresses,in aclditionto thc exptndon itsclf, Sincc thc strcsslntl tcmpcraturcdistributirtnalso dcpcndr on thc rhape of thc conrponcnt.tlc 233 gure 6-65 is the plan view of a device whose temperature decreases from the re to the periphery. It is supported on four feet. In Figure 6.65a one of the waschosenasthe fixed point. If the deviceis not to ,oLt. o, iam, the suide re 6.65. Plan view of a devrce. temperature decreasesfrom rcentre to the periphery, mounted four feet: Designatedfixed point on one ; simple slide along a line that is the symmetry line of the rature field maginary fixed point in the centre device formed by the ion of the lines of expansron lmaginary point fixed b only be placed along the symmetry line of the temperature field, that is on opposite foot. Figure 6.65b shows a method of providing guides along netry lines, without a designated fixed point. The intersection of the linei gh the guides constitutes an imaginary fixed point from which the device expand evenly in all directions.In that case,two guides,for example I and2, ld be omitted. gure 6.66 shows the location of inner casings in outer casings when a mon centre must be maintained as, for instance,in turbines. If the deformed pe of these components is not completely rotationally symmetrical, then the les must be placed on the symmetry lines to prevent jamming of guides -Such due say' oval deformation of the casings,(see Figure 6.66:D. oval rmation is causedby temperaturedifferences,especialryduring the warmphase.The imaginaryfixed point lies on the longitudinalaxisof tie casingor ft. Figure 6.67 shows an austenitic steel high-temperature steam inlet pipe a tich must be fitted into a ferritic steel outir casing b while protruding into a tic steclinnercasingc. Because of markeddifferences in thL two coeificients expansion and also because of the considerable temperature differences wccn thc componcnts; particular attention must be puia to relative expan- 1n. An imaginaryfixcd point is provided by the rotationallysymmetrical c s . i l n i l r r i l n g c m c n t c n s u r i n g t h e u n i m p c c l c de x p a n s i o no f t h e a u s t e n i t i c p o n c n l t k l n g u n y l i n c t h r o u g h t h c i n r i r g i n u r yf i x c d p o i r r t . B c c a u s c thc aa i L JL+ 6 Embodimentclcsrlrr Guidelines for embodiment design 235 the fixed point of the inner casing are not identical and no definite rature distributionscan be assignedto the components.The double e of freedom is obtained with the help of the piston-ring seal e which ts the independent axial and radial movements of the inlet prpe. lative expansion of components far, we have been considering expansionin a relatively stable environment. ry often, however, the relative expansion of two (or more) components has to taken into account, especiallyin the caseof mutual loadingsor when certain ces must be maintained. If in addition the temperature varies with time, )n the designeris faced with a very difficult problem. The relative expansion the two componentsis: 00 F i g u r e6 . 6 6 . L o c a t i o no f i n n e r c a s i n g si n o u t e r c a s i n g s : (a) Arrangement of guidesdoes not allow for expansion;oval deformation of the h o u s i n g sc a n c a u s eg u i d e st o j u m (b) Arrangement allowing for expansion;guideslie along symmetry lines; no jamminr with oval deformation \i/ __-u-/l\ d r . t : G 1. 1 1 '/ 0 ^ 1 1 1 1 - a z . l z . / 7 n z ( t ) . te relative expansion relative mean temperature difference does not vary with time, and if the :ients of linear expansion are identical, then all that has to be done to ise the relative expansion is to even out the temperature or else to select rials with different coefficients of expansion. often both are necessary. s can be seen in the case of a flanged connection consisting of a steel stud an aluminium flange (Figure 6.68). Because the aluminium has a higher icient of expansion, a temperature rise will increasethe load on the siud, :h may lead to failure. This can be prevented, on the one hand, by increasing length of the stud and using a sleeve and, on the other hand, by usine Figure 6 67 Inlet pipe a of a steam turbine made of austeniticsteel that takes the stearn through the ferritic steelouter casingb to the inner casingc. Expansionplanesthrough g u i d e w a y sr / d e t e r m i n ea n i m a g i n a r yf i x e d p o i n t . P i s t o nr i n g s e a l sa t c p e r m i t t h e a x i a l a n c lr a c l i i rcl x p a n s i o no f t h e e n d o f t h e i n l et p i p c ( B B C ) tclnpcrilturc distribution at that pointis l'airlyunilitrm,thc rcspcctivc raclialirrrrl producca rcsultingcxp[n$i()n axiillcxpilnsions ulongthc indicatcdlincs. lly crlntrasi,tlrc inscrtiolrol'tltc inlet pipe into thc inncr casilrgnlustilll()w indcpcndcrtlcxplrtsionillon!llw() uxci. bccttnc thc fixcd point ol'tlrc irrlct;ri;x' l r c 6 . ( r l l . . ( ' o r r n c c t r obny n r e a n so f a s t e e ls t u d a n d a l u m i n i u mf l a n g e [6.147]: ttlud crtdirrt.gcrcrl bccauscalurrriniunrflaugc has gre:rterexpansion i n c r l r p o r l t t i o ttlt l l n v i t t c x 1 ' r i t n s i osnl c c v cw i t h l r c ' o c l f i c i c r irltf c x l r u p s i e nc l o s ct o ( . 1 ! i l o b i r l i r r r c tch c r c l l t i v c c x p t r n s i o no l t l t r r r g ct r r r tsl l r r t l 6 Embodiment dcsrlrr components with appropriate coefficientsof expansion.If relative expansion r\ to be avoided altogether, then we must have: dr"r: 0 : ar' lt'/0^t-Gz'lz'/0-zWith /1 : lz* lz and,t: 1: everywhere display the same temperature-depenclentproperties. This can be remedied if the expansions are kept under control by the carefully ned introduction of appropriate cooling or heating. a3'\'/0^3 l2l\the relative lengthof sleeveto flange becomes: a3'/06- Guidelines lor embodiment design c\'A0^1 c1'/0^1- a2'/0^2 s t et ' l With steady-state expansion, /0^t: /0^2: /06 and with ( a r : 1 1 x 1 0 - 6 ) ,I n v a r ( a 2 : 1 x 1 0 - 6 ) a n d a l u m i n i u ma l l o y ( a z : 2 0 x 1 0 6 ) r r s t h e c h o s e nm a t e r i a l s( a s i n F i g u r e 6 . 6 8 b ) ,w e h a v e 7 : l z l l t = 0 . 9 . The designer will be familiar with the complicated expansion problerrrr associatedwith the pistons of internal combustionengines.Here, the tempcrrr ture distribution over and along the piston differs even in the near-steadystrrtr' and, what is more, differences in the coefficientsof expansion of piston urrtl cylinder must also be taken into account. One solution is the use of iul aluminium-silicon alloy with a relatively smallcoefficient of expansion (smallt'r than 20 x 10-6), of expansion-inhibitinginsertsthat are also good heat conclrrr'. tors, and of a flexible piston skirt. The bimetal effect provided by steel inscrts also helps to match the shape of the piston skirt to that of the cylinder [6.12t'l: Fisure 6.69. teady relativ e expansion the temperature changes with time, for instance during heating or cooling )cesses,we often find a relative expansion much greater than that which is in the steady,final state. This is becausethe temperatureof the individual ponents can differ considerably. In the common case of components of rl length and equal coefficients of expansion, we have: CIl: (12: C and \: lr:1 d . e r: G ' I ( 2 0 ^ 1 ( t ) - / 0 ^ x r ) . The_heating of components has been examined by, among others, Endres and r l m [ 6 . 5 9 , 6 . 1 8 0 ] .N o m a t t e r w h e t h e r w e a s s u m ea s t e p o r l i n e a r t e m p e r a t u r e nge in the heating medium, the heating curve will be characterisedby a time stant. If, for instance, we consider the temperature change /0^ of a ponent dunng a sudden temperature increase/0* of the heatins medium. , u n d e r t h e a d m i t t e d l y a p p r o x i m a t ea s s u m p t i o nr h a t t h e s u r f a c ea n d m e a n peraturesof the components are equal-which, in practice, is approximately only for relatively thin walls and high thermal conductivity-we obtain the shownin Figure6.70, with: Z0^: /0* (l - e-ttr) / is the time and Z is the time constant such that: Figure 6.69. Piston of internal combustion engine made of an aluminium-silicon alloy with steel insertswhich inhibit circumferential expansion ; moreover the bimetal effect ensures optimum adaptation of the piston skirt to the cylinder (Mahle design)from [6.126] If, on the other hand,the choiceof materials is restrictedin practice,then thc designermust rely on temperatureadjustments. In high-powergenerators.l()r instance,large lengthsof insulatedcopperrod must be embeddedin the steeI purposes alonethe absolute lnu\l rotors.For insulation and relativeexpansions possible. only small Here thc solution keep the tempcralrrrc trc kcpt as as is to lcvcl to a minirnumlry cooling[(r.ll6. 6.23.51, Morcovcr,if thcscfast-rurtrtirtg thcrmll imbuluncclt rotorshilvc largcdimcnsions, may ()ccurtcvcn thoughthe tcrrrpcrilturc distributionis rclirtivclyunaform, lhc rtltttr.bcclusc<ll'itsctlrttpli rnrrlcrisltthrt cirlcdstruclurciurdlhc virriotrs havegoncirrtoil. nrirynrlt irlwlrrr f :c'm h.A re: specific heat of the component; mass of the component; heat transfer coefficient of the heated surface of the component: and heated area of the component. te the simplification involved, this approach may be considered funtal. ith two componentshaving different time constants,we obtain temperature 9s that, at a given critical time, will have a maximum difference. At this we have maximum relative expansion, and must provide clearancesto thc expansionor run the risk of excessivestressesbeyond the yield point. l idcntical tcmperature curves appear if the time constants of the two t n c n t sc a n b c c q u a l i s e d .I n t h a t c a s e ,t h e r e i s n o r e l a t i v ee x p a n s i o n .T h i s i v c c i r n n o t a l w a y s b c a c h i e v e d ,b u t i n o r d e r t o r e n d e r t h e t i m e c o n s t a n r s l x i n r i r t c l yc q u i r l . t h a t i s , t r l r c c l u c ct h c r c l i r t i v cc x p a n s i o n ,t h c d e s i g n e rc a n , thc rclirtiolrship: 6 Emhodiment desilrr 238 l.1e Guidelinesfor embodiment desisn v - n r zl - r A 2n.rl 7 Z9^tlnox lZ.Sn,- ,J +fA --a (heated tromoutside)t2' -____l/ Holrow shal Horowshar 6-A(heated frominside) \f__, Prare constantsof a step Fisure 6.70. The effects on two components with different time te"mperaturechange, A8*, in the heating medium vl T : c ' o '-- ' - . Ah _/ ''L-' \-{-a], (heated on oneside) o *:>_] fi:|iro.rborhsides) S \ where V: volume of the component; e : density of the component, either adapt the ratio of the volume V to the heated surface area A; or else adjust the heat transfer coefficient h by means of , say, lagging. Figure 6.71 gives the relationship VIA for a number of simple but representrttive bodies. - Brade v tbt, ' A= t4 =' ,ffi r{ tr' V_ t b t - t A 2tb 2 !_ A,t_A" A CI C re 6 .7 | . Volume-surface area relationship of various geometrical bodies, arrows to heated surfaces As a result, the clearance between the spindle and the sleeve will diminislt dangerously. fn a, tne sleevesare sealed axially but can expand freely radially. Moreovct, their volume to area ratio is such that spindle and sleeveshave approximatclV equal time constants.As a result, the clearanceremains more or less uniform lt ali temperatures and can therefore be kept small. The surface of the valvc spindleind the inner surfaceof the sleevesare heated by steam leaks, so that rvc have: (VlA)rrin6": rl2 (VlA),t".u": (ruz- rl)12r,; . c hilvc w i t h r ; : r a n c l ( 7 / , 4 ) , n ; , ' , 1: 1 "( V l A ) , t e c \ cw rl2: (rri - r2)l2r /n- r' V2 00 rc 6 72. Spindlc sealsof steam valves; l i x c t l s l c c v cr c r l r r i r c sr c l i r t i v c l yl a r g cs p i n d l cc l c i r l i r r . r cbce c a u s ei t h a s n o t b e e n I t l t o i r l k r wl o r c x p r r n s i o n r : r r l i r r l l lvr c c l r r r t lr r x i l r l l vs c l r l c rsl l c c v cP c l r t r i t ss l t t i r l sl l t i t t r l l c l c i r n r n c cb c c i r t r s sc p i n t l l c r l c c r c l t ; r r cl r c c r tr l c r i r r t c t ll o l t i r v ct l r e s l r r r r r . ' t t i lcco l l s l i l l t l 6 Embodimentdesirrr Figure 6.73 showsvarious steam turbine housings.With appropriate design it is possible to adapt the volume to area ratio of the housing and also the hcrrt transfer coefficient and size of the heated surface to the time constants of thc shaft, and thus keep the blade clearancesapproximately constantwhen startirrl (heating) the turbine. 247 Guidelines for embodiment design ence a gradual plastic deformation that, after a given period, may lead to re. The ensuing time-dependentfracture stressis much lower than the 0.2 cent proof stress at the same temperature determined by short-term riments (see Figure 6.74). Critical temperature and creep strength depend ly on the materials used and must both be taken into consideration.with s, the criticaltemperaturelies between300' and 400"C. casing Single N (partial) blade Separate statronary carrier mm2 I Double casing \- II (full) bladecarriet Separate stationary t\ h Figure6.73. Steamturbine housingswith differenttime constants fino @ There are several well known methods for reducing the heat translt'r coefficient of a component (for example by insulation) and thus for slowing down the heating and reducing the relative expansion. The ideas we have just put forward are applicable wherever temperaturcs change with time, and particularly wherever relative expansion goes hand rn hand with clearance reductions that are likely to endanger the functioning ol' turbines, piston engines and machines operating in hot environments. Heat treatment t h 930'C/0il + 2n730clan I 200 \ \r og I \\ I I \\ o62(105) 1 I Critical temperature Ot 6.5.3 Designing to allow for creep and relaxation I Behaviour of materials subject to temperature changes When designingcomponents subject to temperature changeswe must take into account not only the expansion effect, but also the creep properties of tlrc materials. The temperaturesinvolved need not necessarilybe very high, thoLrglt they usually are. However, there are some materials that will, even at temperir. tures well below 100"C, behave in much the same way as metals do at very high temperatures. Beelich [6.9] has examined this subject at some length and in what follows u'c shall base ourselveslargely on his findings. Materials in common use, pure metals no less than alloys, have a polvcrystalline structure and a temperature-dependentbehaviour. Below a crilicul t e m p e r a t u r et,h e s t a b i l i t yo f t h e i n t c r - c r y s t a l l i n cb o n d s i s l a r g e l yi n d e p e n d c n lo l t i m c , a n d t h c y i c l d p o i n t c a n b c u s c c lt o c l c t c r m i n ct h c s t r c n g t ho f c < l r n p o n c r r l s . C o m p o n c n t s a t t c m p c r a t u r c sa t t o v c t h c c r i t i c t t lt c m p c r i l t u r c a r c s t r u n g l y i r t l ' l r t c n c c d b y t h c t i n r c - d c p c n d c r r t r c h i r v i o u ro f t h c m a t c r i n l , I n t h i s t c n r p c r a t r r r c r i l n g , c , r r r a t c r i i r l sw i l l . u r r t l c r t l r c i r t l l u c n c c t t f k l u d . l c n l p c r i l t u r c i u r d t i r r r e , 0c 400 Temperature (1osl 600 ' re 6 74. Characteristic values determined by high-temperature tensile strength and :p experimentswith 21Cll 5% Cr-Mo-V steel at various temperatureslcritical perature is the intersectionofthe curvesof0.27a proofstress and the stressfor 0.27c strainin 105hours when working with synthetic materials, the designer must allow for their asticbehavioureven at temperatures below 100"C. In general, the modulus of elasticity changesinversely with the temperature re 6.75a). The smallest changesoccur with nickel alloys As the modulus of elasticity drops, so does the stiffnessof the components, of synthetic components in particular (see Figure 6.75b).In their case,the igner must know the temperatureat which the modulusof elasticitydrops nly to rclativelylow values Crcep p o t l c n l st l r i t l i t r c P u t u n ( l c r l t l a d sl i r r k r t t gp c r i r x l sa t h i g h t c n r p c r l t u r c sw i l l , a Aa 6 Embodimentdesign 6.5 Guidclines lbr embodiment design in addition to the strain given by Hooke's Law (e : ol E), also experienceplastic with time. This property of materials, which is known as deformation (eo1u,,) 'creep', depends on stress,the effective temperature 6 and time. We say a material creeps if the strain of the components increasesunder constant stress[6.9]. The creep curvesof various materials are well known [6.64. t.rl -l al "'l 6.eol. I 2.10s N mm2 L! 1t q at1 #:ffi81'ffi 10( N mm2 l0l N ' I c-rt*r2 Ir, . , lr'rirr* -T- t l-.-rnermosets (ol* bl\ N p ; i,lilf ,, i b_"t \ . .^q Ilu- E E = =r = 0 i-Thermoplasticr \ 100 200 300 0c 500 TemPerature t -... 0 l n ' 3 LD-polyethvlene tl',. 4 Epoxy resin(pur 5 4 + g l a s s r e i cnnf og \ I 100 -100 0 \ /irl o\1/ 100 2000c 300 Temperatu re Primary Secondary Tertiary creep creepregroncreepregion region (Accelerated (Transitional (Steadycreep) creep) creep) Figure6.75 Relationship of modulusof elasticity to temperature, of variousmaterials ( a ) m e t a l s( b ) s y n t h e t im c atcrials r e . 6 . 7 . 6S. t r a i n( a ) a n d c r e e pr a r e w r t n d u r a t t o no f l o a d ( s c h e m a t i c ntatron) ; characteristics of varlous creep phases Creep at room temperature '.t':"]sanrtevenausteniticsteelsshowvarying degreesof ,:t:?^::'.1:-r_1*lt on,length b Before we can design componentsloaded to near the yield stress,we must kno\\' how they react in the transition region between the elastic and the plastic statcs [6.90]. With persistent static loads in this transition region, we can expect primary creep in metals even at room temperature (see Figure 6.76). Thc resulting plastic deformations are small and merely affect the dimensional stability of a particular component. In general. steels show little creep when s u b j e c t t o s t r e s s< 0 . 7 5 ' q y 2 o r < 0 . 5 5 ' o F , w h e r e a s , i n t h e c a s e o f s y n t h e t i c materials, a reliable assessmentof the mechanicalbehaviour can only be madc by consideration of the temperature and time-dependent characteristics. Creep below the critical temperature Previous studies 16.90,6.971of metals have shown that the customary calculir t i o n s , b a s e d o n h i g h - t e m p e r a t u r ey i e l d s t r e n g t h a s t h e m a x i m u m p e r m i s s i b l t ' s t r c s sf o r s h c l r t - t e r ml o a d s .a c l d i t i o n atl h c r m a l l < l a c las n c ll o a d v a r i a t i o n s ,s u f f i c t t u pt o t h c c r i t i c a lt c n r p c r a t u r c . W i l h c o t r r p o n c n t tsh a t r l r u s th i r v c l r i g l rt l i r n c n s i o l t i sr lt i r b i l i l y ,h o w c v c r . t h t ' c h t r r l c t c r i s t i cosl l h c r r r i r t c r i ; t l c i c l r r r i r t c hr lv c r c c p c x p c r i n r c r t t rst t r r s ltr l s o l r t ' t i r k er r i r r t ot ' o l r s i t l t ' l r t i o rcrv, c r rr t n t r x l ( ' t n l c l hVi p h l c t t t l ) c l t l t t t ( ' sI ,l t u r l l o v c rrlt t t r l ,1.^l.r11g b of operationandworking,"-;;;;i;. st turai.n*;: #;,,i; try ere vated :*:i:,,1".": T: :,"1,. -.: ?1."i: ruc " may lead tJ u .urr.J t",iil*";;; :l*::-3:l..11a1sform1tions rependence of I properties of themateriurr, :9-"n:"-of-the s [ 6 . 1 0 7 ,6 . 1 3 2 ] . *;i;hlr r.iTtr" .#"'r,l,il above the critical temperature this temperature region, mechanicalloads wilr causedeformations in metals j::"?"f: lT:f::t'll,"..y,..Jd,rt..neir,,inat is thematerials willcreep. rn :ral,thisprocess canbe dividedintoihreeff,ur". 6.g7l(FiCr*;i;6;l 16.g0, changes,*reueginnin d ;f i?" tertiary :. lln:.T: phase must be consider"d dung".our. "1n;:,:*J,,::1"1i,_"_,e l..rlJirr";;"i;t..; ffiru,:i s at approximatelyl per centpermanentstrain.Fig"ure o.zzlir"*, the 10s crcep strengthso1o7n1111s at 500"C of various steels. Rcluxation I t l i t t l c d s y s l c n l s ( s p r i r r g s ,b o l t s , t c n s i t l n w i r c s . shrink l'its), thc ncccssary l i r t l p r o t l u c c si r r t o v c r i r l l s l r i r i n r , ( c k r r r g l r t i o r r , l / ) . l l c c i r u s co l . c r c c p i r n r l f r E , m h o d i n t e ndt e s i g r t 244 245 Guidelinesfor embodiment design 300 N mmZ 250 I 200 I I b tsn g Critical temperature ol ot 100 (cast 17Cl1 5o/"Cr-lVo-V 10 2 t, ? L 6 8102 in N/mm2 Stress 6 8101 correspondingto a lVo permanent strain of various steels after 105 Figure6.77.Stresses hoursat 500'C[6.1461 settling of the material due to plastic flow at the bearing surfaces and split lines. the ratio of plastic to elastic deformation gradually increases.The phenomenort 'relaxation' [6.60, of decreasingelastic strain at constant overall strain is called 6.24t,6.2421. Loaded components are usually preloaded at room temperature. Becausethc modulus of elasticity varies with the temperature (Figure 6.75), the preload decreasesat higher temperatures even without a change in length of the loadecl super-imposed stresses(normal or tangential to the surface). Studies of thc relaxation of bolted flanges [6.60, 6.241, 6.242) have shown that plastic deformation also occurs at the split lines and bearing surfaces(settlement) and irt the threads (creep and settlement). To sum up, we can say that, with metallic components: - The drop in preload depends on the relative stiffness of the parts loadetl againsteach other. The more rigid the connection, the greater the drop in thc preload due to plastic deformation (creep and settlement)' - Although settlement can be appreciably offset during the tightening of boltctl flanges or the assemblingof shrink fits, the designershould, where possiblc, provide for few but accurately machined surfaces (split lines, bearipg surfaces). - T h e r e i s a t e m p e r a t u r el i m i t b c y o n d w h i c h t h e m $ t e r i o lc a n n o t b e p r o p c r l r ' u s c d ( F i g u r c 6 . 7 f t ) , I n a d c l i t i o n ,t h c d c s i g n e rt h o u l d t l w t t y s c h o o s c m a t c r i i t l r M12steelbolts/nuts 0 o-.-o 45C/10C .---{ 34C11"/" Crl3SC * -----+ 42C I 1okCr-Mo/35C b x-x C/l 5o/"Cr-lvo-V 21C 1 5"/" Cr-Mo-V/25Cl1oio Cr- 200 250 300 350 400 450 500 Temperature 5500c 600 : 6.78. Remaining pre-load stressoo,"(103)after 1000h in bolted joints at the iling temperature. Preload : 0.2o/oinitial strain [6.242] in which the appropriate yield point is not reached even with superimposed operational stresses. In the short term, high initial pre-loads (initial clamping forces) give rise ro higher residual clamping forces. ln the longer term, the residual clamping forces become relatively independent of the initial preload (Figure 6.79). Joints that have already undergone relaxation can be tightened up if the toughnessof the material permits. As a rule, creep of about 1 per cent, which leads to the tertiary creep region, must not be exceeded. If joints are subjected to an alternating load in addition to the static preload (Figure 6.80), then, as experiments have shown, the amplitudes tolerated during relaxation-dependentdecreasesin the mean stress are considerably greater than those tolerated at constant mean stress.However, relaxationdependent decreasesin the mean stresswill often lead to a loosening of the hen using bolted joints made of synthetic materials, the designer tries to advantase of their small electrical and thermal conductivities, their ance to corrosion, their hieh mechanical damping, their small specific ts etc. In addition, such joints must, of course, have the appropriate h and toushness. Specialattentionmust also be paid to preloaddecay, lest the functioning of jointsbe seriouslyimpaired. S p e c i asl t u d i e s[ 6 . 1 3 6 ,6 . 7 3 7 ,6 .2311have shown that in svnthetic.unlike l l i c .m a t c r i a l s : the prcltlad rcmainingaftcr a givcn timc ilnd at room tcmperaturcis dclcrmincdhy thc nratcrillitsclfand its tcndcncyto ubsorhmoisturc:and 246 6 5 Guidelines for embodirnent design 6 E t n b o c l i r n e n td e s i g n 110 N mm7 120 M l 2B o l3t 4 C l 1 C " kr M 1 2N u 3t 5 C Testtemperature 450'C Relaxation A static +o. N/mm2 dynamic o35 .90 . 100 1t? lm!lo F^ "' s TI tAe -Fu "AB = E 6 Fp.lt9 99^ltrengthand^durability of boltedconnections arexperimental remperature of 450'C[6.2a|l;sizeMt2; boft34alt%Cr sreel;nut35Csteel;'staiil stress 3.5-5 fie'load N/mm2; (a) creep experiment (b) alternatingfatigue (wohler) experiment (c) fatigue relaxation expenment (seeFigure 6.79) 0 No of loadapplicatjons iV* 10-2 101 _ l l f T l 0I - 100 10, h-----? Figure 6'79. Effect of different levelsof initial joint pre-load on the residualctamping force with time, for both static and alternating(fatigue) loading - continual changesin the absorption and releaseof moisture have a particular-ly deleterious effect. 4 Design features In order to increasethe potential life of componentssubject to long-term loacls. the designer must familiarise himself with the behaviour with time of the material involved. According to [6.90], it is dangerousto use short-term valuc\ to predict Ioad responsesfor periods of l0s hours or longer. It is impossible to avoid thermal stressesin all componentsby specifying tht. use of highly alloyed materials. Appropriate clcsign features are often m()rc u s e f u l t h a n c h a n g e si n t h e m a t e r i a l . T h c c l e s i g nm u s t b e s u c h a s t o k c c p c r c c l l w i t h i r rp c l n r i s s i b l cl i n r i t s .w S i c h c i r r r bc clonc ['ly rncansof: - : t h i g h c l i t s l i cs t r i t i l tr c s c r v c ,w h i c l r l t c l p r l o k e c p d o w n i r t l t l i t i o n l rflt 1 r t l st l u c t o l c n l l ) c r i l t r r r lc' l r r c t r r i r t i <(tsncsc l r i g r r r c6 t { l } ; Figure6.81.Austenitic-ferritic steelflaneed joint for operatingremperatures of 000"i [ 6 . le e l - insulatit'r' or cooling of the component as in steam turbines and in gas t u r l ' r i n c s( s c c F i g u r e 6 . 8 2 ) : - t h c i r v . i d i r n c co f m a s sc o n c e n t r a t i o n s w h i c h , i n u n s t e a d yp r o c e s s e sm , ay lead t o i l r c r c i r s c rt!h c r n r a l l < l a c l i n ga; n c l -' tltc l ) r c v c l l t i ( ) no l c t ' c c P i t t t t t t w i r n t c dr l i r c c l i o l r sw l r i c h c a n c i r r r s cl u n c l i < l n a l 6 Embodiment desigrr 248 5 Guidelinesfor embodiment desisn 5 In-other words, the part which is moved during dismantlingshould not pr..iccl ially beyond the fixed part [6.154]. Designing against corrosion damage often happens that corrosion can only be reduced, not completely avoided; )reover the use of corrosion-proof materials may not be economical.It follows Figure 6.82. Double casing steam turbine with shrink rings that hold the inner casing together. Relaxation of the shrink rings is reduced by cooling with exhaust steam. As the machine increasesits output, the shrink rings exert an increasing pressure thanks to growing temperature differences between the steam inlet and outlet. The shrink rings are Jeated on heat-inhibiting segments which, with the help of shims, permit the original shrink fit to be restored after relaxation (BBC design) failure (for instance the jamming of valve spindles) or dismantling problems ( s e eF i g u r e 6 . 8 3 ) . In Figure 6.83a the material of the cover creeps into the relief groove. Thc cover, which heats up more quickly, pressesagainst the centering surface an(l also creeps at point y. The cover shown in Figure 6.83b is a better design since, despite the creep, it can be dismantled easily. In addition the cover has been made hollow so that it cannot exert a significant radial force on the centering surface. Causes and effects of corrosion 'hile the formation of metal oxide layers in dry environments and at higher mperaturestends to increasechemicalresistanceto corrosion, relativelv weak trolytes are formed in conditions below the dew point and these generally to electrochemicalcorrosion [6.r9i). corrosion ii also fostered bv the fact at different components have contacting surfaces with different properties, for stance due to the inclusion of various noble or base metals, to differencesin talline structure, and to residual stressesset up, for instance, by heat tment and welding. In addition, wherever the design calls for slits or holes. ere appear local ditt.erencesin electrolyte concentrationeven in the absence of rences in electric potential resulting from the use of different materials. Accordingto [6.89,6.r93,6.r94,6.20g,6.212]we mustdistinguish between iform and locallyconcentrated corrosion.The latter has a variety of causes d effects, so that we must further distinguish between corrosion in crevices, ntact (bimetallic) corrosion, transition zone corrosion corrosion [6.ri7),'erosion trgue, stress corroslon cracking and corrosion accompanying and vitation. The preventive measures depend on the respective causesand effects. various a m p l e sa r e g i v e n i n 6 . 5 . 4 . 4 . force Sealing Uniform corrosion presenceof moisture (weak electrolytes) combined with oxygen from the e n t , particularly below the dew point. F i g u r e6 . 8 3 .C e n t e r i n ga n d s e a l i n go f a c o v c r l 6 l 5 4 l l lhc rclicf g,rtxlvcatrd at.l' (af Dismantling is impeclcclbccauscthc nrltcriitl c'rccgts,into ftlrccs.('rccp docs with tmtllcr clttmping scitl t)cttcr provirlcs il ccigc r",,li'rg Cunuc* it.rj tlclign irrrprrlvctl ttl not inrpcdc tlisnrlntling tltlrtks rtensivc, uniform corrosion of the surface;in steel, for instance,approximatelv I m m . p c r a n n u m i n a n o r m a l a t m o s p h e r c . S < l m e t i m c sm o r e ' p . o n o u n . . o :rlly, cspccilllyin zorrcskcpt l'rcclucntly below thc dcw point arrclhcncc 250 6 Embodiment desigl subject to moisture concentration. Uniform corrosion is fostered by greater activity of the medium, higher flow velocity, and intensive heat transmission. Remedies - Provision of uniform service life by means of appropriate wall thicknessesancl materials. - Design based on a concept that obviates corrosion or makes it economicallr acceptable (see Example 1 below). - The use of small and smooth surfaces involving geometrical shapes with rr maximum volume to surface area ratio (see Example 2 below). -The avoidance of moisture traps (Figure 6.84). - The avoidance of local temperature differences by good insulation antl prevention of hot or cold bridges (see Example 3 belbw). -The avoidance of flow rates greater than 2mls. - The avoidance of areas of high and differing thermal loads on heaterl surfaces. - The application of a protective coating, possibly in conjunction with cathodit protection. Guidelinesfor embodirnent design 251 Local corrosion I corrosion is particularly harmful because, unlike uniform corrosion. it ces very hrgh stress concentrations and also because it cannot alwavs be ticipated. For that reason, the zonesin question require specialattention right m the start. roston tn crevrces ost often, the accumulation of acidic electrolytes(moisture, aqueous dium), followingthe hydrolysisof corrosionproductsin crevicesetc. In rust_ acid-proof steels, there is a breakdownof passivitydue to depletionof ygen in the crevice. ffects ased corrosion in hidden areas.Increasedstressconcentration in areasthat ln any case, under greater stress.Danger of fracture or separtion , without or warnlng. edies wr0ng righl wr0ng right The provision of smooth, crevice-freesurfacesand connections. The provision of weld seams without permanent crevices; the use of butt seamsor through-welded fillet seams(Figure 6.g5). The sealing of crevices, for instance by providing protruding parts with moisture-proof sleevesor coatings. wr0ng IL--_-j L right tr-tr Drainhole c Figure 6.t14.Drainage of componentssusccptiltlct<lcttrrtlsion; (a) clcsignof l'klrlrscncouragingantl irrtllctlittgcorrttrittn ( l l ) w r o n g a r r dr i g h t i r r r i l t t g c n l c t totl s t e c ls c c l i o n r ( c ) b r l c k c t s r u i r d co l c h i t t t n c sl c c t i o t tw i t h t l l n i n h o l e r c 6 . 1 i 5E, x a m p l eosf w e l d e di o i n t s ; , s u s c c p t i b l cl o c o r r < l s i o ni n c r c v i c e s r) c o r r c c tt l c s i g na c c o r d i n gt o [ ( r .I 9 7 1 ) u c v i c e - l l c c w c l t l i r l gt l f ' p i p c s ;i t l s oi n t p r o v c sr c s r s t i u l c tco s t r c s sc o r r o s r g rcr r l c k i p g 252 - ( r E , m b o d i r n e n td c s i r l 5 Guidelines for embodiment design t)_t The enlargement of crevices so that throughflow prevents the accumulation ol moisture. Cont act (b imetallic) co r r o sio n Cause The contact of two metals with different potentials in the presence of lrrr efectrolyte 16.196,6.208). Endangered atea valve to venting Effects The baser of the two metals will corrode more rapidly than the nobler round the contact area, and the faster the smaller its surface area. Once again, the stress concentration is increased and corrosion products may be deposited. Suclr deposits have secondaryeffects of various kinds, for instancethe production ol slime, contamination of the medium etc. Remedies - Use combinationsof metals with small potential differencesand hence a smlll contact current. - Prevent aetion of electrolyteson the contact area by providing local insulatiorr between the two metals. - Avoid electrolytesaltogether. - If necessary,resort to planned corrosion by introducing still baser materials irr the form of 'sacrificial anodes'. Transition zone corrosion ,ure6.86. Increasedcorrosion at the transition from the gaseousto the Iiquid state 197]due to concentrationof the m e d i u m i n the region of the water line oi a vertically anged condenser.This can oe re medied by raising the water level rosion fatigue SE ive attacks on a component subjected to mechanical fatigue loadins i a b l y r e d u c e i t s s t r e n g t h .T h e g r e a t e r t h e l o a d i n g .t h e m o r e i n t e n s et h ! rrosion and the shorter the life of the comDonenl Causes Changesof state of the medium or its componentsfrom the liquid to the gaseous phase and vice versa tend to increasethe danger of corrosion of metallic surfaccs in the transition zone. That danger may be increasedfurther by encrustations ure without distortion, as in fatigue failure. Because the corrosion pro, especially in slightly corrosive media, can only be seen under a micro, this type of corrosion is often mistaken for normal fatisue failure. [6.re]1. Effects This type of corrosion is concentrated in the transition zone and is the more pronounced the more sudden the change of state and the more aggressivethe medium [6.177]. Remedies -The gradual input and removal of heat by a heating or cooling element. -The reduction of turbulence, and hence of heat transfer coefficients at tht' i n l e t o f t h e a f f e c t e dm e d i u m , f o r i n s t a n c eb y m c a n s o f g u i c l ep l a t e s . - T h e p r o v i s i o no f c o r r o s i o n - r e s i s t i njga c k c t si r t c r i t i c l l l l o i n t s ( s c c E x a m p l c s i and 4). - T h c a v o i c l a n c oc l ' t r i r n s i t i o nz o n c p r o l l l c r t t sl t y l p p r < t p r i i r l cr l c s i g nI c i r l r r r c s Fiuurc (r.fi(r. The minimisation of alternating mechanicalor thermal stressesand especially avoidance of oscillatory stressesdue to resonancephenomena. The avoidance of stressconcentrations. The provision of compressivestresseson the surface by shotblasting,roller burnishing, nitriding, etc to increasethe working life. The avoidance of contact with corrosive media (electrolytes). The provision of surface coating (for example rubber, baked enamel, hot dip galvanisation,aluminium etc). s,scorrctsio n crac kinp a i n s c n s i t i v cn t a t c r i a l st en d to develop trans- or inter-crystallinecracks if i c t c l r s i l c s t r c s s c sc < l m b i n c w i t h a s p c c i l ' i ct r i g g c r , T h c s c r n a t c r i a l sa r c : 254 desirrr 6 Embodiment unalloyed carbon steels, austenitic steels, brass, magnesium, aluminium allors and titanium alloys. Effects Depending on the medium [6.197), various very fine and rapidly developinr trans- or inter-crystalline cracks appear in the component. Adjacent parts arc not affected. Remedies - The avoidance of sensitive materials, which may not, however, be possible becauseof other requirements. -The substantial reduction or complete avoidance of tensile stresses on the attacked surfaces, for instance by preloading or shgt-blasting. - The reduction of residual tensile stressesby annealing. - The application of cathodic coatings -The avoidance of corrosive influences bv lowerins the concentration antl temperature. Corrosion accompanying erosion, cavitation and abrasion Corrosion may accompany erosion and cavitation, in which casethe breakdown of the material is accelerated.The basicremedy is the avoidanceor reduction ()l' erosion and cavitation by hydrodynamic means or specialdesignfeatures. Only when this is not possible should such hard surface treatments as metal spraying or hard chrome coating be considered. Abrasion spots can appear, for instance,as a result of thermal expansion, or of pipes vibrating againsttheir supportsetc. In either case,the oxidic protection layer on the surfaces of the rubbing parts may become damaged. Exposecl metallic areas have a more negative electrochemical potential than thosc covered with a protective layer. If the fluid medium is an electrolyte, thesc relatively small exposed areas will be broken down electrochemicallyunlessthc protective layer can be regenerated. Remedies - Reduce the vibration of the pipes by reducing the flow velocity inside thenr and/or change the distancesbetween the supports. - Increase the gaps so that no rubbing contact takes place. - Increasethe wall thicknessof the pipes, thus increasingtheir stiffnessand thc tolerable corrosion rate. -Use pipe materials that readily accept protective coatings. I n g e n e r a l ,t h e d e s i g n e rs h o u l d a i m a t e n s u r i n gt h e m a x i m u m a n d u n i f o r m l i l c o f a l l c o m p o n e n t s[ 6 . 1 7 7 ] . I f i t s h o u l d p r o v e c c o n o m i c a l l yi m p o s s i b l et o m c c t t h e s er e q u i r e m e n t sw i t h t h e a p p r o p r i i l t cc h o i c co f m a t c r i a l sa n d l a y o u t , t h c n t h c designcr must providc for thc rcgullr monitoring of all urens and comprtncnts particularly pronc to corrosion. ftlr insluncc hy virual inspcction und rcgulirr 255 Guidelinesfor embodiment design asurementsof the wall thicknesses, directly by mechanical or ultrasonic ods andiorindirectlyby meansof corrosionprobes that can be scrutinised d replacedat regularintervals. Corrosionshouldnever be allowedto proceedto the point where it threatens ty (6.33.{. Finally, the reader is referred back to the principle of the division of tasks 4.2), with the help of which even difficult corrosion problems can be solved. , one component might provide protection against corrosion and provide a , while another provides support or transmits forces. As a result, the bination of high mechanical stresseswith corrosion stressesis avoided. and choice of materials for any one component becomeseasier [6.155]. Examples of designing against corrosion damage ample I ye is used to absorbCO2 from a gaseousmixture under pressure,and the 2-enriched lye is then forced to surrender much of its CO2 by expansion neration). The position of the expansionchamber in a gas-washingplant is rmined by the following factors If the lye were expanded immediately behindthe washingtower (Figure6.8'7 , int A) the pipework to B would have to withstandlowerpressures and would Preferred location forexpansion in C02) C02outlel Gasunder pressure in C0r) Pump Figure 6.87. Influence ofthe point chosen for the expansion of COr-enriched lye on the choice of material for the pipework from A to B ingly allow a saving in wall thickness. However, because of the release of the agressiveness of the lye permeatedwith CO2 bubbleswould increaseto an extent that the cheap-unalloyed pipe steel commonly used would prove and hence have to be replaced with a more expensive rust and material. For that reason,it is far better to keep the CO2-enrichedrye pressureuntil it enters the regeneration tower (point B). desiqner has to choose between two methods of storing compressed gases re 6.t3tt): 3(I cylindricalcontainers,each with a capacity of 50 litres and a wall ) thickncss of 6 mm: 256 6 Embodiment dcsigrr Figure6.88.Influenceofcontainershapeon corrosion16.l77linthecaseof gasesstoredrrr 200bar: (a) in 30 cylinderswith a capacityof 50 litreseach; (b) in a spherewith a capacityof 1.5mr' (b) 1 sphericalcontainer with a capacityof 1 .5 m3 and a wall thicknessof 30 mm. Solution b is less prone to corrosion for two reasons: -The surfaceexposedto corrosive attack is approximately 6.4 m2, and is aborrt five times smaller than it is in a. In other words, less material is lost througlr corrosion to the same depth. - For an anticipatedcorrosion depth of 2 mm in 10 years, the loss of strength irr a is such that the wall of the container must be increasedto a thickness ol 8 mm, while corrosion to a depth of 2 mm in the 30 mm wall of container b is relatively insignificant. The sphericalcontainer is therefore the better design 251 5 Guidelinesfor embodimetrt dcsign re 6.89. Outlet of a ner for superheated and CO2 under original design insulatedoutlet avoiding rt nsatlon ) other corrosion-resistant nts with separate ponents ) ) 1 lnterchangeableand/or wallthickness 2 greater (depending on losses) material 3 moredurable L Gos of the two. Example 3 Figure 6.894 shows the original design of a container holding a mixture ol super-heatedsteam and co2 16.1711.The outlet is not insulated and cooling leads to the formation of a condensate with strong electrolytic properties Corrosion will attack at the transition zone between the condensate and thc gaseswith the result that the outlet may break away. Figure 6.89 shows two solutions: a using insulation and b using separate components made of more durable materials' Example 4 In a heated pipe carrying moist gases,the inlet to the heated area is particularlv prone to corrosion (Figure 6.90a). A lesssudden transition (Figure 6.90b) or art extra protective sleeve (Fig. 6.90c) offer remedies. 6.5.5 Designingto standards I Ob.jcctivcsof standardisatitllr A n c s s c t t t i i tll' c i t t t t r c< l l ' t l t en t c t h t l t l sw c l t i t v cl r c c n d i r c t t s s i n gi s t h c h r c i l k t l < t w trtl l Region threatened bycorrosron ngmeorum 00 re 6.90.Corrosionin a heatedpipe[6 1771 severecorrosion at the inlet due to suddentransition suddentransition avoided protective sleevecoverscritical zone and mitigatessuddentransition x into simple problems. Thus, in the conceptual phase, the overall nction is broken down into simpler sub-functionsto facilitate the searchfor -f'r.rnr;tioncarriers or the use of design catalogues. In the embodiment p h i t s c . t o o . i t i s h c l l l l ' u l t o w o r k s e p a r a t e l yo n i n d i v i d u a l a r e a s o r a s s e m b l i e s b c t i l r c r c c o r r r l t i n i n gt h c r r r i n l < l r r n o v c r i t l l l i r y o u t c l c s i g n .l f w e e x a m i n c t h i s 258 6 Embodiment design approach in the light of the minimisation of effort, we are bound to ask to what extent generally applicable function carriers can be determined and documented so that the designer can have ready accessto tested solutions-that is, to known elements and assemblies. This question has also been raised in connection with standardisation which, according to Kienzle [6.100], can be defined as follows: 'Standardisation lays down the definitive solution of a repetitive technical or organisational problem with the best technical means available at the time. It is therefore a form of technical and economic optimisation limited bv the time factor.' Further definitions can be found in [6.33]: 'Standardisation determines the best solution of recurring problems,' or in [6.36]: 'Standardisation is the systematic unification by those concerned of material and immaterial things for the benefit of the community.' Standardisation considered as the unification and determination of solutions. for instance in the form of national and international standards (BSI, DIN. ISO), of company standards, or of generally applicable design catalogues,and also of data sheetsis becoming of increasingimportance in systematicdesign Here, the fact that the objectives of standardisationare to limit the range of possible solutions in no way conflicts with the systematic search for a multiplicity of solutions, because standardisation is largely confined to the determination of individual elements, sub-solutions, materials, computation and testing procedures etc, while the search for a multiplicity solutions and their optimisation is basedon the combination or synthesisof known elementsand data. Standardisation is therefore not simply an important complement to, but the prerequisiteof, the systematic approach, in which various elements are combined as so many building blocks. It is, however, important to stressthe limitations of all types of standardisation. As Kienzle stated: 'Standardisation. . . is a form of technicaland economic optimisation limited by the time factor'. The data in standardsare time-dependentand must be continually updated to reflect technological changes. In what follows, we shall be examining the possibilitiesof, need for and limits of, standards in the design process. In addition, the reader is referred to the c o m p r e h e n s i v el i t e r a t u r e [ 6 . 1 3 , 6 . 3 3 , 6 . 3 4 , 6 . 3 6 , 6 . I 0 I ) . 2 Types of standard In technical devices, standards of various origin, content, range of application and complexity are used in the design of even the simplest components. Thus B d n n i n g e r [ 6 . 4 ] h a s s h o w n t h a t n o l e s st h a n 3 0 s t a n d a r d sw e r e i n v o l v e d i n t h c d e s i g n o f a s i m p l e c o m p o n e n t f o r a p a r t i c u l u r p r e c i s i o n - e n g i n e e r i ndgc v i c c . D e s i g n e r sa v e r s e t o s t a n d a r d i s a t i o n( ' S t a n d a r d i r a t i o na s a s t r a i t j a c k e t ' 1 6 . 4 1 ) should considcr how many standarclsthcy uro unwlttingly in thcir daily work. lt' 6.5 Guidelines for embodiment desien 259 they do so, they will find that standards are the indispensable foundation and prerequisite of all types of design work. The following discussion of types of standard is meant to: - draw the attention of systematic designers to this important method of acquiring an organised body of information; -encourage them to make wide use of standards; - invite them to suggest new standards or, at the very least, to influence the development of standardisation;and remind them of the crux of standardisation, namelythe systematicarrangement of facts with a view to their unification and optimisationin the light of functional considerations. y their origin we distinguishbetween: nationalstandards of BSI (BritishStandards Institution)or the DIN (German StandardsInstitution); European standardsof the CEN (Comit6 Europ6ende Normalisation)and CENELEC (Comitd Europ6ende NormalisationElectrotechnique); recommendations of the IEC (InternationalElectrotechnical Commission); and recommendations and more recentlyuniversalstandards by the ISO (InternationalOrsanisationfor Standardisation). By their content we distinguish, for instance, between communication stanrds, classificationstandards,type standards,planning standards,dimensional dards, material standards, quality standards, procedural standards, opertional standards,test standards,delivery standardsand safety standards. By their scope we distinguish between basic standards,that is general and terdisciplinarystandards;and specialstandards,that is standardsused in list fields. The level of a standard is determined by its breadth, depth and range of lication. A full standard covers every possible aspect, a partial standard does not de all the details, and an outline standard provides a rough and ready mework in areas where technical development might be impeded by full ndardisation. It is usual to develop outline standards before partial standards these before full standards, which are, in any case, relatively rare. One rd can, and usual does, play several roles. Besides the national and international standards we have mentioned. the igner can also have recourse to the rules and regulations published by essionalengineeringorganisations[6.50, 6.2271.These are important as they the way for further standardisationafter initial trials. The designer can also have recourse to a variety of internal company standards regulations as follows: [6.3I, 6.5I,6.66,6.238].Thesecanbe classified compilations of representative standards, that is, a selection from general standards that is applicable to the special requirements of a particular c<lmpany-for instance,stock lists and comparisonsof old with new standards (synoptic stanclarcls) ; 260 6 Embodiment desigl -catalogues, lists and data sheetson bought-outparts, including their storage and also data on the acquisition (ordering/supply) of raw materials, semifinished materials, fuels etc; - cataloguesor lists of in-house parts, for instance machine elements, reperlt parts, assembliesetc; - information sheetsfor the purpose of technical and economic optimisation, f<tr instance on production capacity, manufacturing methods, cost comparisons ( s e e6 . 5 . 6 . 7 ) ; - rules and regulations for the calculation and embodiment design of machinc elements, assemblies,machines and plant, if necessarywith a selection ol sizes and/or types; - information sheets on storage and transporr capacity -regulations poncerning quality control, for example inspection and testinr procedures; -rules and guidelines for the preparation and processing of information, fttt instance of drawings, parts lists, numbering systems and electronic datrr processing;and -rules laying down organisational and working procedures, for instance the updating of parts lists and drawings. 3 Using standards Though there are no absolutely binding standards in the legal sense at the timc of writing, national and international standards are widely treated as regulations, adherenceto which is of great advantagein the caseof legal disputes.This is particularly true of safety standards 16.43,6.54, 6.68]. In addition, all company standardsshould be consideredbinding within their sphere of application, not least for economic reasons. The sphere of application of a given standard is largely set by Kienzle's definition (see above). A standard can only be valid and binding if it does not conflict with technical, economic, safetyor even aestheticdemands.Even in thc case of such conflicts, however, the designer should guard against rejecting <lr replacing the relevant standards out of hand, without assessingthe possiblc consequences. Moreover, he should never make such assessmentby himself, but shoultl always consult the standardsorganisation and the head of his department. In what follows, the reader will find a number of recommendationsand hints for the correct use of standards. First of all, we recommend adherence to national standards since tltc preferred sizes laid down in them help to determine the dimensions of irll c o m p o n e n t s .I f t h e s eb a s i cs t a n d a r d su r c i g n o r c d , t h e n u n p r e d i c t a b l el o n g - t c r t t t c o n s c q u e n c c s( f o r i n s t a n c e ,i n t h c s p a r c p a r t s s c r v i c c ) ,a n d g r a v e t c c h n i c a li t r t t l cconomicrisks may cnsuc. J'hc usc ol' stlndarcls shoulcl bc exumlned uguinst thc chccklist in (r.l itlrovc. 5 Guidelines Ior embodiment design 261 ton the anticipated overall function or sub-function be fulfilled by the use of a ndard solution? If it cannot, the problem (specification) and the chosen function structure uld be re-examined before the search for a solution is besun. 'orking principle n existingstandardshelp the development of suitable solution principles or pts? If they impede this development, then the consequencesof ignoring or anglngthem or of introducing new standardsmust be sublected to a detailed lysis. and form design basicand specialstandards-especially constructional,dimensional, mateand safety-must be fully taken into account. Testing and inspection res also influence the embodiment. Standardsshould only be ignored in borderline problems. ery blishedcomponent, work and environmental safety standards and regula_ must be rigorously observed. Safety standards must always be given ce over rationalisation procedures and economics. field of ergonomic standardshas not yet been adequatelyopened up, so that designerwould do well to consult the general literature (6.5.1) and work in collaboration with production and safety engineers. uction , observance of production standards is particularly important and that of ry regulationsis binding. The designershould only deviatefrom production rds after a broad assessmentof all the industrial and relevant market rchase and sales) aspects. 'ity control standardsand inspectionrulesare essentialfeaturesof quality control. assembly must be ensured by the observation of standard tolerances. and fits, and also of test standardsand inspection rules. )ort , insidcas well as outside the factory, is rendered safer, simpler and cco!l()micali f t h c r c l c v a n t s t a n d a r d sa r c o b s e r v e d . 262 6 Embodiment design 5 Guidelinesfor embodiment design Operation The correct operation of engineering products involves the use of various standards, for example, standard symbols and standard operating procedures. Maintenance Standard symbols (for instance, circuit diagrams) should be used and servicc standards should be provided. Expenditure Costsand deliverytimesmust be minimisedwith the help of companystandards The above list must not be consideredexhaustiveor universallyapplicablethe designer'swork is much too variedand complexfor that, and the rangeol generaland companystandardsmuch wider than we havebeen able to cover in our summary.By working his way down the checklist,fhe designercan tell fairlr quickly to what extent a particular standardfits the variousheadings. It may also be helpful to searchspecialindexesfor the appropriatestandarcls and rules. Fundamental principles for the application of standardsto thc designer's work are alsoset out in a numberof specialcontributions[6.11,6.53. 6.eel. Finally, we should like to refer the reader to the use of preferred numbers anrl preferred series of numbers I6.L6,6.171in the graduation of sizesand in typc rationalisation, especially in the development of size ranges and modular products (see Chapter 7). 4 Developing standards Since the designer bears much of the responsiblity for the development, manufacture and utilisation of products, he should play a leading role in thc revision of existing standards, and the development of new ones. To make a useful contribution to the development of standards, he must first determinc whether the revision of an existing standard or the development of a new standard is technically or economically justified. There is rarely a clear-cul answer to this question. In particular, completely reliable assessmentsof thc economic consequencesare seldom possible because of the complex effects of in-house costs and market influences,and, in any case,would involve considcrable research. The evaluation criteria set out in Figure 6.91, once again arranged rrr accordance with the checklist, can prove of great help in the assessmentof existing or newly proposed standardsif they are used in conjunction with thc u s u a l e v a l u a t i o np r o c e d u r e .N o t a l l t h e e v a l u a t i o nc r i t e r i a w e h a v e m e n t i o n c t l a p p l y t o t h e a s s e s s m e not f i n d i v i d u a l s t a n d a r d s .T h u s , t h e e v a l u a t i o n o f : r d r a w i n g s t a n d a r di s i n f l u e n c e db y i t s c l a r i t y , b y t h c i m p r o v e m c n ti n c o m m u n i c l t i o n , b y t h c s i m p l i f i c a t i o no f t h c d c s i g no c t i v i l y a n d t h c o v c r a l l c x c c u t i o n( ) l ' t h c Examples principle andformdesign Lackof ambiguity ensured position Market of theproduct favourably influenced Material andenergy expenditure reduced Complexity of the product reduced, design worksystematically improved and simplified, partsfacilrtated anduseof replacement Salety increased, Clarity of instructions improved, Psychological and aesthetic conditions improved. Materials handling, storekeeping, manufacturing and quality control lacilitated Execution of theorders planning simplified; production improved; capacity increased Inspecti0n quality andtesting simplified; improved Assembly facilitated, Transport andpacking simplified 0peration clarilied Replacement of partsimproved, sparepartsservice and maintenance facil itated Costsot,and/or timespenton,desigh, workpreparation, materials handling, manulacture, assembly andquality controlreduced, Calculations simplitied Electronic dataprocessing reduces costsof slandardisation. re 6.91 Evaluation criteria for the assessment of standards :r it provides, by the degree to which it is generally acceptedand also by the s its development entails. Before he makes an evaluation, the standards ineer or designer should therefore grade the importance of the various iluation criteria and discard those that may not apply. In much the same way with the recommendationsin 5.8, there must be an adequatevalue rarlns to y the development of standards. Finally and by way of summary, the following principles of developing neral, and particularlycompany,standards,can be enunciated: Standardisationshould only be used if it is economicaland useful. There must be a need. Standards must not contain any provisions that conflict with the law (for instance, with monopoly restrictionsor safety regulations). Standards must be unambiguous, framed in clear terms and under- stood. Standarddimensionsmust, as far as possible,agreewith preferred number series. Standardsmust ensure the complete interchangeabilityof parts. If a standardiscd product is modified in such a way that it can no longer be freely intcrchangcdcvcn in respectof a singlefeature, its designation(identification nurnbcr)must bc altcrcd. 6 Embodiment design -All standardsmust be based on SI units. - Matters of fashion and taste should not be standardised. Colours, for instance, should only be standardisedfor use in identification symbols. - Standardsshould only be altered for technical,not for purely formal, reasons - The development of new standards must be generally agreed by all thc departments concerned. Thus when a new standard is proposed, it should first be examined by a working party, and then presented for wider discussion.The proposed standard thus servesas a draft for the final standard. if adopted. 6.5.6 Designingfor production 6 5 Guidelines for embodirncnt desiqn fulfilment of a given function and to the best solution from a production point of view. Another step in the same direction is the application of general and company standards (6.5.5). Task clarilication design Conceptual design Embodiment design Detail I Relationship between design and production The crucial influence of design decisions on production costs,production times and the quality of the producl has been describedin [6.23,6.110]. Designingfot production means designing for the minimisation of production costs while maintaining the required quality of the product. By production, we usually refer to: - the manufacture of components in the narrow senseby acceptedprocesses - [6.3e]; assembly, including transport of components; quality control; materials handling; and operations planning. The designer will accordingly do well to consult the checklist (6.2) under thc h e a d i n g s ' p r o d u c t i o n ' , ' q u a l i t yc o n t r o l ' , ' a s s e m b l y 'a n d ' t r a n s p o r t ' . In what follows we shall first concentrate on the design of components or assembliesin the narrower sense,while paying due regard to quality control and improvement of the overall production procedure. In 6.5.1 we shall thcn examine design features for improved assemblyand transport. Designing for production is greatly facilitated if, from the earliest possiblc stage, the designer's decisions are backed up with data compiled by the standardsdepartment, the planning and estimating department, the purchasing department and the production manager. Figure 6.92 shows how the flow ol information can be improved by systematicmeans and the appropriate organisir' tional measures. In original designs of mass-produced articles, systemalic procedures and exchangesof views are essential.Their use, however, shoulrl also be encouragedin one-off and small batch production, especiallyin the cast' of adaptive designs,where the designeris often forced to make decisionswithorrl c o n s u l t a t i o n ,i f o n l y t o m e e t d e l i v e r y c l i t t c s . I ) c s i g n i n gf < l rp r o c l u c t i o nh u s b c c o n r ci r r ti n c r c i r s i r t g liyn t p r l r t a n ta c t i v i t y i n l h t ' w a k c o l ' g r o w i n g a u l o r r r a t i o nI.l y o b s c r v i t t gt l t c b i r s i cr u l c s < l l ' s i n t l l l i c i t yi r r r t l 'llrr' c l i r r i t y( 6 . 3 ) , t h c t l c s i g n c ri s i r l r c r r t l vp r o c c c d i n gr r k r t t gt l t c c r l r r c c tl i n c s . p r i r t c i p l c so l r ' r r r l l o t l i r r r ctrht .l ' s i g r( t( r , { ) , t r N r ,c t l t l l c r t ( l t i t t tt r r i t b c t t c t i t t t r ls l r l i ' r manulacture Component Assembly control 0uality Product Figure 6.92. Flow of information For an overview of the designer'sinfluence on the production procedure, and the interdependenceof designand production, the reader is referred to Table . That table highlights severalgroups of problems with a direct bearing on the ionalisation of production procedure [6.14], namely: Appropriate overall layout design which determines the production procedure the breakdown of the product into assembliesand individual components use or bought-out, new, repeat or standard). Appropriate form design of componenrs, which determines the production :cdure, the manufacturing methods and the quality of components. Appropriate selection of materials, which detrmines the production procedure, nranufacturing methods, the materials handling and quality control. Altpntpriutc ust'o.f standard and bought-out components, which influence the r t l u c l i o r rc i r l - l i r c i t yt h , c s t o r a g ea n d t h e c o s t s . A l t l t n t l t r i u t ad o c t u r r t ' n ! u l i o rw r ,h i c h r n r r s tl r c i r r l i r p l c rtlo 1 h c p r o c l u c t i o np r o c e d i t r c . t ( ) l l t c r t l r r t r r f l t c l t r r i rnt rgc t h < l cllrsr r t lt o t l t c t l u l r l i t vc r l n t r r l l . 266 6 Embodiment desitrr Table 6.4. Relationship betweendesign and production Overall layout design Component form design: Production - - - Shape and dimensions Surface finishes Tolerances Limits and fits - Materials selection: - Type of material Treatment Quality control Semi-finished materials Availability - Production procedure Assembly and transport possibilities Batch size of similar components Proportion of in-house and bought-out items Quality control Production procedure Manufacturing methods, machine tools Measuring instruments In-house and bought-out components Quality control - Production procedure Manufacturing methods, machine tools Materials handling (purchase,storage) In-house and bought-out parlr Quality control - Standardand bought-out components - Repeat parts Standard parts Bought-out parts - Purchases Storage Stock control Production documentation - Workshop drawings Parts lists Data processing programmes Assembly instructions Testing instructions - Execution of orders Production planning Production control Quality control - 267 conversely' such production limitations as the capacityof Design Assemblies Components Bought-out parts Standard parts Joining and assembly Transport aids Quality control .5 Guidelines for embodiment design machines,assembly etc,naturallyhaverepercussions on rhedesigner,s choice {i:.tlnt"t iJ:T:f"ll the overall layout. The appropriate sub-division of the overall layout can give rise to diJJ.erentiar, :egral, composite androt buitding-broca methods of coristructionD iffe rential co nstr uctio n metho d By differentialconstructionwe refer to the breakdown of a component(a of one or severalfunctions) into severaleasily manuru.t.,rlo'pu.ts. carrier This idea comesfrom lightweightengineering [6.g7],where that appioac]r'ri,as introduced for the purpose of optimising _toia'-"u.yingcapacity. In both cases,we are entitledto speakof the ,principle of sub_di,iision fo.'p.oJuil;,. As an example of the differential method let us take the rotor of a Synch'onousgenerator (Figure 6.93). rre6'93.Rotor of synchronous generatorafter[6.r2] (AE,G-Telefunken); asforgedpart 2 Appropriate overall layout design The overall layout design, developed from the function structure. determines the division of a product into assembliesand components. With the overall layout design the desrgner: - determines the source of the components, that is whether they are in-housr,. bought-out, standard or repeat parts; -determines the production procedure for instance , whether the parirllr.l m a n u f a c t u r co f i n d i v i d u a l c o m p o n c n t sr l r l r s s c r l r b l i ciss p o s s i b l c ; - t f c t c r n t i n c s t h c d i m a t t . s i r t n i. rsn c l t h c r t g r l l r o x i r r r i r l ch u t c l t ; ; i z c , so l s i r r r i l l r r c o r r r l l o n c n l si,r r r r li r l s <tl h c r n c i r n s< t l 'f u i t t i t t 1i r4t t t la . r . r c i l r / r / t ' ; s c l c c t ss u i l i r b l c/ i / . r ;i r n r l i t t l l r r c r r t ' c tsl t t t t l i t t(' ( , t t t r ( ) ly r r o t . t . r l r l t . r , as dtsc constructionwith forged flanges and with welded flanses The large forging shown_atthe top is divided into several rotor discs consisting simple forged parts and two considerably smalrer flanged ,rruiir, b. Each of : Iattercould alsobe subdividedinto shait,disc supporinurg",-*o coupling nge in the form of a welded construction,c. The reasonfor this differentiar ,nstrucrionmight be the market situation (price, d"liu;;t Oaie; of turge of.the to various outputrequire_ senerator Iillll,,;il"u (r.t.r sizcs) r:nrs ancltypesof coupring. "jl:. :1::,'.:1il,l:i:" i fu.rtheradvantage r, ir,t"in"T"ri, n .c *rrnul'rrcturccr .s st.ck ancrnot nJcessariry to a specifiE ..;;; However, lhc illtrstrirli()ll irlso(lcrrronstrirlcs thc Iir'itrrti.rrs'f'rlrc'cliffcrcniiu';n,r)l.l,lit-. 268 6 Embodiment dcsitrr Guidelines for embodiment design beyond a certain rotor length and diameter, the machining costs become tot' great and the stiffnessof the joints too problematical. Figure 6.94 shows the magnet support of a large-scaleDC motor which carr either be cast in one piece or else be built up from sheet metal and welded. Tht. production costs of the second design are some 25 per cent lower than those ol the first, and this despite the fact that the differential construction involvcs several processes.However, the cost reduction is not constant but depends rtn the relative market situation of castings,sheet metal and semi-finishedmaterial castin I laminated and oneprece I welded procedure Productron 1 Cast 2 Cast biock magnet RoI block plates support Stamp plates Stack plates Rivet Welh dousing lvagnet support duringassembly 1953 1951 1956 1962 Production costsol laminated support asa percentage ol costsof castsupporl Figure6.94.Productioncostsof a DC motormagnetsupportafter[6.104](Siemens) . g u r e6 . 9 5 .W i n d i n gm a c h i n e( E r n s tJ u l i u sK G ) : ) w i n d i n gh e a dw i t h i n r e g r a t eddr i v eu n i t winding head with separatedrive unit Another differential construction is shown in Figure 6.95. In the winding machine a, the winding head is integrated with the drive unit on a common shalt. The differential solution b was developed to facilitate the parallel manufacturc of drive units and winding heads to meet various requirements. In this way, it small number of standard drive units can be combined with a large number of winding heads. The replacement of forged and cast constructions with welded constructions incorporating suitable semi-finished parts provides a further example of tlris method. All in all, differential designs have the following advantages,disadvantagcs and limitations: Advantages: - u s e o f e a s i l y a v a i l a b l e a n d f a v o u r a b l y p r i c e d s e m i - f i n i s h e dm a t e r i a l s o r s t a n d a r dp a r t s ; -easier acquisiti<lnof forgecl and cnst purts; - c a s i c r a d a p t a t i o nt o c x i s t i n g f l c t o r y l a y o u t ( d i m c n s i o n s .w c i g h t ) l - incrctsc irr componcnt hatch sizcri reduction in component dimensionsallowing easier assembly and transport; simpler inspection (smaller components urrd lu.g". batch sizes); easier maintenance, for instanceby simple replicement of woin pa.ts; easier adaptation to special requiremenis; and reduced risk of missing delivery dates. sadvantageo s r limital.ions: greater machining outlay; greater assemblycosts; greater need for quality control (smaller tolerances, necessaryfits etc); and limitations of function by joints (stiffness,vibration, sealing). ral const ruction metho d intcgralconstruction we referto the combination of severalparts into a single l n c n t , T y p i c a l c x a m p l c sa r e c a s tc o n s t r u c t i o n si n s t e a d o f w e l d e d. o n r t r i . _ c x t r u s i o l l si n s t c l d o l c < l n n c c t c cslc c t l o n s ,w c l c l c di n s t c a d of bolted ioints 6 Ernbodiment desigrr etc. In lightweight engineering this type of construction is often used to avoi(l stressconcentationsand to save weight [6.87]. Figure 6.96 showsan example chosenfrom electricalengineering.Here, a casl and welded construction has been replaced with a single cast component Though the castingis fairly complicated, it leads to a cost reduction of 36.5 per cent. Naturally, this percentage will vary with the size of the batch and with market conditions. 6.5 Guidelines for embodimcnt clesign :7t end discs. Variant c is an integral construction in that two cast holkrw botlir.s In.Variantd thecastconstruction is splitup agairr(a ;ilffi il w:,dil':#;f;lll :n1::*,l3"ll: Ti I"":a,r,"ii, ",i;; show -"tnoosaves matert"iiir'in.'it"i, ff;?l:;:T:illi 1) .thattheintegral l:?::lP_:l::d.,rc":her. was chosen because of difficulties in procuring large castings. The advantages and disadvantages of the integial consf,uction method are easily determined by a reversar of the advantages and disadvantagesof the differential method. Comp osite construc tion met ho d By composite construction we refer to: -_ the inseparable connection of several, differently made parts into a single component needing further work, for instance, the combination of cast and forged parts; - the simultaneousapplication of several joining methods for the combination of components [6.ieS1; o. -_ the combination of various materials for optimal exproitation of their Figure 6.96. End cover of electric motor after [6.104] (Siemens); ( a ) c o m p o s i t ec o n s t r u c t i o n (b) integral construction Castandwelded construction Castcomponenl D 0 Figure 6.98 gives an example of the first method: the combination of caststeel mponents and rolled steel sheet into a welded constnrction Further examples are bogies with cast centres and welded arms, and also the :lding of cast bar joints used in steel structures. Examples of the secondmethod are combinations of adhesivesand rivets or of hesivesand bolts. Another example is the rotor of a hydroelectric generator (Figure 6.97). Four different constructionswith the same generator output and identical radial loads were investigated. Variant a has numerous individual support discs and mlr' therefore be considered a differential construction. In Variant b the degree ol division is reduced by the use of cast-steelhollow shafts, two support rings antl radialloads: 14poles :63 108N -+l t]f; M.gnct whcel of a hydro-electricgenerator of compositeconstructronfrom l'1," I 1 . |5( n I r ( ; ' l ' c l c l u n k c n ) l r i A t r r t(' r , t ) 7 ,l t o t o r c r r r t r t r t r r ' l i o ltot t ; t l l t t l ; t 'r c l l t ' l t t r l r r t ' [ c ( ' l I t ( ' l l ( ' n ( r' ; r l r r t ( S i r ' r r r t . r r)r S P r t l i l r r r l l c t ls t c c l s l r c c t S t t l t p , r lt t : r r l s l t . c l 272 6 E , m b o d i m e n td e s i g t t The combination of several materials into a single part is exemplified by synthetic components with cast-in thread inserts; by composite sound-absorption panels which have two plates separated by a plastic core; and also by rubber/ metal components. Another economical design of the composite type is the combination of steel and pre-stressedconcrete 16.7al. B uildin g - b I o ck constructio n method If the differential method is used to split a component in such a way that thc resulting parts and/or assembliescan also be used in other products, then thel can be considered as building-blocks. These are particularly useful if they arc cheap to produce. In a sense,the utilisation of repeat parts from stock may also be considered a building-block construction method. 5 Guidelinesfor embodimcnt desisn PS Guidelines 1-1 (D Right Wrong O: Pa Choose simple tor shapes patterns (straight andcores lines, rectangles) C Pa patlerns, if Aimatundivided (eg by possible withoul cores means ofopen cross sections) C Pa tapers lromthespllfline Provide 0 Pa Arrange ribssothatpattern canbe removed; avoid undercuts 0 w i w ^q_l u I **'a*'*tu*^ q**ffi* F--r_r V ffi ffisw af,^1 % \ MLY) 3 Appropriate form design of components With his form design of components the designer exerts a great influence oll production costs, production times and the quality of the product. Thus his choice of shapes,dimensions, surfacefinishes, tolerancesand joints affects: - the production procedure; - the machine tools, inchtding bench tools and measuring instruments; - the choice between in-house components and bought-otll components; - the selection of materials and semi-finished materials; and -the quality control procedures. Conversely, production facilities influence the design features. Thus, thc available machine tools might limit the dimensions of componentsand necessitate a split-up into several connected parts or the acquisition of bought-out components. There are special guidelines for designing for production, and these arc d e s c r i b e da t l e n g t hi n t h e l i t e r a t u r e f 6 . 2 I , 6 . 2 4 , 6 . 1 2 8 , 6 . I 4 5 , 6 . I 7 0 , 6 . 2 1 3 , 6 . 2 3 1 , 6.248,6.2491.In keeping with the aims of this book, we shall be presenting thc reader with none but essentialdesign suggestionsarrangedsystematicallyin thc form of charts. Our classifyingcriteria will be the processs/eps(PS) used in thc manufacture of the component. In addition, we shall be assigningobjectives-'reduction of costs'(C) and 'improvement of quality' (Q)-to the various desigrt guidelines. When designing components,the designer should always bear thcsc processsteps and objectives in mind. Form designfor primary shaping processes The form designof components to be shapedby primary processes,for examplc casting and sintering, must satisfy the demands and characteristicsof tlrc processesused. ln cast components (primary shapcs obtaincd from thc fluid statc) thc designer must allow for the following pr()ccssstcps: /r(rtlenr(Pa), casting (Cal 'l'lrc and mac'hining(Ma). Figurc (r.99lists thc most important dcsign guidclincs. l i t c r a t u r c c i t c d c o r t t i t i n sf u r t h c r i n l i t r m u t i t l n . Pa Ensure accurale location olcores 0 Ca (bubbles, Avoid vertical sections blowholes) andreduced crosstotherisers sections 0 Aimaluniform wallthicknesses andcross-sections andatgradual changes ofcross-seclion; select material allowing otadequate wall thicknesses andcom0onent sizes 0 Setsplitlines toavoid misalignment andto permrt easy removal oftheflash c Arrange caslings toease machini ng C 0 Ma Ma Ma Provide adequate supporl surlaces Ma Avoidsloping machining and boring surlaces Ma Combine processes machining by appropriate arrangement of machining andboring surlaces Ma Avoid unnecessary machining by breaking uplarge surlaces 0 tu% *{W* i% @S-=@* uz# llash' 16-^!+r t!rz:r,li 0 c C 0 C .-iq l*t ffi. F--n F==-U+ ::,ll -,tl Mo \Ns NSs f,ltal fl /6-) u'g c Mu uu% M i i g u r c 6 . 9 9 , l ) c s i g ng u i d c l i n c sw i l h c x l n r p l c sl i r r c a s tc o m p o n e n t si,n a c c o r d a n c e with 6 . 7 1 ,6 . 11 8 ,6 . 1 4 5 .6 , t 7 0 . 6 . 2 4 7 1 6 Embodiment desigu 274 In designing sintered components (primary shapes obtained from the powder state), the designer must allow for tooling (To) and sintering (Si). In particular. he must be guided by the latest findings of powder technology. The essential euidelines are shown in Fieure 6.100. PS TO Guidelines edges andsharp Avoidrounded angles (D -o= O= C 0 Right Wrong r - 5 Guidelines for embodiment design 275 Design guidelinesfor drop forging have been collatedin Figure 6.101.They low for the processstepsof: tooling (To), forginS (Fo) and machining(Ma). Figure 6.102listsdesignguidelinesfor the cold extrusionof simplerotationally 'mmetricalsolid and hollow bodies.They allow for the processstepsof: tooling lo) and extrusion(Ex). It must be stressedthat only certaintypes of steelcan : usedeconomically.Like all other cold forming methods,cold extrusiongives PS Guidelines I t3> L Si Si Si Si sharp angles edges, Avoidsharp transitions andtangential 0 limits and dimensional 0bserve relations: HeightHAVidthW<25 l> 2 mm Wallthicknesses Holes d> 2 mm 0 small-toothed Avoid Proliles 0 small Avoidexcessively tolerances 0 rH@ t-ffi@ u+@ To Avoidundercuts, C To tapers Provide c T\1zI F-T ..f dil - atabout half To Aimforsplitlines perpendicular height tosmallest height --v To Avoidbentsplitlines c c Wrong Right *w* *w* o s WW ww 0 To Aimatsimple, if possible parts symmetrical, Fo rotationally C protusions, Avoidgreat after[6.611 for sinteredcomponents, with examples Figure6.100.Designguidelines Fo Aimatshapes thatoccurdurrng pressing Forlarge unrestrained numbers adapt tolinished shape C 0 Form design for secondary shaping processes Fo Avoidexcessively thinsections 0 -0 Avoid large curvatures, excessively narrow ribs,frlletsand holes excessively small 0 ;0 Avoid incross sharp changes sections andcross sections that project intothedie excessively 0 split-lines inthecase ol Stagger parts cup-shaped of large depth 0 The form design of components to be shaped by secondary processes, for example, forging, extrusion, bending, must adhere to the guidelines listctl below. Special consideration for the design of ferrous materials can be found irt [6.37] and of non-ferrous metals in 16.a2). With hammer forging (free forging), the designer need only allow for thc actual forging process, since no complicated devices (for instance, dies) arc involved. The following design guidelines should be observed: - Aim at simple shapes,if possible with parallel surfaces (conical transitions arc difficult) and with large curvatures(avoid sharp edges).Objectives:reductiort of costs, improvement of quality. -Aim at light forgings, perhaps by separation and subsequentcombinatiort. Objective: reduction of costs. --Avoid excessivedeformations or excessiveclifferencesin cross-sectionstltrc, f o r i n s t a n c e ,t o t h e p r e s e n c eo f c x c c s s i v c l yh i g h a n d f i n e r i b s o r o f e x c e s s i v c l v n a r r o w i n d c n t a t i o n s .O h j c c t i v c : i m p r o v e m c n to f q u a l i t y . - Try to placc bossesand indcnlrtiunr on on€ tidc only, Objcctivc: rcduction ol cotits, :o linesothat thesplit lia Select misalignment iseasily detected andremoval issimple olllash KKK we -a_*av&_*ffi rr w , v lL u l utu l *G=l* tu%i%" ffi C p a r t si n a c c o r d a n c ew i t h | u r c 6 . l ( f l . l ) c s i g ng u i r l c l i n e sw i t h c x a n r p l e sf o r c l r o p - f o r g e d 9 5 . 6 . t 7 ( ) ,6 . 2t 4 . 6 . 2 5I I Guidelines for embodiment design 6 Embodiment dcsilrr 276 I I PS wr0ng Guidelines i ) ' - TO Ex Avoidundercuts 0 Ex andexcessively Avoidtapers differences smalldiameter 0 c Ex Provide rotational lysymmetrical u partswithout pr0trusi0ns, material split andjoin otheruise in cross changes Ex Avoidsharp 0 edges andfillets sharp section, Ex Avoid small, longorlateral holes andthreads 211 0 Right ww ww ww MM MWW wwM ilwM tuwM wwwwww after[6.631 for coldextrusions, with examples Figure6.102.Designguidelines rise to work hardening, in which the yield strength is raisedwhile the toughness of the material drops significantly. The designer must take this factor inltr consideration. The best materials for cold extrusion are case-hardeningarrrl heat-treatablesteels. For drawing, the following design guidelines are recommended in [6.170]: - Allow for tooling (To): Choose the dimensionsin such a way that the smallest possible number of drawing steps is needed. Objective: reduction of costs. -Allow for tooling and drawing (To/Dr): Aim at rotationally symmetrical hollow bodies; producing the corners of rectangularhollow bodies leads to it high loading of the materials and tools. Objectives: improvement of qualitv, reduction of costs. - Allow for drawing (Dr): Choose tough materials. Objective: improvement of quality. -Allow for drawing (Dr): For the design of flanges see [6.148]. Objectivc: improvement of quality. Bending (cold bending) as it is used for the manufacture of sheet metirl components in precision and electrical engineering, and also for casings, claddings and air ducts in general mechanical engineering [6.1] involves twtr separate steps, namely cutting (Cu) and bending (Be). The designer mtrsl accordinglyallow for both. The designguidelinesshown in Figure 6.103 apply ttr the bending processalone; cutting is covered under the next heading. Form designfor separution O f t h e s c p a r a t i n gp r o c c c l u r c sn r c n t i o n c d i n 1 6 . 3 9 1o n d [ 6 . 3 1 t 1w . c shall ortlv ' n r a c h i n i n gw i t h g c o n r c t r i c l l l y d c f i n c d c u t s ' ( t u r n i n l q ,b o r i n g , n r i l l i r r g ) cunsiclcr S Guidelines Wrong => le parts bent Avoid complex (material waste); rather splitand J0rn C 3e Allow forminimum values ol radii(bulging bending inthe c0mpresst0n area and overslretching inthetenslon area) llange height andtolerances 0 le Provide icient suff distance pre-pierced holes and between bend 0 le Aimatholes andnotches to cross when thebend it isnotpossible to provide gap theminimum 0 lo Avoid sloping edges andtapers in theregion ofthebend 0 to Provide clearances atthecorners when allsides aretobebentup 0 Right I = | (l maleria ) , -r-. --- fi='r' ar @d & @fr [E 6.103.Designguidelineswith examplesfor bent parts,after [6.11 'machining with geometrically undefined cuts' (grinding). In addition we lbe considering separation by cutting as defined in [6.a0]. In all separating ;essesthe designer must allow for tooling (To), including clamping, and hining (Ma). ign for tooling involves: 'he provision of adequate clamping facilities. Objectives: improvement of uality. \ preferential sequence of operations that does not necessitatethe relamping of components. Objectives: reduction of costs, improvement of uality. 'he provision of adequate tool clearances. Objective: improvement of quality. sign for machiningin all separatingprocessesinvolves: The avoidanceof unnecessary machining,that is reductionof machined arcas, fine surfacefinishesand closetolerancesto the absoluteminimum (protrudingbossesand cut-outsplacedat sameheightor depth are advantugcous),Objcctivc:reductionof costs. 6 Embodimcnt clcsilrr 218 location of machined surfacesparallel or perpendicular to the clamping surfaces. Objectives: reduction of costs, improvement of quality' -The choice of turning and boring in preference to milling and shaping -The Objective: reduction of costs. Figure 6.104 represents the design guidelines for components machined br turnlng; Figure 6.tOS tor components machined by boring; Figure 6.106 for PS Guidelines (-) To toolrunoul adequate Provide 0 TO toolshapes Aimforsimple c Guidelines To Aimforstraight millrng surlaces; C form tools areexpensive; select milling forgang dimensions t0 Provide runouts foredge mills; edge milling ischeaper thanend milling c Adapt runout tomilling tool diameter Avoid longmilling cuts (eg byselecting curved surfaces slots). c To Ma Arrange surlaces ononeleveland parallel totheclamping, andtight To Avoidgrooves TO C oninnersurlaces tolerances 0 clamping, loradequate Provide 0 eg machining, Ma Avoidexcessive highcollars byseparate replace parts c andsurface working length Ma Adapt function, finish totherequired c @ .mft_''*' G j[ft."'-' Aftrftirff, Lu-il+*Jl wL-N_s ni a--l io Figure 6.104. Design guidelineswith examplesfor componentsmachinedby turning. in aciordance with [6. 128, 6. 1701 PS Guidelines possible, tools useborlng To Where holes Ma onblind (D = > L J : c 0 flats 0 andfinishing To Provide starting angled through breakino Ma lorholes surfaces To holes, Aimforcontinuous blindholes avoiding Wrong (.) PS q) -) .= Right ffiffi % ffi@M ffi@M C by boring,in ntilcltittcd fitr com;xrncnls Figurc(r.l()5.Dcsignguidclincswith.cxirnr;rlcr l c l ' o r d l n c cw i t h 1 6 I. 2 8 .6 . 1 1 5 6. . I 7 ( l l 279 Wrong Right 0 Right Wrong -\ -= 5 Guidelinesfor embodimentdesign c 0 t+l M, f=-'f @h 6.106.Designguidelines with examples for componentsmachined by milling, in ancewith [6.128,6.1701 iomponentsmachinedby milling; and Figure 6.107for componentsmachinedby lrinding. , In the designof cut-out components,the characteristicsof.the tools (To) and ,f.thecuttingmethod(Cu) [6.85]mustbe takeninto consideration (Fig. 6.108). iorm design joining for )f the joiningmethodsdiscussed in [6.a1]we shallonly considerweldingunder he aboveheading.For separablejoints the readeris referredto 6.5.7. , Welding involvesthree processsteps,namelypreparation(pr), wetding(We) nd finishing (Fi). The following designguidelinesapply: -Pr, we, Fi: avoid the imitation of cast designs;preferablyselectstandard, easily obtainable or prefabricatedplates, sectionsor other semi-finished materials; make use of compositeconstructions(cast/forgedcomponents). Objective:reductionof costs. - we: adaptthe material,weldingquality and weldingsequenceto the required strength,sealingand shape.objectives:reductionof costs,improvementof quality. - we: aim for small weldingseamsand small dimensionsto reducedamage through heating and to simplify handling. objectives: improvementof quality,rcductionof costs. -wc/Fi: minimiscthe amountof welding(heat input) to avoid or reduce 280 PS Io 6 Embocliment dcsrr:rr Guidelines Avoid edge limitations Wrong => distortion and corrective work. Objectives: improvement of quality, reduction of costs. Further guidelines are given in Figure 6.109. 0 C rEJ---tlEl f--- TO Right 281 5 Guidelines for embodiment desisn forgrinding Provide runouts wneet s 0 PS flL+-rEr L_F---_I- Pr +%r? Guidelines and Preter withfewparts solutions weldseams c t)--@t To grinding by Aimlorunimpeded ofsurfaces appropriate selecti0n To Giveprelerence toequal blend possible) andto Ma radii(ifnorunout C 0 il weldable seams Pr Aimtoreasily We loadspermit Fi c c Pr Avoid ofweldmaterial build-up We andintersecting weldseams C 0 We Reduce residual stresses dueto 0 0 equaltapers TO Guidelines prefer Aimforsimple cuts, angular corners, avoid curves (-) O !> O= Wrong c .GS v/ TO vr' Cu Avoid sharp-angled shapes and 4 4 4 ru mm c 0 posrtive We Ensure location ofthe priortowelding Fi components Allowsufticient material lor machining welding atter 0 0 tr w f, Wt I Iolerance 6 . 1 0 9 .D e s i g ng u i d e l i n e sf o r w e l d e dc o m p o n e n t s i.n a c c o r d a n c w e i t h [ 6 1 4 5 .6 . 1 7 ( 6 2t3l C I Appropriate selection of materials and of semi-finished materials 0 excessively tighttolerances permitling Preler shapes subsequent cutswithout danger of damage. We Aimlorgoodaccessibllity Fi Aimforsharp-edged transitions C to facilitate thecutting ofthe 0 template andtoensure easy grindi ng layout Cu Avoidwaste bycareful ol partsonstandard plate cut-out widths Cu Right 4 choice of shrinkage byappropriate weldseams andwelding sequence, andalsoofconnecting (flexible secti0ns 0l lowstiffness tongues andcorners) F i g u r e6 . 1 0 7 .D e s i g ng u i d e l i n e sw i t h e x a m p l e sf o r c o m p o n e n t sm a c h i n e db y g r i n d i n g .i n accordancewith [6. 1701 PS Righl Wrong o) --= F-rl -1=- E 0 liigurc6.1(llt,l)csigrrguirlclirtcs l'orcut-outcomFrnonlr,In uccortllnccwith l().l?(ll [n optimum choice of materials and semi-finished materials is difficult to make tecause of interactions between characteristics of the function, working princi)le, layout and form design, safety, ergonomics, production, quality control, lssembly,transport, operation, maintenance,costsand schedules.When expenlive materials are involved, their careful selection is nevertheless of the utmost tconomic importance. In general, the designeris advisedto consult the checklist ,Figurc 6.2) and to evaluate the materials accordingly. Thc selected material and the resulting processing and machining of the : o n r p o n c n t s ,t h c i r t l u a l i t y a n d t h e m a r k e t c o n d i t i o n si n f l u e n c e : 282 6 Embodimentdesirrr - the production procedure; - the choice of machine tools, including bench tools and measuring instruments. -materials handling, for example, purchasing and storage; - quality control; and also -the choice between in-house and bought-out parts. The close relationship between design, production procedures and materials technology calls for close cooperation between the designer, the productiorr engineer, the materials expert and the buyer. The most important recommendations for the selection of materials for primary shaping processes(for example casting and sintering) and seconclarr shaping processes (for example forging, extrusion etc) have been set out br Illgner [6.91]. Few designers are completely familiar with the selection ol materials needed for such new manufacturing methods as ultraS'onicwelding electron-beam welding, laser technology, plasma cutting, spark erosion arrtl electrochemicalprocesses.These topics are discussedin [6.28, 6.55, 6.88, 6.13(). 283 Guidelines for embodiment desien 'es to show that minimisation of weight, which often involves a great deal of gn and technical effort, does not necessarilylead to minimisatLn of costs. over, even when the calculated cost reductions due to the incorporation of finished materials and simplification in manufacturing methods are nor t, the actual savings may be much greater because of the consequent ction in idle time and time spent on operations scheduling. 1- 0 e20 Fo 520 6.1831. Closely connected with the selection of materials is the choice of semi-finishcrl materials (for example, tube, standard extrusionsetc). Becauseof the comm()rl method of costing by weight, the designer tends to think that cost reductiorr invariably goes hand in hand with weight reduction. However, as Figure 6.111) makes clear, that belief is often mistaken. I platethickness (2 16mm) F i g u r e6 , 1 1 1E . l e c t r i cm o t o r housingof weldedconstruction b (Siemens); 3 platethicknesses (a) currentdesign ( 5 ,1 0a n d1 6 m m ) (b) proposeddesign re example also shows that, for the minimisation of production costs, the ral production overheads must be considered in addition to the cost of :rials and labour. Finally, it draws attention to another series of problems the modern designerhas to solve: he must adapt his designsto the demands = potential of numericallycontrolledmachinetools. Let us stay with the l-cutting machine. with his decisionto reduce the numbei of plate L Figure6.110.Costareasfor lightweightandeconomical from [6.2191 constructions, constructionconstruction The following examplethrows further light on this problem.Figure 6.111 showsa welded electricmotor housing.The old layout involvedeight different platethicknesses to achievethe requiredstiffness with a minimisationof weight In the modifieddesign,however,the numberof plate thicknesses wasdeliberately reducedalthoughthis increasedthe weight. This changein the design involvedthe replacementof standardflame-cuttingby numericallycontrollcd machines. The extraoutlaywasto be justifiedby keepingthe programmingantl re-equipmentcostslow and by maximumutilisationof the plate matcriirl throughstackingbeforecutting.A costanalysis showedthat,despitean incrcasc in wcight due to oversizingof sonrcof the housingports,the new dcsignwirs chcaper than the old thanks to klwel labour costs and lowcr procluctiolr overhcuds.Adrnittcdly,lhc actuulsuvinj wm not very grcut, but this cxlmplc nesses,the designer opened the way for a computer-aided layout of the rial on the plates, and hence for the highly cost-effective introduction of lrically controlled flame cutting machines [6.10]. further example of the economic use of semi-finished materials is siven in re 6.r12 which showsthe plate-cuttingplan for a weldedmotor hous=ins. To L__ r e 6 . 1 1 2E . l e c t r i cm o t o r housing,Weldedconstruction i t h p l a t cp l a nf r o r n[ 6 . 1 1 5 1 (Sicnrcns) b c Plate-cutting plan 6 Embodimentclcsrr,rr 285 Guidelines for embodiment design allow the use of circular blanks for the end wall bearing shieldsd, the end wrllr are made from four parts b, which are then welded together. The resultirrg aperture, even after machining, is smaller than the bearing shield made from l hc blank. In addition, this arrangement provides the support feet c. rential construction will be more cost-effectiveand quicker to execute than integral construction. He must accordinglycombine the functional approach h an analysisof production and supply conditions anclwith careful costing and evaluation. Ultimately, designingfor production means minimising costs 5 Appropriate use of standard and bought-out components arrive at the most cost-effective solution, the designer has recourse to value analysis The designer should always try to use components that do not have to lrc specially manufactured but are readily available as repeat,standard, or boup,ltr out parts. In that way, he can help to create favourable supply and storrrlc conditions. Easily available bought-out parts are often cheaperthan parts mrrrlc in-house. The importance of standard parts has been stressedon several occasionsThe decision whether components are to be made in-house or bought-orrt depends on the following considerations: - number (one-off , batch or mass production); -whether production is for a specific order or for the general market; - the market situation (costs,delivery datesof materials and bought-out parts ); -the possibility of using existing production facilities; -the manpower situation; and - the available or desired desree of automation. These factors influence not only the decision whether in-house production to be preferred to sub-contract production, but also the designer's ovct approach. Unfortunately, most of the factors vary with time. This means thirt particular decision may be justified at the time when it is made but may' longer be right if the market or manning situation and the production capac have changed. Particularly in the caseof one-off or batch products of the hcrr engineeringindustry, the production and market situation has to be re-examir at regular intervals. ysis 16.222,6.228,6.2301.In the discussionand evaluationof the cost- :tivenessof a particular solution, collaboration between the various denartts concerned-that is, exchangesof view between the salesteam, the buying r, the design team, the production team and the costing department lvatue ysis team)-proves invaluable. The resulting pooling of expert knowledge experienceprovidesa roundedassessment of the requirements, layout aid t designs,materials,productionprocedures, stock-keeping provisions,stan- how? | why? | why? how? 1stlevel | 2ndlevel | how? ; why? 3rdtevel | how? 4thtevel C' I C, I C, I C, I C5| c6 I C; I C8 6 Appropriate documentation The effect of production documents (in the form of drawings, parts lists assembly instructions) on costs, delivery dates, product quality etc is olt underestimated. The layout, clarity and comprehensivenessof such documc have a particularly marked influence on highly mechanised and automir production methods. They determine the execution of the order, product planning, production control and quality control. 7 Costing and cost evaluation lirr prriluctionwill not hclp thc dcsigncrto nrlslc of clcsigning Thc guiclclirres i r l l t h c c o r n p l c x p r o b l c n r s i r r v o l v c d u n l c r ; h c u l s o t i t k c s i n t o i r c c o t n t li t l l I rclcvirrrtdirlir witlrorrtwhiclr lrc cunnrtlprcdicl willt ccrtirinlvwhcth('t il 6, I l.-1. Assigrrnle nt ol asscnrtrlies andcornponents to subfunctions [6.22tt,6 230] 286 6 Embodimenttlrrrr:n dardisation and market conditions, and hence leads to more reliable and quickcl decisions than the designer could have arrived at by himself. It also helps to split up the overall function into sub-functionsof decreasirrg complexity and to assign these sub-functions to function carriers, that is lrt assembliesand components. (Figure 6.113illustratesthis procedure schematit:rl. ly.) From the calculated costs of componentsit is then possibleto estimatc tlre cost of fulfilling the required overall function and the sub-functions. Strr'h 'function costs' provide a basis for the evaluation of design variants, covering rrt they do market considerations(are all functions absolutely necessary?),dcsirlll considerations(the choice of suitable function structures and solution concr'l)tr and the sub-functions they entail), and production considerations (form desiln of individual components). For the minimisation of costs,it is advisableto proceed at the earliestpossilrlo stage to the optimisation of the economic factors and hence of the productlrl procedure. The choice of a suitable solution concept usually cuts produclrotl Guidelines for embodiment desisn 115 gives the cost distribution of a synchronousgenerator[6.104].It showsthat mpts to lower the labour and overhead costs of the rotor shaft R1 by design ures alone are not likely to be very successful, whereas a reduction in ight or the choice of economical materials will lead to a significant drop in ;ts. with the stator housing s3, on the other hand, a changeof design leading an alternative manufacturing method looks promising in view of the high R1 R2 R3 R4 B5 R6 92Jl c^^ costs more significantlythan do direct production measures.Moreover changesin the designare deferreduntil the productionstage,they often ent;r high alteration costs. Figure 6.114 illustratesthis point. In short, prorlrrc 3ca M1 optimisation should be begun just as soon as the available information pernrits, Figure6 115.Coststructureof a synchronous generator,from [ 6 . 1 0 4(]S i e m e n s ) ExamplesR1: rotor shaft;R2: rotor body;R5: rotor winding; SJ:statorhousing;55: bearing;56: spider;M2: mountinssetc O6 of alterationcostson. tltc Figure6.114.Influenceon costreductionby, anddependence from [6.22t3] designand productionphases, Cost structure Useful aids to cost reductionare providedby estimatesof the coststru( Without a grasp of the cost distribution-that is of the relative shitrc m a t c r i a l s , l a b c l u r , a n d p r o d u c t i < l nt l v c r h c a d si n t h c c o s t s o f a c o m p o n c t l t I asscmbly-the dcsigncrcann()ttcll whut meo$uroshc must takc to l<lwcrI cstimittcs costs.Hcnceit is importanttrl providcthc rolcvantdocumenlatiun: rlf old ollmrter for nditpiivcdcsigns.liigr rlriginuldcsignslncl rccllcul:rtiotts 281 s4 s5 s6 s7 s8 s9 M2 10 Production costs addition to the cost structure, the designer must also take into account the ute costs of the components, materials, semi-finished materials. and of rd and bought-out parts. The relevant figures must be prepared as quickly accurately as possible by the estimatingdepartment, at leasi in the form of ive costs. Figures 6.116 and 6.117 show such relative cost comparisons. often suffice for arriving at decisions and have the advantage over abolute computations that they fluctuate less and hence are more generally cable over a longer period of time. iable preliminary calculations are still no more than a dream of modern rs, who are far too often left to their own devices. In view of the sreat ence of these decisions on the production costs, this state of affairs has intolerable and should be tackled by the engineering industry at the possibleopportunity. For the cconomical design of components and simpler technical artefacts, It)l suggcstsan evaluation procedure and gives a comprehensivelist of the i t n t t c c h n i c a la n c lc c < l l r < t m idca t u . 288 6 Embodiment desilrr 289 Guidelines for embodiment design plaleslaminate I windI soak Other assemblies Materials Testing Figure 6.116. Comparative costsof 30 mm round bar steel,from [6.1901 (a) Costsby weight (no requirementsas to strength and weight) (b) Ratio of costsper unit weight to yield strength (c) Machining coststo produce constantsurfacefinish ( S t e e lI : 2 0 C ; S t e e l2 0 : 1 6 5 C l l 2 % C r - M o - V ) I ''' 3 2 2 8 7 t , 2 01 6 1 2 I 4 0 1 2 3 4 5 6 7 I I 1 0 1 1 1 21 3 Dellvery timein weeks Manufacturing timein weeks re 6.118. Production procedure of an electric motor from the series shown in Figure (AEG Telefunken) / e 1,1 o i E 17 acturing the components and assembliesare indicated by the lengths of horizontal lines. The diagram not only makes clear where improvements can made by the choice of more quickly procurable raw and semi-finished ials or by keeping these materials in stock, but also where different facturingstepscould be taken in parallel. Thus by allowingthe stator 1l 1.0 inations to be built up in parallel with the construction of the housing (two -consuming operations), a significant reduction in the overall production ule is possible. breakdowns of the production procedure are now common practice in project planning and control of all large companies. They facilitate the ination of critical times and of possible design improvements [6.94, F i g u r e6 . 1 1 7C . o m p a r a t i vceo s t so f b o l t e dj o i n t s ,f r o m [ 6 . 1 9 0 1 Production time Accuratc production timc cstinrntcs hclp the dcsigner tu improvc <ln clclivcr datcs, Figurc 6.1 lfl givcs an cxnmpler the production sequcncc l'or il m c d i u r 'l'hc powcr clcctric molor, tinrcr sp€nt on rcquiring thc mntcrial itntl oil 7 Designing for easeof assembly Types of assembly a s s c t t t b l yw c r c f c r t o t h e c o m b i n a t i o no f c o m p o n e n t si n t o a p r o d u c t a n d t o 290 6 Embodiment desilrr the auxiliary work needed during and after production. The cost and qualirr of a product depend on the type and number of assembly operations and orr their execution. The type and number, in their turn, depend on the layorrt design of the product and on the type of production (one-off or batclr production). The following guidelines for designing for ease of assembly can therefore lr,, no more than general hints. In individual cases, they may be influenced,'r overridden by reference to the following headingsof the checklist (Figure 6.lr function, working principle, layout and form design, safety, ergonomics, pr,, duction, quality control, transport, operation and maintenance. According to [6.3, 6.200, 6.224] the following essential operations rrrt. involved: - Storing of parts to be assembled, if possible in a systematic m a n n ( ' r Automatic assembly further necessitatesthe programmed supply of p a r t s u r t l connecting elements. -Handling of components, including: oidentifying the part by fitter or robot; o picking-up the part, if necessary in conjunction with individual s e l e c t i tr r t and dispensing;and omoving the part to the assembly point, if necessaryin conjunction wirlr separation, manipulation etc. -Positionireg (placing the part correctly for assembly), and aligning (firrll adjustment of the position of the part before and possibly after joining). -Joining parts by the provision of appropriate connections. According rrt 16.41],the following operationsmust also be includedhere: o bringing together,for exampleby inserting,superposing, suspending folding: o filling, for exampleby soaking; o pressingtogether,for exampleby bolting, clampingor shrink fitting; o joining by primary processes,for example by fusing, casting irr vulcanising; o joining by secondaryprocesses, for exampleby bendingor by auxilirr components;and o joining by the combination of materials, for instanceby welding, solder or glueing. - Adjustingto equalisetolerances,to restorethe requiredplay etc [6.2001. - Securingthe assembled partsagainstunwantedmovements underoperatiorr - loads. Inspecting. Depending on the degree of automation, various testing it measuring operationsmust b e p e r f o r m e d ,p o s s i b l yb e t w e e ni n d i v i d u a la s s c r bly operations. Thcseopcrationsare involvcclin cvcry asscmbly proccss,thcir imp<lrtirrrcc. scqucnccand frcqucncyclcpcndirrg on thc numberul' urrits(onc-ol'fasscrnl)l\., batchasscmbly) (manunl.part:lut()nriltic and thc dcgrccof uutrtmotlon or l'rrllv tuutomirlic lsscnrbly). Guidelines for embodiment design 297 General guidelines for ease of assembly tional production involves the simplification and automation of the assemblv rcess[6.32, 6.236]. It is advantageous,first of all, to standardisethe necessaryassemblyoperarns. Such standardisation means using the minimum number of assembly niques and, above all, assemblytools. The designercan make a considerable tribution here, for example by choosingone bolt size to satisfyvarious load ditions even if some may be larger than he might otherwise specify. A lurther requirement is the provision of simple assembly operationLs(6.3.2). rr instance, in one-off production it is advantageous to use standard tools her than expensivespecial tools. The cost of individual operations, however, pends greatly on the available assembly equipment and the staff, so that it is ible to make general pronouncements on what is simple and cost- If the design permits parallel assemblyof different sub-assemblies. then a nsiderablereductionin the overallproductiontime is possible. In general, the designer should always aim at a reduction in the number of mbly operations. Since these operationsdepend on the number of individual ponents,he must try to: decrease the number of identical components, for instance, by replacing a large number of small bolts with a smaller number of larger ones; combine several components into one larger component (integral construction); use pre-assembled (bought-out) assemblies; and facilitate the combination of severaloperations by appropriate arrangement of locating surfacesand connectors, to ensure, for instance, the simultaneous tightening of several bolts (see Figure 6.tI9). Even though the precisesequenceof assemblyoperationsis determined by the oduction planning department and not by the designer, the latter should rtheless try to provide for a logical sequence, thus obviating mistakes and ring simple repair and maintenance. Guidelines for improving assembly operations ease of assembly, the designer should consider each assembly operation rately. s operation, which is particularly important in automated assembly,is itated by the use of easily stackedcomponents.The appropriate design asuresinclude: thc provision of compatible stacking surfaces;and thc pr<lvisionclf shapesensuring the correct orientation of non-symmetrical p l r t s ' ( h o l c s , p i n s , g r o < t v c sc t c ) . 292 6 E m b o d i m e ndt c : r r , r r 5 Guii d e l i n e s f o r e m b o d i m c n t d e s r s n 293 up Guidelines Wrong Right Alml0rsimple assembly operations Aimlorfewassembly operations andsimple components ls olperation is particularly important in automated assembly.By appropriatc lice of design features, the designer must: avol d the entanglement of individual parts; prevent the nesting of individual parts; and provide, if necessary,specialfeaturesto ensurepositiveholding of the coml component. ovtnS t e m covement of parts from the store to the assemblysite is greatly influenced the rsize, weight and type (one-off, mass production) of the part. In general, weve er, the designer should aim for: short't distances,for instance by appropriate splitting of the product into easily A S S Embled T modules: good 7 ergonomic and safety provisions, for instance by avoiding visual obstr ructions or potential danger spots; and also Provide lorparallel assembly stmp stmplehandlingmethods,for instanceby the provisionof easilymanipulable joints. trans transportequipmentor of easilyaccessible re e shr shapeof componentsmust accordinglybe designedfor easytransport. tstttol nonmg Reduce number ol identical components rdres resen[6.3] divides positioning into orientating and aligning. In both, the signe rer should: a l mr .for symmetry if no preferential position is demanded; distir tinguish permissible or prescribed positions, for instance by surface marks or br by the shape of the locating surfaces; (integral Combine components construction) a l mr .for the automatic alignment of the joints; and if that is not possible tvide adjustablejoints. prov tn8 trkl-.: H/&/ Combine several operations /'V/r/;V bolldlrection Figure 6.119. Design guidelineswith examplesfor easeof assembly,in accordancewitlr [6 3l Identfying Avoid muddlingsimilarparts, for instanccby thc usc of: - distinctshapes; - distinctdimcnsions of similarshope$t or - clistinctfinishcs. for easy joining entails the appropriate choice of joining method and lors so that: jointts that have to be disassembledfrequently, for instanceto replacewearing parts s, are equipped with easily separable connectors; only' those joints that are rarely or never disassembleduse connectionsthat are e expensiveto separate, for example shrink fits or welded joints; postt tioning is combined with joining where possible, for instance by means of ting connectors; locat any lack of fit between stiff components is allowed for by the insertion of 'ble flexil or compensating elements; and general, procedures should involve a minimum number of simple operain gc tions ns and minimum tool use, provided, of course, that the function can be full'il Il'iIlccl. tc t,l.t itt.s(rtion of a part into the locating surfacesis facilitated by: ((1.$y l.$Y (l('('('J.f t o t h c l o c a t i n g s u r f a c e ; 294 6 Embodiment rlr -visual checks,for instance inspection holes; -simple movements (for example translational) at the locating surfaces; -short movements at the locating surfaces; -special insertion facilities, for instance, by chamfering; - avoidance of simultaneous fitting operations, for instance in stepped shaft - idelinesfor embodimentdesisn Guidelines Wrong Right toring: . Easy stacking holes, by choosing locating surfaces of different lengths; and by avoidance of double restraints. Adjusting -provision of sensitive,repeatable adjustments; and - avoidance of adjustments that affect previous adjustments: - rendering the results of adjustments measurable and controllable. 295 . clear orientation ling: land ldentitying . Dysnape Securing t--l--l -TT[- qJ LJ righlhdthread To lock the joints against unwanted movements due to operating loads, advisable: bysrze -to chooseself-locking,for exampleappropriatelypreloaded,joints; or - to provide suchadditionalform-fitting or frictional locksascanbe asseml without greatcost. f-'r-l -TTrr- -{-;iil:}-@- left-hdthlead tr u rTI,.l l:L Ef, Picking up . Nohidden nesting Inspecting holes Provision otsteps, By appropriate measures, the designer must also provide for: - simple checks on critical requirements (for example air vents in rota machines). Inspection rules without adequate inspection facilities pointless; and - inspection and further adjustments without dismantling already a s s e m b l parts. Figure 6.120 gives examples of components designedusing these guideli Further examples will be found in 16.223). Moving . Possible sliding orrolling Ease ofmanipulation bsitioning: 4 Evaluating easeof assembly The above design guidelines for ease of assembly are quite generally applica Their importance, or in other words their practicability, is however stror influenced by the nature of the production process and hence by the naturr the assembly operation. Thus, storing, picking up and positioning are particu ly important in automated assembly plants for mass-produced articles, w moving, positioning and inserting are of special importance in heavy enginc ing. Another crucial consideration in all types of production is whether a gir product is assembledonce and for all by skilled company engineers,or whct its maintenance and repair call for rcpcatcd asscmblyoperations by lcss cx1)(' outsidestaff. Even if the designguidclincswc huvc given arc scrupulously <lbscrvcd, ol'individuillcrscs spccialcharactcristics murl rtlllbc tukcnintu account.l.or I il noprelerred | . Symmetry Dosttton I I | . I Marking ofpreterred Position i, . Automatic alignment L# r# ---1--- oladjustment . Ease in with examples for improvingassembly operations, 6. 120.Designguidelines a n c cw i t h6 . 3 . 6 . 2 2 3 1 6 Embodiment Coping with design faults, disturbing factors and risks 291 Evaluation criteria for assessingease of assembly Joining: method Joining . lorrepeated assembly . simple, if function allows . positioning combine and l0rnrng . fortolerances. Allowing for instance bycompensating parts Simple executionof assemblyoperations o Storing o Handling Identifying Picking up Moving o Positioning Orientating Aligning o Joining o Adjusting o Securing o Inspecting lnserting . goodaccessibility . improved lead-ins . avoidance ofsimultaneous littingoperations Securing: . if possible simple, without additional elements Freedomfrom possibleassemblyerrors Avoidanceof damageto components Maintenanceof safeworking conditions Observanceof ergonomicstandards Avoidanceof specialtraining of the assemblystaff Coping with design faults, disturbing factors and risks .1 Identifying design faults and disturbing factors Figure6.120(continued) evaluation of easeof assemblyit is helpful to consult the evaluationcriteria gi in Table 6.5. To begin with, all the components must be examined to determinc t cost-effectivenessand scheduling. Next, the overall assemblyshould bc scr n i s e di n t h e l i g h t o f t h e g e n e r a le v a l u a t i o nc r i t c r i ag i v e n i n T a b l e 6 . 5 . E x p c r i c h a s s h o w n t h a t d e s i g nf o r e a s eo f a s s c m b l yc a n n o t b c q u a n t i f i c d a c c u r i r t c l y , t h a t i t s c v a l u a t i o nm u s t , a s a r u l c , b c o f u q u a l i t u t i v ct y p c . l n p l a n n i n ga s s c r r r for lnass-productiolt ()n I tltc dcsigncrwill do better to buschis cvaluatir)ns prtxluctionol' grrotolypcs, design process involves a series of creative and corrective steps. so far we been dealing with the former alone. The designer, however, is forced to amine his ideas and solutionstime and again, and to that end he makes use lection and evaluation procedures.These facilitate the systematicsearchfor links. Even so, the designer can make mistakes, for instance by ignoring faulty features as appear with progressiveembodiment. Now, it is essential ify potential faulty behaviour at the earliest possible stage and, moreeven in the case of satisfactory behaviour, evaluate the possible effects of , ing factors. idcntification of design faults and disturbing factors is facilitated by the riite abandonment of the optimistic and creative approach in favour of a I a n d c r l r r c c t i v eo n e [ 6 . 1 5 3 , 6 . 1 5 9 ] . T h i s c h a n g ei n p e r s p e c t i v ei s o f t e n : u l t b c c a u s cs u b j c c t i v c i d e a s k e e p i m p e d i n g o b j e c t i v ee v a l u a t i o n s . 298 6 Embodimenr il 299 g with design faults, disturbing factors and risks I Fault-tree analysis The influence of faults and disturbing factors can be determined systemati( by recourse to what is known as fault-tree analysis 16.73,6.1271. From the conceptual phase, the designer knows what overall function rr individual sub-functions have to be fulfilled. The established function structrr can thus be used to identify all the functions to be checked. These functions rr now negated one by one-that is assumedto be unfulfilled. By reference to I checklist(Figure6.2),lhe designercan seekout the possiblefaults or disr bancescausingparticularfunctionalfailures.The OR or AND relationshilr thesefaults and their effectscan then be examined. The conclusions will help him to improve his design and, if necessar\. re-examine the solution concept or to change the rnethod of productit assembly,operation or maintenance. Let us take a concrete example [6.t-st; The design of a safety blow-off valve for a gas container (Figure 6.121) rrr be checked for possible design faults during the conceptual phase. From r f h,:1!!y, Pno* /, o!-4r/ line condition 0perating Valvelunction - Faults - normal cl0sed ^^,--l I lul I lldl - closed reaKage 00esnol 00esnol o0en close 6.I22.Operating conditions, valve main functions and faults of the safety valve pipe Blow-ofl Figure 6.121.Safety blow-off valve for a gas container specificationand the function structure,it is possibleto specifythe operar conditions depicted in Figure 6.122. The blow-off valve is intended to open the operating pressure,pop, eXCe€ds1.1 times the nominal working pressrl pno-, and to close when the container is again at nominal pressure: The n functions are therefore 'open valve' and 'close valve'. The overall function c also be described as 'limit pressure'. Let us now assumea possible failure ol' overall function, namely'valve does notlimit pressure'(Figure 6.I23). The vi functions shown in Figure 6.122, and their timing, are negated. They havc OR relationship with the overall function. Each fault thus identified is rrc investigatedin terms of its possiblecauscs(Figure6.124).The fault wc chosento investigatein more detail is 'doesnot open'. An identifiedcausemav havcto bc ussociatcd with furthercauseswith whit it has an OR or AND rclationship. tnd whlch mny hovc to bc of Figure6.123.Construction fault-treebasedon faults identifiedfrom Fisure 6.122 ingly. On the basis of the information gained from fault-tree analysis,the is able to revisethe specification(Figure 6.I25) before he proceedsto embodiment phase. As a result, his design will be greatly improved and tial faults avoided. r second example concerns the embodiment phase. Possible faults of an valve (Figure 6.126), are shown in Figure 6.127 and the appropriate production,assemblyand operationremediesin Figure6.128. that fault-treeanalvsismust.bv its verv nature, must.however.be stressed confined to important design areasand critical processes.What is essentialis thc dcsigncrshouldlearn to usethis approachby rote. In other words,he and look for potential hlbitually ncgatcthe identifiablesub-functions 300 6 Embodiment dcsilrr Coping with design faults, disturbing factors and risks 301 Figure 6.126. Sectionaldrawing of an alarm valve Figure6.124.Detailfrom completedfault-tree(Figure6.123) for the fault 'doesnot open' Valve cone blocked ication Specif tot blow-cffnlve Salety (valve headwithplanesealing Valve surface without taper) | ,9.73 N0rigidj0intbetween valveheadandspindie Easy maintenance orexchange 0fsealing surfaces Setting pressure Correct notmarked unknown Settlng too Force/preload easilymoved miscalculaled [,lovement of preloadscrew nolmeasured correctly W r o nsgp r i n g Dimensional errors in components Blow-cff aperture blocked Erosion dueto wrongmaterial Corrosion Seleclion ol wrongmaterial Inclusion ol wrongmaterial Spring sticks to block Seatdistorted bycreep Temperature toohigh Clearance toosmall fit Faulty of componenls Deformation Covered byother components B owoffpipe(if present) blocked Vatve seat blocked Figtre 6.127. Fault-tree analysisfor alarm valve (Figure 6.126) 25 Valve liftlimited 26 Damping of valvemovement 27 Insta lationin a c osed,ice-proof area 28 frlction N os l i d i nsge a l sa,v o i d Influence of disturbing factors fo01-pr00f (eg diflerent Ensure mounttng above approach should be applied not only to the examination of faults but to the search for potential disturing factors. Failures are often due to the r. According to Rodenacker [6.169], disturbancescan be causedby fluctuaof the input values, that is by qualitative differencesin the flow of material, rgy or signals entering the system. If these have a deleterious effect on the result, they may have to be offset by special design features. Di:;turbanceJ may also result from the function structure if the interrelaip l-rctwcensub-functions is ambiguous, or from the working principle if plrysicirl cl'l'cct is not correctly anticipated. The selected layout and form ,f.r irn(l tlrc associatcclmaterial propcrtv Iluctuatiottsand tolerqnccsneeded 29 flange sizesfor inletandoutlet) x) Requirements wererevlsed afterconsttucti0n 0l laulttree liigrrrc (r.125 I{cvision of sgrccilicltirrn irltcr fnttll.trcc rturlytis by reference to the headingsof the checklist, namely working principle, t and form design, etc (Figure 6.2). 302 6 E m b o d i m e n t t l es r r r l Coping with designfaults, disturbing factorsand risks 303 .2 Designing for minimum risk ite provisionsagainstfaults and disturbingfactors,the designerwill still be Blow-off aperture blocked A,R Testin setposition D Desrgn P Production A Assembly (useand 0 Operation matntenance) R Record required lockfor D Provide adjustment screw D Check choice A, R Check environment of material ol valve P,R Check D,0 Forbid on useof production pipe (choice blow-otf of materials D Avoidcovering andfits) holewithballor D,RFixtemp conefilter pointing range A Install valve 0, RCheck operating Iemperature D Spring mustnot stickto block Figure 6.128. Various corrective measures as a result of the fault-tree analysis (Figure 6.127\ with gaps in his store of information and with evaluation uncertainties-for nical and economic reasons?it is not always possible to cover everything h theoretical or experimental analyses. Sometimes all the designer can hope do is to set limits. Thus despite the most careful approach, some doubt may ain whether the chosen solution invariably fulfils the function laid down in specification or whether the economic assumptions are still justified by the ly changingmarket situation.In short, a certainrisk remains. One might be tempted always to design in such a way that the permitted limits not exceeded, and to obviate any impairment of the function or early mage by running the equipment below full capacity. The practical engineer that with this approach he very quickly comes up against another risk: the solution becomes too large, too heavy or too expensive and can no longer pete in the market. The lower technical risk is offset by the greater economic Coping with risks d with this situation,the designermust ask himselfwhat countermeasures can take-provided, for production and assembly may not have the effects they were assume(l have. Finally, the influence of external disturbing factors such as temperatutc moisture, dust, vibration etc cannot be neglected,and measuresmust be take to avoid potential faults. 3 Procedure All in all, the analysis of faults and disturbing factors and their eliminat should be conducted along the following lines: -Identify, and then negate functions. -Look for possible causesof faults in ambiguousfunction structures,such o imperfect working principles; o imperfect form designs: o imperfect material, energy and signal flows; o abnormal factors causing undesirable reactions of the system in respcct s t r e s s e s .d e f o r m a t i o n s .s t a b i l i t y . r e s o n a n c e .w e a r . c o r r o s i o n . e x p a n s i r s e a l i n gp o w e r , s a f e t y .e r g o n o m i c s .p r o d u c t i o n . q u a l i t y c o n t r o l . t r a n s p o r l operation and maintenance (Figure 6.2). - Determine what conditionsmust bc mct for faults to ensue(for exilnll AND or OR relationships). - Introduccappropriatc dcsignrcmcdicsby improvingon theoriginalsolulion, <lr hy clcvisingchccksof production,arrcmbly, tr$nsport,opcratitln iutd mnintcntncc.On thc wholc it in bcttor to lmprovcon thc origilralsohriion of course, that the solution was carefully chosen in the place and the appropriateguidelineswere scrupulouslyfollowed. The essentialapproach, which we shall be examining in greater detail, is that designer must, on the basis of the analysisof faults, disturbing factors and spots,provide a substitutesolutionagainstthe possibilitythat the original tion misht not cover all uncertainties. In the systematicsearch for solutions, several solution variants were elaboand analysed.To that end, the advantagesand disadvantagesof individual tions were discussedand compared. This comparisonmay have led to a new improved solution. As a result, the designer is familiar with the range of ible solutions; he has been able to rank them and also to take stock of the nomlc contralnts. principle, he will select the cheapestsolution, provided only that it has icient technical merit for, though it may be more risky, it will afford him ter economic leeway. The chancesof marketing the resulting product, and of judging the validity of the solution, are greater than those of marketing lier product, which might jeopardisethe entire developmentor, becauseof 'riskless' design, cannot provide information about performance limits. While he is well advisedto adoptthis strategy,the designershouldassiduously reckless developments that might lead to damage, breakdowns and a dcal of unnecessaryirritation. If risks cannot be eliminated by theoretical or experiment in good time or with justifiable outlay, the designermay f<rrccdto opt for the cheaper and riskier solution, but he should alwayskeep a costly,lcssriskyaltcrnativein reserve. 305 C o p i n g w i t h d e s i g nf a u l t s . d i s t u r b i n g f a c t o r s a n d r i s k s 304 To that end, he must develop the less cost-effective solution proposllr elaborated in the conceptual and embodiment phases into a second or thirtl solution reservedfor critical design areas,and ready for immediate use in casctrl need. Provision for such development should be built into the chosensolution ll the latter should not meet all expectations,it can then be modified, if necessitrv step by step, without great outlay in money and time. approach not only helps to reduce economic risks for :r ihir ryti"-atic also to introduce innovations one at a time, and to provitlt' but tolerable outlay, of a detailed analysis their performance, so that further developmentscan lrc made with minimum risk and at minimum cost. This approach must, of cours(', be coupled with a systematic follow-up of the practical experiences gairr"l through it. By designingfor minimum risk, the designerthus tries to balancethe technicrrl againstthe economic hazards and so presentsthe producer with useful experi' encesand the user with a reliable product. t could be obtained if forced convection was substituted for natural vectioncooling(Figures6.I29b and 6.130). This raised the difficult question of whether natural convection cooling would rtheless meet the required operational conditions and, if not, whether the elaborate and more costlv alternative with its additional cooling circuit uld be accepted bv the customers. 'minimum risk' decision-that is to construct the housing in such a way The t either cooling system could easily be used-helped the designer to gain rience for only a small increase in cost. / I Hemp-packing / I c {fll5'J - Theory Experiment 2 Examples of designing for minimum risk Example 1 A study of possible improvements in the performance of a stuffing box shorr..'tl that, tb inirease the sealing pressure and the iurface speed, the resulting frictional heat on the shaft must be removed rapidly in order to keep tltC 1 d> temperaturein the sealingareasbelow the limit. .Ib that the packingringsbe mountedon the shaft ttrat end, it wassuggested as to rotate with it and rub againstthe housingrather than the shaft. The hc generatedby friction could then be extractedthrough the thin wall (Figtt e.ngd. Theorticaland experimentalstudiesshowedthat a markedimprov I / lL- +Fffi, :- 0 .< _2- b h -4 Figure 6.129.Cooled stuffing box in which the packing revolves with the shaft. The appropriate designof shaft and pressring ensures the internal connection ofthe packing rings; a very short, heat path facilitates good heat extraction; (a) heat extraction by natural convection currentsin the surrounding medium, dependent on the prevailing air flow ( b ) h c a t c x t r i t c t i o t lt r y f o r c c t l c(Ilvcclion tluc t() scpilriltc cortlingrir flow crtsuring hllhcr flow vckrcilics nnd lncremcd hcnl cxt rtct ion ?3 V4 Figure 6. 130. Theoretical and experimental temperature determinations at the seal plotted against the peripheral speed on the shaft (a) Layout as in Figure 6.I29a (b) Layout as in Figure 6.r29b m/s (c) Conventional stuffing box with packing attachedto the housing ple 2 the development of a series of high-pressure steam valves operating at peratures of more than 500"C, the question arose whether the customary hod of nitriding the valve spindles and bushes should be retained despite the that the nitrided surface expands with temperature (thereby reducing the ial clearance), or whether very much more expensive stellite hard facing ld have to be substituted. When the problem first arose,there was a lack of uatc information about the long-term behaviour of such layers at high 'minimum risk' solution adopted was to selectthe wall :raturcs.The of the valve spindleand bushesso that, if and thc clinrcnsions krrcsscs 306 6 E m b o d i m e n t ( l e\ r t n 307 Coping with designfaults, disturbing factorsand risks necessaryand without changingthe other components,stellite-treated pirrtt could be substitutedfor the otherswhenevernecessary. As it turned out, tlrc operatingtemperaturerangewas considerablylower than had beenanticipaterl. so that nitriding provided a satisfactorysolution and also helpedto identify tha operationallimits. Once theselimits were known, the more expensivesolution could be reservedfor more demandinsconditions. Example3 Reliable design calculationsfor large machine parts, particularly in onc olf production,dependon the analyticalmethodsand the postulatedconstraint\, It is not alwayspossibleto predict all characteristics with the necessary degrt'c of accuracy.This applies,for instance,to the determinationof the criti,rrl whirlingspeedsof shafts.Often, it is impossibleto predictthe preciseflexibrlrry of the bearingsandfoundations.However,the differencebetweenhighercritirul whirling speedsin high-speedinstallationsis small in the rangeof flexibilitrtr normallyencountered. In the situationdepictedin Figure6.131,'minimumr isk' designcan once again be applied to advantagebecausethe spacingo1 tho Figure6.132 Figure6.133 e 6.1,32. Support which, by selecting different spacers, allows the distancesbetween bearings to be varied. rre 6.133. Plain bearingswith laminated springs1, allowing adjustmentsof ibility (BBC). (Laminated springs also have good clamping properties thus narrowing critical range) g the many suggestionsput forward for a device to wind a strip into a layeredring, two seemedparticularlypromising(Figures6.734a and b). solutionshownin Figure 6.134ais the simplerand cheaperbut alsothe ier of the two, becauseit is not certain whether the inner rotating mandrel 1 is invariably able, despite the increased friction produced by the knurling the pressure of the springs 2, to move the strip 3 forward. -B e solutionshownin Figure6.I34bis lessrisky, because the pressurerollers Theoretically Flexibility of predicted range bearings and which cannot foundations befurther re0uce0 Figure 6.l3l. Critical whirling speeds(qualitative)for a shaft plotted againstthc flexibility of bearingsand foundations bcarings,which has a major influcrrccon the critical speed,can be acl.j (Figurc6.132).Intcrposcdsprirrglaminations (Figurc6.133),morcovcr,llk altcrationof thc cffcctivc l'lcxibiliticn.Both measurcs,takcn tugctlrcr scpuratcly,will produccthc requircdcffcct to that thc sccondor third critir whirlingspccdctn bc clinrinnlcdfnrm thooporrtln3ipccd rtngc ol'thc nrlclr to the ends of the springs and the feed roller 5, which moreover can be riven, make the advance of the strip more certain. This solution, r, is the more costly of the two, and also more susceptible to wear use of the greater number of moving parts 'minimumrisk' solutionis that shownin Figure6.I34a,but with a feed-in r as in Figure 6.134b, and arranged in such away that, if need be, it can be n without alterationof the other parts (Figure6.134c). is additional element proved essentialwhen the machine was tested, and readily available. large electrical machines, fans attached to the shafts ensure the efficient ing of the windings and the laminations. quantity of air involved cannot, however, be predicted with accuracy of thc uncertain inflow and outflow relationships.Hence, in the first s. thc blaclcswcrc maclc adiustable to facilitate the correction of the air 308 6 Embodimentdcsrlrr 7 Evaluating embodiment designs 309 6.135.Blades of a coolins fan in an electricalmachine: the anele can be altered to djust the air flow (AEG) slot for balancing weights; b grub screw for fixing the blades; c rotor; d blade 7 Evaluating embodimentdesigns 63 F i g u r e6 . i 3 4 . (a) Proposedwinding device ^1rotating mandrel; 2 pressuresprings;3 strip to be wound; 4 parts of the ejection mechanism ( h ) P r o p o s e dw i n d i n gd e v i c e 1 rotating mandrel; 2 springswith pressurerollers; 3 strip to be wound ; 4 parts of ejectit mechanism;5 feed-in roller loaded by spring and possiblydriven (c) Chosen solution by springs;3stripto be wound;5feed-inroller tensioned 1 rotatingmandrel;2pressure spring6 and drivenby belt 7 throughput(Figure 6.135).With enoughexperience,it was then poSsiblc Here, asin Examlllc substitutea non-adjustable and cheapercastconstuction. the possibilityof makingcorrectionsensuresa 'minimumrisk' solution. The sameapproachcan alsobe usedfor morecomplexventilationsystcrlls. which the air quantities and pressurelosscscannot be predicted with accurir( shouldmcct ri to showthat the designer All theseexamples are intendccl thc l'irststcpbut ulsothc sccondor third, whiclrc not simplyby considcring oftcn bc donc at rclativclysrnilllco$l, Expcricncchus shownthat cmcrgc nlcasurcst() corrcct unlilrcsccnftrultr are mgny timcs morc costly itnd consunring. 5.8 we discussedthe subject of design evaluation. The basic procedures tlined there apply equally well to the conceptual and to the subsequent . As embodiment progresses,the evaluation will, of course, rest on more more concrete objectives and properties. In the embodiment phase, the technicalproperties must be evaluatedin terms the technical rating, R,, and the economic properties separatelywith the help the calculated production costs in terms of the economic rating, R.. The two ings can then be compared on a diagram (5.8.1.6). The prerequisitesof this approachare: That all the embodimentdesignshavethe samedegreeof concreteness, that is the sameinformationcontent(for instance,rough designsmust only be comparedwith rough designs).In many casesit suffices,while keepingthe overall perspectivein mind, to evaluateonly thoseaspectsthat show marked differences from one another.Oncethat hasbeendone,their relationshipto the whole, of course, must be examined,for examplethe relationship betweenpart costsand total costs. That the production costs (materials,labour and overheads)can be determined. If a particular solution introducessubsidiaryproduction costs and demandsspecialinvestments, then, dependingon the point of view (producor user's), these factors must be allowedfor, if necessary by amortisation. er's addition, optimisation can help to achieve a minimisation of production In und opcratingcosts. 310 6 Embodimentdesirrr If the determination of production costsis omitted, then the economic ratilrrl can only be evaluated qualitatively, as it was in the conceptual phase. In tht' embodiment phase, however, costs should, in principle, be determined more concretely. As we mentioned in 5.8.1, the first step is to establish the evaluation criteritt They are derived from: - the requirements of the specification: o desirable improvement on minimum demands (how far exceeded); and o wishes (fulfilled, not fulfilled, how well fulfilled); -the technical properties (to what extent present and fulfilled). The exhaustivenessof the evaluation criteria is tested against the headings t'l the checklist (Figure 6.136), specially adapted to the level of embodimcrrt attained. Heading Examples Function principle: in accordance withtheselected working Fulfilment risk,susceptibility to disturbances efficiency, Layout design Form design fits,scope lor modifications requirements, weight, arrangement, Space Salety Ergonomics Produclion protection Direct methods, industrial safety, ol theenvironment safety 7 Evaluating embodiment designs 311 rs described in7.l. They are instancesof the common case in which what to be evaluatedis not the overall design but designvariants in chosendesisn Fisure 6.137 shows the three variants of the bearins bracket to be aluated. In variant Vllhe bearing bracket is fitted directly to the front wall, as welded construction. In that case, the rotor can only be assembledvertically the top, which must be facilitated by the appropriate housing shape. life,wear, Durability, deformation, sealing, operating stability, resonance shockresistance, lVlan-machine relati0nship, handling, aesthetic considerations Risk-free methods, setting-up time,heattreatment, surface tolerances treatment, 0ualitycontrol Assembly Transport 0peration possibilities Testing Maintenance Costs Schedules repair Servicing, checking, andexchange adjustable, resettable Unambiguous, easy, c0mf0rtable, v3 6.137.Design variants for bearingbracket. Simplified diagram (AEG-Telefunken) packing transportati0n, means lnternal andexternal of despatch, properties, Handling, behaviour, corrosion operational consumption ol energy (economic rating) Evaluated separately date Production schedule andcompletion Variant I/2 consists of a separate bearing assembly of welded construction, ich is bolted to the machine base. The rotor is assembledhorizontally, with bearing bracket unbolted. In variant V3, the bearing bracket is incorporated in the end plate as a cast component which replaces the front wall, closes the lateral housing rturesand is bolted to the housins. Figure6.136.Checklistfor evaluatingembodimentdesigns At least one significant evaluation criterion must be considered for euch heading, though sometimes more will be needed. A heading may only ltc ignored if the correspondingproperties are absent from, or indentical in, all tltc variants. This approach avoids subjective over-valuations of individual pro. p e r t i e s .I t m u s t b e f o l l o w e d b y t h e p r o c c d u r a ls t e p so u t l i n e d i n 5 . 8 . 1 . I n t h e c m b o d i m e n t p h a s e ,e v a l u a t i o nn l s r lc o n s t i t u t c sa n e s s c n t i a sl e a r c hl o r wcak spots. As an examplc of thc cvaluation of dc*i3nr ln thc embodimcnt phasc, wc shlll c o n s i d c r t h c b c i r r i r r gb r i r c k c t l i r r u s c w i t h l h ! h o u * i n g o f t h c r i r n g c o l ' c l c c t n c The evaluation of the technical properties (technical rating) is made with the of an evaluation chart, Figure 6.138 (see also Figure 5.51). Since evaluation teria for such desisn areascannot be fullv derived from the specificationof the ire machine, it is useful to consult the checklist (Figure 6.136). The following ings of the checklist apply to the case under consideration: tion- Lay o ut design- Form design bearing spacing permits higher speeds,quieter running and shorter ovcrallconstruction. tligh stiffne.rsof the loaded areas of the bearing and housing reduces dcformltionof thc air ventsand leadsto smootherrunnins. JIL 6 Embodimenl rlcrrr,rr 313 7 Evaluating embodiment designs Good damping characterlsllcsrender the machine less susceptible to damaging vibrations. Ergonomics o t @ @ o O O O S S> @ ES ll * =: s'= L- O Incorporatior? of the bearing into the housing reduces the danger of injury through projecting edges and gives a more satisfactory form design. ILS ides minimising the production costs (which are assessed separately in >_ o E += = o o o rmining the economic rating) the designer should, because of limited uction capacity: = o O E O @ = S @ O il O O N N O O Sq =- ll ^ ^i'F6 try to make do with few productionfacilities; and aim for a high level of sub-contracting. o- so ity contr ol-A ssembly-Tr ansp or t S =s .F= @ E B o O = E E = E B o c trr 6 E E E o , o = N o s- l l N @ E = o o o E , =E E r= E9 Ea = oo =E il^ =- s= N o O E= >s o n-Maintenance r= o = B o oil i l N E E c f design of the bearing bracket influences the assembly of the rotors, the tting and inspection of the air vent, and the assembly of the seals and the end s of the housing. The rotor assembly, in particular, also influences the sport facilities because of its great weight. Hence the desisner should aim for easeof assembly and inspection bo 'a -:qJ (D o c a E E o c o = N N O O >6 tqo oYAta o6= E= @ O '= o c oo E = O o E E L E c o o a o O a O O o <6 O c o o o,^ o c c c E -E c E - U i6o )= 6 ttt o o F >R == I- >ll ('iJ O ooS E result. For a full assessment,the technical and economic ratings of the three variants o?- .-PE @6 z c 5rr^ ll servicingof the sealsbetween housing and shaft, and also the inspectionand ning of the motor windings, is affected by their accessibilityand the ease h which they can be dismantled. Hence the designer should aim for an easy exchange of seals and simple of the interior of the motor. In Figure 6.138, the three designvariants have been evaluatedwith the help of criteria, and the appropriate parameters values entered. Weighted values obtained by means of weighting factors. The evaluation shows that the nical ratings impose the order V3-V1-V2, Vl and V2being relatively close each other. Variant Iz3 has not only the highest technical rating but also the t balanced value profile. The economic rating is obtained from an analysis of the production costs. To 'ideal ate the ratings. the costs' were set at 80 per cent of the cheapest iant. Figure 6.139 gives the economicratings of the three variants obtained as EE ao Variants N Figurc(r.139.Economic ratingof the threedesign variantsshownin Figure 6 .t 3 7 Percentage production costs Economic ratingfi. Vt V2 106% r 0 0% 0,75 0.8 V3 11t 0,16 3r4 6 E m b c l d i m c n tr l r s r r , r have been incorporated in a rating diagram (Figure 6.140). The reader will sr.,. that though variant 73 has the highest technicalrating, it is also the most cosrlr In the case under review, the costs of the evaluated design area amount to ollr 5 per cent of the overall cost of the motor, so that the economic ratings of tlr,. cheapestand most expensive variants alter the overall costs by no more thrrrr about 3.3 per cent. Moreover, with variant v3 the cosis of the housing ,,,,. Developingsizerangesand modular products t,0 .1 Sizeranges V? 0,8 V1 I | 0,6 v3o 0,4 Figure6.140.Comparison of the technicaland economicratingsof the designvariantsshownin Figure6.137 0,2 0 0,? 0,4 ^ 0.6 0,B 1.0 4+ slightly reduced. In view of the pronounced effectsof deformation and dampirrr: on the smooth running of the machine, variant V3 was judged to be the best ol the three despite its relatively low economic rating, the more so as cast enrl plates were known to produce good results in practice. 1erangesprovide a rationalisationof designand production procedures[7.31]. For the manufacturer they have the following advantages: The design work can be done once and for all and can be used for a host of applications. The production of selected sizes can be repeated in batches and hence becomes more cost-effective. Higher quality is possible. is implies the following advantages for the user: competitive and high quality products; short delivery times; and easy acquisition of replacement parts and fittings. Disadvantagesfor both manufacturer and user are: limited choice of sizes, not always with optimum operational properties. By size range we refer to technical artefacts (machines, assemblies or ponents) for a wide sphere of applicationsthat: fulfil the same function; are based on the same solution principle; are made in varyingsizes: involve similar production techniques. If, in addition to the range of sizes, other associatedfunctions have to be ted, then modular products (see 7 .2.2) will have to be developed side side with size ranges. The development of size ranges may be original or on an existing product but must, in either case, be carefully graded. We to the initial size as the basic design and to the derived sizes as sequential [7.31]. In the development of a size range, similarity laws play an essential, and l-geometric preferred numbers a useful, role. Similarity laws lnrctric similarity ensuressimplicityand clarity of design.The designer lws, howcver, that technical artefacts stepped up in geometric proportions srl-callcdpantograph constructions)are not satisfactoryexcept in very rare : s . l l t p i r r t i c u l a r ,p u r c l y g e o m e t r i c a lm a g n i f i c a t i o ni s o n l y p e r m i s s i b l ew h e n t16 7 D e v e l o p i n g s i z e r a n g e si i n d m o d u l a r p r o ( i r , I ;imilarity laws permit, which should always be checked. These laws are usr',| i,/erysuccessfullyin model testing 17.28, 1.30, 7 .32,'7.361.In general, howevt r lhe development of size rangeshas a different objective from model technologr, namely to achieve: -the same level of material utilisationl -with similar materials if possible; and -with the same technology. lt follows that, if the function is to be fulfilled equally well throughout the rani', lhe relative stressesmust remain the same. We speak of similarity if the relationship of at least one physical quantitv rrr the basic and sequential designs is constant. It is possible to define bu'r, similarities with the help of the fundamental quantities length, time, forr,' quantity of electricity (charge), temperatureand luminous intensity (Table 7. I t Iable 7.1. Basicsimilarities Table 7.2 lists important similarity relationships in the development of size anges for mechanical systems.They are by no means exhaustive and must be pplemented from case to case, for instance in bearing developments bv mmerfeld's number and in hydraulic machines by the cavitation number and pressureindex. Table 7.2. Special similarity relationships Similarity Kinematic atic Dynamic Similarity Basic quantity Invariants Geometric Temporal Force E,lectrical Thermal Photometric Length Time Force Charge Temperature Luminous intensity E 1 , : L 1 lL 1 y 9t: tltt cpp: FJFl qo: QJQo E,1: $ylt)11 et:Jrllo Thus we have geometric similarity if the ratio of all the lengths of any sequentnl design to all the lengths of the basic design is constant. Here, the norr. dimensionalparameter to be held constantis gl: L1lLs, where L1 is any lengtlr of the first member of the size range (sequentialdesign);and Ls the corresporrtl. ing length of the basic design. In the same way, we can describesimilarities irr time, force, electricity, temperature and luminous intensity. If two or more of the basic quantitesare in constantproportion, then we hirrc special similarities. Now, model technology has defined dimensionlessp:ur. meters for important and recurring similarities. Thus, in the case of sinrrrltaneous invariance of length and time , we have kinematic similarity , and in t lrr,. caseof simultaneous invariance of length and force we speak of static similaritt , A very important similarity, namely dynamic similarity, appears whcrr ;r constant force relationship is combined with geometric and temporal similrrr ities. Depending on the forces involved, we arrive at different dimensiorrl,'sr parameters. Thermal similarity deservesspecialmention because,in the case,'l geometrically similar size ranges and the same utilisation of materials, it canrr,rl be squared with dynamic similarity. * F u n d a m e n t aplh y s i c aql u a n t i t i e a s r e a s l i s t c di n t h c o r r g i r r l (l l c l r l u n t c x t . ' l - h cl r ; r ' , r , p h y s i c aql u a n t i t i esse l e c t efdo r t h c S I s y s t c rtni i l l c rs l i g h l l vr r r t t l ,l o n gw i t ht h c i l b a s i ct r r r t . ( r r r c t r c t) il r r r t ' ( s r ' c o r trtul ); :r s(sk i l r r g t i t r tct l)c: c t l i ct ' r t t tr, l s h o w ni n b r a c k c t su,r c :l c r r g t h ( q r n r l t c r c ) ;l h c r r l o t l V r r i r r r r itcc r r r l l c r ; r t r r r (ck c l v i r r l ; l t t t r l l t t t t t i t t o t t ist t t t ' t t s i t v( c l t t r l t ' l l r ) I l r ' t l i l l c t c r t c c s t l o t t o l ; r l l r ' t ' tl l t u n t i t t c i D l r ' :r l t ' r t t t l r e t l 317 7 [ Sizc ranges Invariants gr,9t Definition Description "O;Sry 9r-.9t Hooke Ho: 9L' 9r' 9F Newton Nr: F n'. r, Relative elastic force -!- Relative inertia Q'vt'L' Ho C a u c h *y Ca:';;: Froude Fr: o. v2 T Inertia force/elasticforce v2 Inertia force/gravitational force gr NN ** E e'8'L Reynolds Re- E,lasticforce/gravitational force L'''Q q Inertia force/frictionalforce in liquids and gases 9L,90 Biot Supplied or removed/conducte quantity of heat 9 r ' 9 r po Fourier Conducted/stored quantity of heat In some texts, we find Ca : v . t/ pt E . This is appropriate if Ca is intended as a velocity relationship. Not named. Similarity at constant stress In heavy engineering systems,inertia forces (forces due to mass, acceleration etc) and elastic forces resultinq from the stress-strain relationship play a p r c t k l r r r i n a n tr o l e . l l ' t h c s t r c s s c sa r e t o r e m a i n c o n s t a n t throughout a size range, then 0-t'l:-cortstlrnl. 318 7 D e v e l o p i n g s i z e r a n g e s a n d m o d u l a r p r o t l r r ,L In that case the stressparameter becomes: ,r^:o':1!: os fo Eo 3le Size rangcs le 7.3. Similarity relationshipsfor geometricalsimilarity and cquirl strcsses: tlt'pcrrrl, : of important quantities on length r W i t h t h e s a m em a t e r i a l , t h a t i s a t E p : E t l E o : 1 . th Ca: Qv2lE:constant and the same material, that is a : E :const we need: t; : cp,: €11€1y: l, orup, : + t, or cpzr: er z L t )L r With this so-calledCauchy condition, all changesin length must increasc rl .he same ratio as the appropriate lengths. The elastic force parameter th( r) )ecomes: Speeds, n, rr.r nding and torsional critical speedsftr,, a", trains e, stresseso, surface pressuresp due to inertia and elastic forces, speeds y stiffnessess, elastic deformations A L otAt : ezt , ooAo with eo : gu.cPp,: I and cpt: g2y fttIat rt0a0 eights W, torques I, torsion stiffnesses,s, vFt: nd moments of area I, J QtVtat gts te: The utilisation of the materials and safety level are only constant if the influence of the dimensions on the material propertils can be ignored. with Qp: QrlQo: l, Ev : Vt/Vo: tlltt, Qr." (Pra assmoments of inertia I' , J' QoVoao (Ptt Q12 rsP M o m e n t sM , , M, l'he inertia force oarameter is: 9r." rarnss, stresses o, surfacepressures p due to gravity orces F (Ppp.: f, : c()r.tst n the caseof geometricalsimilarity the following relationshipsoccur : rtt rnd Size ranges developed in accordance with these laws are geometrically similar nd provide for the identical utilisation of the materials.Such deveioDmentsare ible whenever gravity and temperature have no decisive influence on the sign. If they have, the use of semi-similar seriesis advisable(see 7.1.5). L{A et r "^_ rrh_A ve have Decimal-geometric preferred number series Eet: ELlg2, A. dynamic similarity, that is a constant ratio between inertia and elastic forr* vith geometric similarity, can only be attained if cp,: Ey: ?pp: Qzr: Qpr: ELlrPt-: E2L .Ience the velocity ratio becomes: cg,: grlclt: qrlEr: I With the same material, the same result can also be derived from the Cauclrv rumber (Table 7.2), for when p and E remain constant then the dynarrrrc ;imilarity will only remain constant if the velocity v also remains constant. For all important quantities such as power, torque etc, ancl u'itlr pr:ch:const. andcpr:gr:gn:9,: l , i t i s n o w p o s s i b l ct o c s t a b l i s ht l r . ' ; i m i l a r i t y r e l a t i o n s h i p ss h o w n i n T a b l c 7 . 3 . I t s h o u l d b e r c m c m b c r c c lt h i t t t l t c u t i l i s i r t i o r o t l t l t c r t t i r l c r i a l sl n c l t h c s i r t c t r c v c l o n l y r c r t t a i r tc ( ) n s t i r n ti l t l t c i r t l ' l u c r t c o c l t l t c r l i r n c n s i t ) n (s) l l t h c r r r l r l c r i , r l r l o p c r t i c sc i r r rl r c i u r r o r c t tl h r o r r g l r o r tr ltr c r i r c t i l n F c . we are familiar with the most important similarity relationships,we still to determine the best method of choosing the individual steps of a size nge. Kienzle [].24,7.251and Berg [7.5 to 7.9] have arguedrhat a decimaltric series is the most useful. A decimal-geometric series is based on multiplication by a constant factor cp is developed within one decade. The constant factor E determines the step of the series and can be expressedas: E : \EJ-q,: \,M re n is the number of steps within a decade. For 10 steps, the serieswould have a factor: ,o:Vlo:1.25 i s c i r l l c cR f 1 0 . T h c n u m b e r o f t e r m s i n t h e s e r i e si s z : n * 1 . 'l'lrblc 7 , 4 s c t s o u l t h c m : r i n v a l u c so f f o u r p r e f e r r e dn u m b e r s e r i e s[ 7 . 1 2 ] . 'l'ltc t t c c t l l o t g c o t n c l t ' i cs c i r l i n gi s o t l c r r l i r u n c li n c l a i l y l i f c a n c l i n t e c h n i c a l 7 Developing size rangesand modular t.4. Main values of preferred numbers [7.121 (logscale) Diameter d in mm 50 Basicseries Basic series JLl 7 1 Size ranges R10 R20 R40 R5 R10 R20 R 4t) 1.00 1.00 1.00 1.06 l.r2 1.18 r.25 r.32 1.40 1.50 4.00 4.00 4.00 4.00 4.25 4.50 4.75 5.00 5.30 5.60 6.00 1.60 1.70 1.80 1.90 2.00 2.12 2.24 2.36 6.30 12s 140 160 180 200 17+ zS0 zg0 315 355 400 450 s00 60 58 56 ct t.t2 1.25 1.25 r.40 1.60 1.60 1.80 2.00 2.00 1 2.50 1A 2.50 2.80 3.15 3.15 3.55 4.50 s.00 5.00 5.60 6.30 6.30 1.rc 8.00 8.00 900 6.30 6.'70 7.r0 7.50 800 8.50 9.00 950 2.50 2.6s 2.80 3.00 3.15 3.35 3.55 3.75 :e. The resultingseriesconform with the Weber-Fechnerlaw which re physiologicalsensationproduced by a stimulusis proportional hm of the stimulus. the [7.33] has shown how, in the development of friction drives, t er instinctively chooses the main dimensions by means of geomctrit ;. Our own work on turbine shaft oil scraper rings has confirmed tltt ;s. In Figure 7.1, shaft diameters are plotted using a logarithmic str t the number of newly designedoil scraperrings (or rings on order) ovt'r of 10 years. The results show that there were 47 diameterswith pcirkr r r l e s sr e g u l a r i n t e r v a l s ,w h i c h c l e a r l yd e m o n s t r a t e sa g e o m e t r i c asl c i r l i r r v e r , t h e n u m b e r o f n o m i n a l s i z e sw a s d i s t u r b i n g l yl a r g e - s o m e d i f f c r c t l f c w m i l l i m c t r c s a n c lg a v c r i s c t o v c r y s m a l l p r o c l u c t i o nb a t c h e s .[ - t r c k r r r c 7 . 1 a l s os h o w s ,i l ' l l r c f ' c r r c cs li z c si t r c s c l c c t c cwl i t h t h c h c l p < l tt h c l { . t l r c n u r r r h c l o l v i r r i i r n t sc i r n b c r c c l u c c t l o l c s s t l u t n I t i t l l , g i v i r r l r l c r i r l r l y n r o r c b i r l i r r r c c ti rl r r r lh i g h c r r c ( p l i r c n l c t l lp c r t t t t t t t i t t i tsl i z c . l l r r t l t h c c r s c l r r l s c l sl r r c l rp r c l c r r c r l n u n r t r c r st l c l i b c r i r l c l y .u t l l u c ht t t r t r cs t t i t r t t r l rt 'r r c worrltlhlvc crttcrgctl ty itscll, 52 50 48 46 lrl, L7 /.0 38 36 34 37 30 o 2B 5 26 2L 22 70 18 to l4 12 10 I 6 L 7 ?LN 727 2io 340375 3bo Figure 7. I . Frequencyof seal di ameters d of scraper rings for tu rbine shafts; continuous line: actual situation; broken line: suggestedsizerange The use of preferred number series thus provides the following advantages .12): Appropriate scalingleads to the selectionof nominal sizesin accordancewith demand. The finer series have common numerical values with the coarser. With proper gradation it is possible to approximate an arithmetical series. This facilitates jumping from row to row and hence provides the different steps needed for matching the distribution of the market requirement. The preferred number series contain both decimal powers and also doubles and halves. -There is a reduction of the dimensionalvariants by the choice of dimensions bascd on prcferred numbers with a consequent saving in manufacturing i n l r r r c t i o n s ,c q u i p m e n t a n d m e a s u r i n gt o o l s . - Sincc thc llrocluctsancl qr.roticntsof terms of the seriesare in turn terms of a g c o r r r c l r i c i rsl c r i c s ,l r n l r l y s c si r n r lc i r l c u l l t i o n sr c c l u c cr n a i n l yt o m u l t i p l i c a t i o n l r r r r tl l i v i s i t r r rA s . r i s c t l n l i r i n c riln t h c l l r c t ' c r t c rnl t t n t b c rs c r i c sw i t l t a g o o c l -)zL - - approximation, geometric gradation of component diameters will generatt. circumferences, circular areas, cylinder contents and spherical surfaces thlrt are, in their turn, terms of the preferred number series. If the dimensions of a component or of a machine are terms of a geometricrrl series, then linear magnifications or diminutions will give rise to preferrctl numbers in the same series provided, of course, that the magnification or diminution factor is also selected from the series. Automatic growth of the size range will be compatible with existingor futurr' ranges. particularrangeof sizesinto severalsets. If we define a characteristic number N of a ranse suchthat: Greatest term of the ranse 3.r5 I ?.5 2 R a t.0 1 1 Er In general, when trying to rationalise a product size range, the designer uill select his increments once and for all. To that end he makes an approprilrrr. selection of step sizes, for instance in respect of power and torque. T-lt;rr selection can be based on several considerations.First of these is the markcr situation, which as a rule requires small incrementsso that the varied demarr,l. of customerscan be met most effectively. The second considerationis efficierrr design and production. For technical and economic reasons,the selected step sizes must be fine enough to meet the technical demands (for instanct,. power), and yet coarse enough to allow large-batch production based on;r simplified range. The selectionof optimum step sizesthus involves an integratt'tl approach to the'market-design-production-sales' system, and requircs information about: -market expectations(sales)in respect of individual sizes; -market behaviour in respect of simplified ranges and the resulting gaps, -production costs and times of the various step sizes and the effect on thc overall production costs; and -properties of each product in the size range. Since the optimum selectionof step sizesmust be based on all the factors wo have mentioned, it is not always possibleto opt for a constant step factor; mo16 often technical and economic considerations will demand the breakup ol N: I 6.3 5 El 7.1.3 Selectionof step sizes : rnn Smallest term of the ranse wheren is the numberof the stepsin any particularrangeand z: n * I is tlro numberof terms,then the factor: q:VN' Thc rangccan bc split up by mcansof a<'onstant r,>ravariublefac:tor,thirt ir, by stcps within and/or bctwccn coarscror fincr prcfcrrcd nunrbcrsclicr (R .5-R 4t)).1'hc rcsultingstcp charactcristics arc shownin l-'igurc7.2, 'I'ypc A hus u constuntfuctor (for instunccqt= 1.25corrcsponding to l{ l0; ovcr thc cntirc rungc. 323 Size ranges 7 D e v e l o p i n g s i z e r a n g e s a n d m o d u l a r p r o d L r tr ' 8s 1 2 3 4 5 6 7 8 9 1 01 2 3 1 5 6 7 8 9 r 0 1 2 3 l . 56 7 8 9 1 0 r 2 3 4 5 6 7 8 9 1 01 2 3 i - 56 7 8 9 1 0 * Termnumber l:i,Tlllflltrffiffi ffi TypeA TypeB TypeC trffiffi ffiffi Type D TypeE Figure 7.2. Step characteristicsofsize ranges[7.16]; (factors assignedto each step) In Type B, the lower part of the range is divided up coarsely (for instance g:1.6 corresponding to R 5) and the upper part more finely (for instance :1.25 corresponding to R 10). Suchdegressive geometricalproduct ranges should be used whenever a coarser grading for the smaller product sizes is ically justifiable.If a degressive gradingis ssed, it should alwaysbe on the combination of several preferred number series with decreasing ep factors and not on a constantly decreasingseries since, on the one hand, at adaptations may be needed can be made accurate enough and, on the other d, adherence to the preferred number series is advantageousfor the reasons have already set out. Type C has a greater increment in the upper range and is used if demand is ncentrated on smaller sizes. It is also known as a progressive geometrical ge. Type D has a smaller, and Type E alarger, step factor in the middle part of range. For simplicity, we can generally take it that the size gradation must be the r the greater the demand and the more precisely certain technical stipulas have to be met. A different gradation can be chosen whenever the market ands and without great design effort. Needless to say, the effects on uction must be taken into account as well. In grading, a distinction must be made between independent and dependent uantities. As a rule, the task itself determines which sizesmust be treated as pendent and which as independent. For example, geometric grading of the output may be advantageous for market reasons and grading of sizes by ferred number series for production reasons. If the two are associated by a r law (Figure 7.3, curve a) then both can be graded by a preferred number ics, cither with exponent p :1 (linear growth) or with p # 1 (non-linear h) (see 1.1.4). In Figure 7.3, the dependent and independent quantities vc bccn plottcd logarithmically. If the preferred numbers have the same rr. lhcn thc spacingis constant(Figure7.4) 324 7 D e v e l o p i n g s i z e r a n g e sa n d m o d u l a r p r o d u r t D5 I vI Dqs = D4 ri l t; n u34 D3 D u23 R r D2 D,, I / rf 2 ). lb.""' '" ,ro los l,o, P Ne g q [ m m-] (D) quantities. (1)anddependent For a power Figure7 3. Gradingof independent on the PN (preferrednumber)diagram(curverrI functionthereis a linearrelationship for othersthereis a non-linearrelationship t\z l'ro However, technical systems may not involve power relationships betu'crrt dependent and independent quantities. In that case, not all the sizes can l)c geometrically graded. Here the designer must decide, depending on the trrsk, whether he grades the independent or the dependent quantities in accordantc with a preferred number series. For economic reasons,it is often advisableto split the range into parts and lrr replace several sizeswith just one for each part (semi-similar size range). hrrt care must be taken to ensure that the stepped line needed for such gradatiorrrl roughly equivalent to the continuous curve. Figure 7.3 shows the resultirrg Sizc ranges stepped line for size relationships based on a power function (curve a ) and for non-linear relationshipswhich are not governed by that function (curve b). Independent sizes In, Izs etc have been assigned to the geometrically graded part of the rangesDtDz, D2D3 etc. This correlation is obtained by replacingthe DtDz, D2D3 etc with their geometric mean values Drz: \/il:D, and then drawing the stepped line accordingly. This is preferable to fixing the line intuitively, which is done far too often. It will be seen that the dependent relationships based on curve a once again result in a geometric grading of the steps,while the non-linear relationshipsbasedon curve b do not (in other words, the 1'- values are not geometrically graded). Here the designer must again decide for what sizesa gradation based on preferred numbers is still appropriate. Further deviations from strictly geometric gradings may, as we have already said, be imposed by manufacturing considerations.Practice has shown that it may be more economical to provide arithmetic or even irregular incrementsfor some component dimensions, so that, in a product size range, semi-finished materials, which are not usually geometrically graded, can be exploited more fully, or the manufacturing processcan be simplified (see 7.1.5). Even though grading based on preferred number series is generally advisable, the designer should not use it rigidly, but decide each case individually after cost analysis. Deviation from geometric grading will also occur if certain dimensions only have to be stepped, while others have to be adapted to specific customer mands. This is called a sliding arrangement [1 .15), and may prove most ive when the special dimensioning involved does not lead to a significant in production costs.Thus for the ball valve seriesdescribedin [7.15] the imensionsof the housing, drive shaftsand bearingsare firmly stepped,whereas e plugs and sealing rings have been given 'sliding' dimensionsfor hydrodynareasons. A similar approach is used in the design of turbines and thermal uipment 17.261. .1.4 Geometricallysimilar sizeranges the basic design, the choice of materials and the necessarycalculationsarc to nd, and if the nominal dimensions lie roughly in the middle of the intcndcd range, then nearly all technicalrelationshipscan be expressedin the gcncral 4..J'o 0 tr lll4-lll4 ^ -t'.--T- (v,=ts-gxo=!t-!tY Illxn | : cxP or tona=l l-Tt.0 +1 lg9rnr=n rl 1,25 3J54 tttr nlrt 63 I I I I 1 0l ? s F i g u r e7 . 4 . T e c h n i c a l relationshipsin the PN d i a g r a m ;n s t c p n u m b c r r n t h c f i n c s t u n d c r l y i n g/ ' N s c n c s ,c v c r y i l l l c r s c c l r ( ) r 'sr p r c l c r r c d n u r n b c ro l t h i s r c r i c s ; c v c r y i r r t c g irr l c x l ) o n c n tl c l t l s b : r c kt o l t t o l l t c t p t c l t ' rt c t l t t r t t t t l t t ' t logy : logc *plog,r. Every preferred number (PN) can be expressedby PN: l)mtn oy' log (PN) : mln hcrc nr is the step number in the PN range and n the number of steps within c r l c c a d c . I l c n c c t h e t e c h n i c a lr e l a t i o n s h i pc a n a l s o b e e x p r e s s e db y : tfly il '*lt - ftl,. ntl m,. ^ 326 7 D e v e l o p i n g s i z e r a n g e s a n d m o d u l a r p r o r l r r ,r The basic design is assignedthe index O, the first successivemember of tlr,' size range (sequential design) the index 1, the k-th member the index k. All relationships can now be represented by straight lines in a doublr' logarithmic graph, the slope of each line correspondingto the exponentp of llr,' technical relationship (dependence) (Figure 7.4). For simplicity, we entcr tlr,' preferred numbers instead of the logarithms and so obtajn a very practicalrl,' visual tool for the development of size ranges,as Berg [1 .7,7.9] has pointed orrr Every intersection represents a preferred number, and is always produced l,r lines with integral exponents. If the abscissagives the nominal sizex, then llr,' factor g*:xtlxo. In geometrically similar size ranges it is equal to the lenltlr factor cpy. Once the basic design has been fixed, all other magnitudcs dimensions, torques, power, speeds etc-can be derived from the knou rr exponentsof their physical or technical relationships(see Table 7.3) and can 1,,' drawn as straight lines with the appropriate slope (thus weight, ew: Et3, rrrll h a v e a s l o p eo f 3 : 1 ) . As a result, the main dimensions of the product can be expressedin diagrrrrrr form without the need for further drawings (see Figures 7.5 and7.6). Such data sheets enable the designer, starting from the basic design. trr provide the salesdepartment, the purchasingdepartment, the planning deprrrt 321 Size ranges 560 s00 ,I E BO = 160 ? 1t0 E rnn I B mm 6 =80 E 11 o v J =:: sll 3U l = - r4n5 I ,l *l= ll,5 _ 28 -E l "€ J: ti' -?n :]R . 16 e 14 I t?s '= 117 I I ? E s10 -9 c ; = 63 80 10012s160 d1+ '71 a63 5,6 63 80 100 125 160200 250 315 400 500 -tR10 R2A 4 -H! -1-- -1.-HB H5---H6-l--H7 -u5. -1 J x7 L-xB_ v5--.1*x5 C l a sosf s h r i ntki Id t o rl d t d : 1 1 + 0 . 3 % " dt_ r gure 7.6 Data sheetfor the gear coupling sizerange in the nominal dianteter ritngc (r, rresponding to the basic designshown in Figure 7.5. Dimensionsgeometricallysimilar ceptions: outer sleeve diameter D of the smallest member for reaions of stiffnessI ndard module s m are not stepped in accordance with preferred numbers ; special adaptatron of prtch circle diameter becauseof the demand for an integral, even number of tccth. The classof shrink fit is shown under the abcissa w i t h4 n o :, a t . , = r i o , - 12 . r - 1 1 - 6 32-.115 = Z n n fit d: Shrrnk ff=t r tosr" *S , , henceD:/5selecled-s=4mm fl==l? frl,/,,/. \r I / ,'']"' l , i g r r r e7 , 5 l l r r s i ct l r ' r r g rtr, r r l . l t ' l r(r' o u l ) l i n l \l r / ( , r . t t l l l ( .r,/ , - l { X l r r r r r r nt and the production department with crucial information on every size in J rangc. I t s h o u l d , h o w e v e r , b e r e m e m b e r e d that the measurements cannot be altsl.crrcclclircctly from the data sheetsto the drawings (which need only be r r l c o n c c i r r r o r c l c r h a s b e e n r e c e i v e d )unless the following factors have also b c c t t l i r k c r ri n t o c o n s i c l c r a t i < l r r : 7 D e v e l o p i n g s i z e r a n g e s a n d m o d u l a r p r o r l r r ,r r I Size ranges 329 1 . Fits and tolerancesare not in geometric step with the nominal sizes,the sizc t rl a t o l e r a n c eu n i t i f o r a d i m e n s i o nD b e i n g g i v e n b y j : 0 . 4 5 . l / D + 0 . 0 0 1 / ) , that is, the factor for the tolerance unit being determined by the relationshrgr Qi: Qtll3. Particularly in the case of shrink and interference fits but also of functiorr. determined bearing clearancesetc, the tolerancesmust, becausethe eluslrt, deformations tend towards er, be adapted accordingly. In other wor'(l\, smaller dimensionsmake more, and larger dimensionsless, severe deman,ls (Figure 7.6). 2 . Technological limitations often demand deviations. Thus a cast wall cannot lrt. reduced below a minimum thickness, and certain thicknessescannot lrc completely hardened by quenching. In all such cases,the limiting dimensi,,rrr must be ascertained,as was done, for instance, with the smallestsleevc l.r the gear coupling shown in Figure 7.6, which had to be strengthenedbr';rrr i n c r e a s ei n t h e w a l l t h i c k n e s s( D : 7 I m m t o D : 1 5 m m ) . T h e s a m ep r i n c i l , l c applies to measurement and machining provisions. J. Overriding standards are not always based on preferred numbers, so tlrc relevant componentsmust be adaptedaccordingly(seeFigure 7.6-fixing rhc module). 4. Overriding similarity laws or other requirements may impose a more l)ro. nounced deviation from geometric similarity, in which case semi-sintil;rt s e r i e ss h o u l d b e u s e d ( s e e 7 . 1 . 5 ) . Once the necessarydeviations from geometric similarity have been dctr mined, if necessaryby checking drawings of the critical areas, they are entercd i the data sheet. Production documents need not be prepared until actuir needed. To illustrate the size ranqe. sav in catalosues or advertisemcnt displays of the type previously reserved for technical drawings have come r increasinguse [7.7, 7.25]. Figure 7.7 shows an example basedon a gearbox range. Figure 7.8 shows the basic design of a geometrical range of torque-limitc providing for equal utilisation of materials. If the lining wears, the dro1.r Display of a gearbox sizerange [7. l4] (Flender) torque must be kept as small as possible.This is done by meansof a I number of peripheralcoil springswith relatively flat characteristiccurves. sizesof the torque-limiterfulfil the similarityconditionsmentionedin Table 7 relationships between forces are kept constant over the entire range ancl utilisation of the material is constant. Figures 7.9 a and b are the relevant data sheets.The identifiable deviatiorr dimension B is determined by the overriding standardwidth of the chain whc (bought-out parts); the deviation of ,4 by the use of standardscrewsand taps rr also by technological factors (wall thickness). Figures 7.10 a and b show r s m a l l e s ta n d l a r g e s tm e m b e r o f t h c s i z c r a n g c r c s p c c t i v e l y . 7.1.5 Semi-similar sizeranges (ictlntctricallysirnilirrsiz.criutgcsburedon e decimal.gcornctric scricscirrrrrol r e 7 . U .B a s i cd e s i g no f a t o r q u e - l i m i t e r( R i n g s p a n nK G ) be realised. Significant deviations from geometrical similarity may be posedby the followingfactors: ovcrridingsimilaritylaws; and ovcrridingtaskrequirements; rtvcrritlingproductionrequirements. In rrll suchclscs,.roni-sitnilarsizerangcsmust be developed. 7 Developing size ranges and modular protlut lr 2s000 20000 16 0 0 0 12500 r 0000 80 0 0 6 30 0 s000 4000 500 450 400 3s5 315 280 250 22t200 180 160 140 125 117 r00 90 E 80 71 : 63 56 ; 50 45 5 /.0 11qn I | | | 2000 1600 12s0 1000 800 neglected, then the relationships derived from the cauchy condition no longer ly. This, as we have explained, is because,while the inertia and elasticforces t constant speed depend on the length factor (gn: gpp:e2r), the weight lncreasesas: 9 p * : p t . s - V r l @ o ' 8 - V i : g p g 3 r , a n d f o r c p o: l , 5 b= 'o rcesmust change.Hence with similarcross-sections the stresses changeaswell WD nd geometric similarity cannot be maintained. This is the case, for instance, ith the construction of electrical machines and convevor svstems. l1q ?50 ?8 77 20 18 440 _? 31,5 J I, = 25 influenceof thermalprocesses similar seriesof problems ariseswith thermal processes.Constanttemperaure relationships96 only apply when there is thermal similarity, regardlessof her the heat-flow is steadyor fluctuating.The first caseis representedby so-calledBiot number, Bi:hLll U.201, where ft is the heat transfer // W ient and /. the coefficient of thermal conductivity of the heated wall. Here it is obvious that, with approximately equal heat transfer coefficients (the ity remaining the same) and with the same materials, only the length can e75 t0 =20 11 12 11, 10 o 63 80 1001251602002s0 315 Nominal D inmm+ dlmension 0 e3r. tant, length is the only variable dimension. If it does vary, the relevant imensionlessparameter cannot remain constant-that is, the relationship of the 500 400 200 ; 160 E 125 ! 100 E80 as er*: Table 7 .2 showsthat, if all other material propertiesand the speedremain I tioo bq . 33r 7.1 Size ranges ry, and indeed must vary in a size range. As a result the dimensionless rameter governing thermal similarity cannot itself remain unchanged.The 16 17,5 10 80 100125160200 D inmm+ dimension Nominal b gure 7.9. Data sheetsfor torque-limiter shown in Figure 7'8 parts I Dimensions adaptedto overriding standardsor the sizesof bought-out of inertiaWD2 and moment W weight T, torque ) Main parameteri: me is true of fluctuating heating or cooling processes represented by the r number: Fo = ltlksLz), re ,t is the coefficient of thermal conductivity, c lhe specific heat and g the sity of the material. If the material remains the same, the time r and the h L are variable. For the Cauchy number to remain constant. the time must ry as a function of the length. Once again we are left only with the length, ich must be variable in a size range. Hence the Fourier number can only in constant if: (Pt:?Lt o e is, if the time varies as the square of the length. o =e e I b igure7.10. Layouts from the size range shown in Figure 7.9 (RingspannKG); ) smallest ) largest Overriding similarity laws {luenc'eof gruvity incrtiil lirrccs.clilsticlirrccs:rndwcilhl rct lolclhcr. und il thc liltlcr cilnnothc All other thingsbeingequal, therefore,thermalstresses due to temperature riationsincreaseas the souareof the wall thickness. similarity relationships the function of a device is determined by physical processesthat do not involve rtia or elasticforces, then the physicalrelationshipsmust be taken into siderationin all designsbasedon similaritylaws[7.18,7.30,7.32,7.36]. In a plain bearing,for instance,the operatingconditionsare set by the crfcld number: So: pq4l(qco) 7 D e v e l o p i n g s i z e r a n g e sa n d m o d u l a r p r o t l t t tl r JZ 'here p is the mean pressure, r/ the non-dimensional clearance, 17the dynantrc iscosity and @ the rotational speed. In a machine that otherwise obeys the Cauchy number, we have .. - PtV2r\ottto - .^ -.2 I - I ,ft,!\9iU"f6r*,: With elastic forces we have ep:1, 'e have: 9v: 7, 9.: with weight we have gp: et', for the rcst, 7 1 9 Y ,c P n =1 a t I : c o n s t ' .1 Size ranges JJ.' tasmuch as inputs and outputs may vary widely in size, as happens with paper nd print products. Figure 7.11 is a schematicrepresentation of a rathe.Here, the size of the an-operated controls cannot be increased with the size of the range; indeed me cannot be altered at all. Thus the operating height must always be adapted to man, and there are some operations that require an exceptionally long turning length or an exceptionally large turning diameter. In all such casesthe machine as a whole must be designed on semi-similar principles, while individual assemblies such as spindle drives, tail-stocks etc can be developed as geometri- y srmrlarseries. 'lltr' y'ith elastic forces, therefore, we have gs,,: eri with weight gs,,: El. ommerfeld number increases with the overall size, the bearing becottt,r rcreasinglyeccentric and, at a given size, may take up the clearancenecessiu\ rr lubrication. In a pipe with laminar flow, the loss of pressure is expressedby: l p' : f ! . 9 r t : 3 2 n ! , d2 d2 ,here f : 641R" in the laminar region, R": dvpfry, / : length of : diameter of pipe, v : velocity in the pipe, e : densityof the fluid, : dynamicviscosityof the fluid. With r7: constant,the pressurelossfunctionbecomes: clzp: Q,lQr 'hus, if the pressure loss is to remain constant, the velocity in the pipe nr lcrease in proportion to the size. As a result, the Reynolds number n tcrease to such an extent that the transition region for turbulent flow is reachc r which case the above equations no longer hold. Electric AC motors that have a discretespeeddepending on the pole numl annot be used to adjust the speed of a finely stepped range of machines (lr rstance pumps) to maintain a constant Cauchy number. The consequenr rould be varying stresses and different outputs and the remedy is a suititl d a p t e d s e m i - s i m i l a rs e r i e s . Overriding task requirements 'he c h o i c e o f a s e m i - s i m i l a rs i z e r a n g e m a y b e i m p o s e d ,n o t o n l y b y s i m i l i r r i t lws but also by overriding task requirements. This situation often arises irr r g o n o m i cc o n t e x t . A l l c o m p o n e n t sw i t h w h i c h h u m a n b e i n g sc o m e i n t o c o n t n t h c c o u r s co f t h c i r w o r k - e s p e c i a l l y t h c c o n t r o l s ,h a n d l e s ,s t a n d i n ga n d s r t t i , l a c c s ,a n d s a f c t y f c a t u r c s - m u s t f i t m i t n ' s p h y s i o l o g i c a ln c c d s a n d p h y s i c l l l i n t c n s i < l r t sI n . g c n c r l l , n o n c o f t h c s c c o n r p r ) n c n t sc a n b c c h a n g c d w i t h t l r o rurminal siz.crlf thc rangc, An ovcrriding rcquircnrcntrnily rlso rppcur for purcly tcchnicll rcils(ln\. i s u r e7 . I l . L a t h ew i t h m a i n nsions and controls wn schematically;the meter/lengrh/height rario y haveto be variedto suit I q,. * cp,,but if possible = cpr,: I for ergonomic eo=* "0 9r=# ng 'p,= -!:' " Da Overriding production requirements development of a size range is aimed at high cost-effectiveness.within the nge, especiallyif it is finely stepped,individual componentsand assemblies be more coarsely stepped to provide larger batch sizesfor even greater cost tiveness. Figure7 .12is the datasheetof a geometrically similarturbinerangeconsisting sevensizes.Stuffingboxesand locatingboltsare steppedmore coarselythan rest, ensuringgreaterbatch sizesand greatereconomy.Figure7.13 shows increasein batch sizesfor an assumedsalesprojection. All these examples make it clear that it is not always possible to adhere to :metricallysimilarsizeranges;instead,the designermuststrive,with the help similaritylaws,to arrive at that sizerangewhich providesthe highestoverall ilisation of the strength of every component. Depending on the physical raints, each size will have to be individually selected.This is best done with help of exponential equations, as we shall now go on to show. Adaptation with the help of exponential equations ponential equations are a simple means of dealing with the requirements n t i o n c d u n d e r 7 . 1 . 5 . 1 - 3 a n d o f d e v e l o p i n gs e m i - s i m i l a rs i z e r a n g e s . As wc havc pointed out, nearly all technicalrelationshipscan be expressedby rwcr l'uncti<lns. 7 D e v e l o p i n g s i z e r a n g e s a n d m o d u l a r p r o t l r t 't ' 334 800 mm 630 wheeldiameter Turbine pipediameter Exhaust part:outletdiameter Reaction s00 Meanpistondiameter 400 3 1 5 .4 250 200 diameter Coupling Mainsealdiameter boxdiameter Stuffing frontPinion diameter: Bearing gearwheel diameter: Bearing z 160 '1 25 E !ogr)"k i-xr:xugL'"k; rr: where 91 is the chosenstep factor of the dimension chosenas nominal in the size range' 10,x0, z0 are the appropriate values of the basicdesign,k is the k-th step, Bnd y., Je and ze are the associatedexponents. Since cp is a constant, we have for all elements ct : ci (zlEtz"t{)e' lu : locPtv.k: c(xocpy'"k)p" 31.1 25 400 710 are geometricallysinrrl Figure7.72. Data sheetfor turbine sizerange: main dimensions bolts are in largcl locating and boxes stuffing standards; by determined are deviations steps than the other comPonents forecast Sales 265 315 400 s00 630 710 800 s 6 3 2 I \umDer 6 I Type boltsperturbine 3 locating 0 4 00 5 0063 6 1 1 0 8 0 a75 ,3',r5 6 3 Number1 8 77 77 r8 I )lZe Fieure7.13.Salesforecastin respectot tuibine sizerange(Figure7.12)and the bolts.Becauseof the large associated stepsizes,largerbatchsizesarepossible With ye - cxoP^zoP' we have: hich is independent of k. Here y", x" and zc are the exponents to be determined, and p* and p, the ysical exponents of x and z. The exponent ye must be determined independently of xc and zc. Let us now consider a practical example: the provision of sprung elastic peline supports for a range of geometrically similar valves (Figure 7.14). The llowing requirements must be met: the stress in the spring due to the weight of the valve must be constant throughout the range; the stiffness of the spring must increase as the bending stiffness of the pipe; the mean spring diameter, 2R, must preserve geometricalsimilarity with the increasingvalve size (nominal dimension d). What law must the spring wire diameter 2r and the number of active coils, n, obey? First of all the appropriate relationships must be set down, so that the exponential equation can be determined (the subscripte shows that only the exponent of the corresponding quantity is involved): F: Number 0 3 15 /q Cd3 F.R to Comblned Size 050 tq C XqP"Z1.P' . EtG"ktt-+ z"k r,) Jc: xepxl Z"pr, 50 /.0 26s : llt By equating the exponents, we obtain: boltdiameter Locating Flange boltdiameter Valvediameter 80 Zoer'"k IoEJ"u : Y,EtG.kt'+z"kr,) 100 20 !x: portdiameter Transfer 63 o80 1B T h u s a p h y s i c a lc ; u i t n t i t Yo f ' t h c k - t h m c r n b c r o f a s i z c r a n g c c a n t l l t c r l r e p r e s c n t c cbl y : I'l -- t'u1u/''-ul" 'l 335 expressedby preferred numbers starting from the basic design (Index 0): r000 ? 7 I Sizcranges I 1 c r l c p c r r t l u t l v i t l a i l h l c t , l n ( l t h c i t t t l c l t c t t t l c r t tv t r l i t t b l c s t l t t t t l ; c i t t t i t l w l t t r 13nf2 .)-- Gra 4nR3 (1) F,:3d" F. * R. - 3r":6 (2) r.: (3) s" : 4r. - n" - 3R" (1') (2') (3',) [ . c t r / b c t h e i n d e p e n d e n tv a r i a b l e . S i r r c c t h c s p r i n g s t r e s s m u s t r e m a i n c o n s t a n t , t h e f a c t o r ( p r : 1 , and the xponcnt r" = 0. Thc stiffnesss of the spring must correspond to the bending l l i l l r r c s sr r f t l t c p i p c s .A c c o r c l i n gt o T a b l c 7 . 3 t h i s i s e n s u r e db y E , : E r S i n c et h e 36 ,)-)/ 7 Developing size ranges and modular Substitutingequations(5') and (6') in equation(2') we obtain: 3 d " +d " - 3 r " : 6 r.: (413)d" L! 6 S u b s t i t u t i n ge q u a t i o n s( 4 ' ) , ( 6 ' ) a n d ( 7 ' ) i n e q u a t i o n ( 3 ' ) , w e o b t a i n : q : (7') 4rr- n"- 3d": 4. ffi- +| n" : 4r, - 4d" = 4(413)d, - 4d" : (413)d" Result: Spring wire diameter 2 r and the number of active coils n must increase as d4t3. In that case. the factor is: Qr: Qn: qdt4/3 The spread of the individual sizes is shown qualitatively in the data sheet r e p r o d u c e di n F i g u r e 7 . 1 5 . Examples ample 1 range of high-pressuregear pumps is to consist of six sizes giving delivery umes ranging from 1.6 to 250cm3 per revolution at a maximum operating of 200bar and a constantinput speedof 1500rev/min.In Figure7.16 steps laid down for the six sizesare plotted againstthe delivery volumes. The l o w i n g r e l a t i o n s h i p sa r e i n v o l v e d : The pitch circle diameters (each pump size has only one) are graded in a c c o r d a n c ew i t h R 1 0 w i t h a f a c t o r o f B u : 1 . 2 5 , t h e s i z e sd e v i a t i n g v e r y slightly from the preferred numbers by virtue of the constant.integral number of teeth and also because the standard values of the modules rn differ very slightly from the R 10 series. The volume delivered per revolution resulting from the tooth geometry is ABTO Noninalsizeof valve tigure7 .1,4 Figure7.15 rigure7.14.Valvesupportedin pipe line by meansof coil springs rigure7.15.Data sheetfor semi-similar coil springs V :2n dnmb, where b : gear tooth width. rasic dimension d of the valves i n c r e a s e sg e o m e t r i c a l l y ,e s : E 6 , s o t h a t )xponent of s becomes: s": d" ( The loading is equal to the weight of the valve F; the weight dimcnsiorr 'elated to the basic size d by E r : E u : . T h e e x p o n e n t o f F r e f e r r e c lt o , / hcrcforc: /'' : 3d' l l ' t h c n r c a ns l l r i n gc l i i r r n c t ci rs t r l i n c r c i r s ci n l l c ( ) n l c t r i c asli r n i l i r r i t yw, c From one size to the next, and at geometrical similarity, the volume delivered lore lncreasesas: : Qt3 : 1'253:2, Qv : cPa,,cP^qt the volume delivered doubles from step to step (Figure 7.16). p u m p p ( ) w er P : A t t . I / i n c r e a s e sa s Ep: E^p(EYlcP) V'rt, : I irnclt1, -_-I l i l v C V ' t { - - r 7 ' . 1{ ) f i ll,. - r/" (t ' l 7 Developing size rangesand modular ]8 250 0 cm3/rev 160,0 1250 100.0 800 630 50,0 10,0 3 15 80.0 25,0 r * l 20.0 50,0 160 4 0 0 12,5 31,5 1 0 0 2 5 0 B,O 20,0 6,3 16,0 5 0 B O 17,5 4.0 m m 1 00 l s 0 8.0 6,3 5,0 boJ 1.f, 40 25 20 315 339 Size ranges and the increasing bending mornents due to increasesin tooth width) with a shaft of constant diameter, the three pumps with the greatest tooth width in each size group must have their output pressure reduced. For overriding economic reasons (identical shaft diameter, identical bearings), the first two pumps of each size group do not have their strength fully exploited. - The delivery volumes of the top three pumps in any size group correspond to the bottom three of the next group up. A delivery pressure of 200 bar can therefore be obtained over the entire delivery-volume range. This particular size range was conceived as a semi-similar series with a small number of housing sizesand severaltooth width sets, so that, at the same drive speed and pressure over the entire range ('overriding task requirements') and also at constant gear tocth size, constant gearwheel and shaft diameters per housing size ('overriding production requirements'), the maximum possible range of delivery volumes could be provided. t I Example 2 In Figure 7.r1 theoutput P of asize range of electric motors with varying pole numbers (speeds) has been plotted against the various product sizes (shaft heights H). The shaft heights are in accordancewith R 20 andhave a step factor I "19 PC1 PTz PC3 PC4 PCs Product size igtreT .1.6.Data sheetfor a size range of high-pressuregear pumps: I/ volume delivt rt' lr revolution; b gear-tooth width; do pitch circle diameter ot gears (Reichert, Hof) ecomes: t P p : c P v: 2 iecauseof the constant rotational speed, the torque is steppedup accorclillfl -Every pump size has been provided with six tooth widths b, exccpl smallestsize which has eight, so that smaller stepsin the volume delivcrctl t b e o b t a i n e d . T h i s m e a n s t h a t f o r e a c h p u m p s i z e t h e g e o m e t r i c a lv ( ) l u c l e l i v c r e dV, = 2 n d 1 1 m b ,w i l l h a v e a f a c t o ro f 9 v , , : Q u : 1 . 2 5 , r d 1i t n c rl r t l r t ' t 'l'hc p,trr c o n s t i r n ta n c lt h c c h o s c nt o t t t h w i c l t hf a c t t l rb c i n g c 1 t 5 :| . 2 5 ( R l 0 ) . ctlrvc fi)r illlv ollc Dullll sizc tltcn lrcctltttcs: I ' r , , *| \ ' , . - { l - l ' 2 5 Io copc rvitI the nlcchllnicill sllcrr('r (terttlliltF Itrlttt lltc ittt'rt'irsittl.l 3'r50 KW 2500 22t0 2000 't800 1600 1400 1250 r120 1000 s00 800 710 630 560 500 450 400 35s 3r5 280 750 /t\ 7r ) 0 B - p o t . mPo.x l'r'r ((10 /,50 500 560 (:;lr;|ll ['r0(lri(;l :;t/(] lrcirllrl l{) - 6 3 0m m7 1 0 - Figure7 17.Outputdatasheetfor an electricmotor sizerange(AEG T e l e f u n k e n[)7 . 1 ] 7 D e v e l o p i n g s i z e r a n g e s a n d m o d u l a r p r o t l t t Lt 341 7 I Size rarrges a: o,tJBbhtD.Ihttt A : 1.12. The output of the electric motor is governed by P at constant angular velocity o or speedn, current density -I and magnetic I I t rr nsity B, the output is proportional to the conductor dimensionsb, h, / and itl', ' the distance Dl2 of the conductor from the shaft axis. The output factor is therefore given by: c P p : Q r 4: l . \ 2 4 : 1 . 6 ( R 5 ) . ln the 4-pole motor (1500rev/min) the output range is therefore 500-3150k\\ Becausepower output varies with speed, and also becausethe dimensions"l e conductor, the diameter of the rotors and the heat removed by ventilati,'rr '' ve to be varied, the slower 6-pole version must be reduced by three steps ( I (280 to 1800kW) 2240kW) and the S-pole version by a further two steps To provide marketable and finer output increments and also to satisfv tlr, 'erriding production requirements, four outputs are provided per shaft hcir'lrt motor size, so that the output curve assumesthe form of a steppecl lirr' n a l l e r o u t p u t s a r e o b t a i n e d b y v a r y i n g t h e s i z e o f t h e e l e c t r i c a l l ya c t i v e p ' ' r t . rd fitting them into the same size of housing. In contrast to what happenc,l rrr r(ample1, the outputs for the different size groups (fixed pole number) clo tr,'l r e r l a p ,( a l t h o u g ht h i s h a s b e e n d o n e w i t h o t h e r m o t o r d e s i g n ss o a s t o m a l r r l , r r r l ficiencv). Figure 7.18 showsthe welded housingsof the motor range in greatly simplili,'tl rm. The stepped sizesof several important dimensions are entered in a tlrrlit Leet(Figure 7.19).Itcan be seenthat the shaft height H,the housingheight i /tf rd the distance between the foundation bolts B and A are all stepped up by t c t o r E y : g H : 1 . 1 2 , R 2 0 . J u s t o n e h o u s i n gl e n g t h B C i s p r o v i d e df o r t h e l o rtputs per shaft size (Figure 7.18). This is possiblebecausedifferent sizesol I ectrically active parts can be fitted easily into one housing size. Without tlri rparationof the housingsfrom the electricalcomponents,the layout would ttr I economic and several housing lengths would have to be provided for c;tt t_a t \a qb- o r r s i l t gi r r c t l t i c k c r r i b s r c t l r t i r c t l , l l c c l i t r s co l o v c l r i r l i r r gs i r r r i l l r r i t \l 'i r u ' s ,o v c t t i t l i r t g l ; r s k r e t l t r i r e l l l ( ' l l t.sr r r t l v c i r i r l i r r gl t r r r t l t r r ' l i ornt . t l r r i r t . r r r r ' r ti rl rt ,r l i r ' i r l t t it rl il t t t r ' t t s i o tittst t t lt t o t t t i t l l rrlt r , ' t a 25.0a 315 + 2 8 0 | 22.t, 900 q I raft height l7 .271 'overriding similarity laws' on the electrical side (for instant't' Because of :spect of the windings) the housing length step factor ps6 cannot be kc )nstant over the entire range of the shaft heights. Figure 7.19 shows t rcreasein step factor for BC with increasingshaft height, the step sizc tr rproaching R 20 for the last two housingsof the range. 'l Let us now look at a few detailed measurementsof this housing desiglr aseplatedimensionsAA and BA have been graded by a singlestep factolur lr c s b c t w e e n R 2 0 a n d R 4 0 . T h i s w a s d o n e t o s a v e m a t e r i a lw h i l e m a i r r t i r i r t r r c m i n i m u r n c l i r n c n s i o n sn c c d c c l f o r a s s c m b l yo f f i x i n g b o l t s . T h c b i r s c p r i c k r r c s sI I A l t a s b c c r r s t c p p c c l i n a c c o r c l a n c cw i t h t h c u s u a l s c m i - l i n i s h r l t c r i i r lc l i r r r c n s i o nbsu, l l l y l r n c l u r g cl i r l k r w sI l 2 0 . [ ' o r t l t c s t r c n g t h c l t i nIgi l r r r l r r i r tl h i c k n c s s/ l / : i s p r o v i t l c r lt i l r l o t r r l t o t t s i r t gs i z c s .O n l y l t l r l l t c l l r t l l , ' DBlpns=9y) I 3 1 5 6a 250 H.zo,o 710 q \n ?u. 18,0 630 200 16,0 EAN I 14.0 Figure7.19 ,ru f I 500r-3 5 5 400 450 500 560 630 710 Product size Figurc 7. lfi Housing for the electric motor sizerange (simplified) shown in Figure 7 17 (AEG Telefunken); ( i r ) c r o s ss c c t i o n s (lt) elevation l r i g r r r c7 , l ( J ,1 ) r r t rsrh c c t t o r l t o r t s i n gr l i n t c r t s i o nosl t h c c l c c t r i cr t t o t o rs i z cr a n g ci n F i g u r e 7 , l 7 ( S v r r r l r o il rssi r rI ; i g u r c7 . l l l ) a Aa 7 D e v e l o p i n g s i z e r a n g e sa n d m o d u l a r p r o d u ( l \ J+L may have to be stepped in accordance with laws that differ from those leading ttr geometric similarity. In every case, however, the designer must, in the first instance, aim at size ranges based on the appropriate similarity laws and th(' preferred number series and only deviate from them after careful consideratiorr of the costs involved. 7 .1.6 The developmentof sizeranges Size-rangedevelopment can be summed up as follows: 1. Prepare the basicdesignfor the range. This can be completelynew or derivt,l from an existing product. 2. Determine the physical relationships (exponents) in accordancewith s i m i l l r r ity laws, using Table 7.3 for geometricallysimilar product ranges,o r u s l l l r ' productranges.Put downthe exponentialequationsfor semi-similar the form of data sheets. 3. Determine the step sizes and add them to the data sheets' 4. Adapt the theoretically obtained ranges to satisfy overriding standarcls,'t technological requirements and record the deviations. 5. Check the product range againstscalelayouts of assembliespaying partictrlrrr attention to critical areas for extreme dimensions. 6. Improve and perfect what documentation may be needed to determinc tlrc range and prepare production documents (when required). The need for developing a semi-similarsizerange may not alwaysappear frrtttl the specificationor from a first survey of the physicalrelationships,but may ortly become clear during the actual development. 7.2 Modular products In 7.1 we discussedthe features and designpotential of sizerange developmeIlt Their aim is the rationalisation of product development by the implementittir of the same function with the same solution principle and if possible with t same properties over a wide range of stzes. Modular products provide rationalisationin a different situation. If a prodt is to fulfil different functions, then many variants will have to be proviclctl , great cost in design and production. Rationalisationis, however, possiblc il t oarticular function variant at anY one time is based on a combination ol'lix i n d i v i d u a l p a r t s a n d / o r a s s e m b l i e s( f u n c t i o n u n i t s ) , a n d t h i s i s p r e c i s e l yw l t r r t modular systenr sets out to achreve. B y m o d u l a r p r o t l u c t sw e r c f c r t o m i t c h i n c s ,a s s c m b l i c sa n d c O m p o n c n l sI l u l l ' i l v a r i o u s o v c r a l l l ' u n c t i o t r st h r < l t r g l t( h c c o n r b i n i t t i t l l to f ' d i s t i n c t b u i l t l i r t l bklcksrlr rrroclulcs. llcclusc srlchrrl(xlulcsnrily contc ilt vrlriotuisizcs.mttdttlitrprotlttclsoltctl 'l'hc shortldhc produccdby sinrillr lccltttttlttcr trrotlulcs ilrvolvcsirc rltUlu,, 7.2 Modular products 343 whenever possible. Since in a modular system the overall function results from a combination of discrete units, the development of modular products demands the elaboration of a corresponding function structure and this calls for greater design effort during the conceptual and embodiment phases than does a pure size range development. The modular system can provide a favourable technical and economic solution whenever all or some different products are required in small batch numbers only, and whenever they can be based on a single unit or on only a few basic and additional units. Besides fulfilling a variety of functions, modular systems can also serve to incease the production batch size of identical parts for use as building blocks in a variety of products. This additional objective, which greatly helps to rationalise the production procedure, is attained by the breakdown of the product into elementary components (6.5.6). Which of the two objectives is paramount depends largely on the product and on the task it has to perform. With a wide-ranging overall function, what matters most is a resolution of the product into function-orientated modules; with a small number of overall function variants, on the other hand, a production-orientated resolution is the paramount consideration. Often, modular development is only initiated when what was originally conceived as an individual or size-rangedevelopment is expected to yield a large number of variants. To that end, products that have already been marketed are often redesigned as a modular system. The disadvantage here is that the products are more or less predetermined; the advantage that their essential properties have already been tested so that an expensivenew development can be dispensedwith. 7.2.1 Modular product systematics Modular product systematics is discussed in [7.10,7.11]. Basingourselveson thesefindings,we shallfirst of all examinethe principlesandthe mostimportant concepts,and merely add a few amplifications. Modular productsystemsare built up of separableor inseparable units. We must distinguish between function modules and production modules. Functionmoduleshelp to implementtechnicalfunctionsindcpendentlyor in combinationwith others. Productionmodulesare designedindepcndentlyof their function and are basedon production considerations alone" Function modulesin the narrower sensehave been dividedinto equipment,accessory, connectingand other modules17.70,7.IIl. This divisionis neitherclear-cutnor adequatefor the developmentof modularsystems. For the classificationof function modulesit seemsadvantageous to definethe varioustypesof functionthat recurin modularsystemsand can be combinedas sub-functions to fulfil differentoverallfunctions(overallfunctionvariants)(see Figurc7.20). I)u.ticluttctiorf.r ilrc funclamcntal to a svstcm.Thcv are not variablein 344 7 D e v e l o p i n g s i z e r a n g e sa n d m o d u l a r p r o c l r r t ' principle. A basic function can fulfil an overall function simply or in combinatiorr with other functions. It is implemented in a basicmodule which may come in ()r)c 'essential'(Figure or several sizes,stagesand finishes. Basic modules are 7.20) lmplementation Variants _ Essential module ----- Possible module - - 0nlyinspecial Ieads cases; tomixed systems Figure7.20.Functionand moduletypesin modularand mixedproductsystems A production-orientated charcteristic islhe complexity of a module. Here we distinguish between large modules which, as assemblies,can be subdivided into components, and small modules that are components themselves A further aspect of module characterisation is their type of combination. The designer should always aim for technically advantageouscombinations of similar modules. In practice, however, the combination of similar with different modules, and also with customer-specificnon-modules, is often unavoidable. Non-modules, as mixed systems,can meet market requirementsvery economically. For the characterisation of modular systems we can also consider their divisibility-in other words, the extent to which a particular module can be broken down into individual parts for functional or manufacturing reasons. For the modular system as a whole, divisibility defines the number of individual units and their possible combinations. For the application of closed modular systems,their range and potential can be expressed by combinatorial plans with a finite and predictable number of variants. Such plans make it possible to choose desired combinations directly. By contrast, open modular systemscontain a great multiplicity of combinatorial possibilities, which cannot be fully planned or represented. A specimen plan provides examples of typical applications of the modular system. The above-mentioned concepts of module development are summarised in Table 7.5. Classifyingbriteria of module: Auxiliary functions are implemented by locating or joining auxiliary modt 'essential'type. in conjunction with the basic modules and usually of the Specialfunctions are complementary and task-specific sub-functions that ncc not appear in all overall function variants. They are implemented by speci 'possible' type. modules of the Adaptive functions are necessary for adaptation to other systems and marginal conditions. They are implementedby adaptive modules whose dimc sions are not fully fixed in advance and hence allow for unpredictable circ s t a n c e s .A d a p t i v e m o d u l e s m a y b e o f t h e ' e s s e n t i a l ' o r t h e ' p o s s i b l e ' t y p e . Customer-specificfunctionr not provided for in the modular system will rccur time and again even in the most careful development. Such systems ill implementedby non-modules which have to be designedindividually for spccilic tasks. If they are used, the result is a mixed system, that is a combinatitlrl ol modules and non-modules. By the importance of a module wc rcfcr trt its ranking within a morlttllt s y s t e m . T h u s , f u n c t i < l nm o d u l c s c a n b c r $ n k c d t t s ' ( ' J . T c l I i ( t / ' o ri l s ' l t o s . r i h l e ' l7.l3l. 345 7.2 Modular products mportance of modules lexity of modules Combinationof modules Divisibilityof modules Applicationof nrorlulcs Distinguishingfeatures - - - Function modules o Basic modules o Auxiliary modules o Special modules o Adaptive modules . Non-modules Production modules Essential modules Possible modules Large modules Small modules S i m i l a r m o d u l e so n l y Different modules only Similar and different modules Modules and non-modules Number of parts per module Number of units and their possible combinations Closed system with combinatorial plan C)pensystemwith specimenplan Tnblc 7.5, ( irnccpls ot' nrrltlullr syrtcnlitlioi 346 7 D e v e l o p i n gs i z e r a n g e sa n d m o d u l r r p r o ( l u \r \ 347 7.2 Modular products 7.2.2 Modular product development In what follows the development of modular systems will be presented in accordance with the steps listed in Figure 3.3. e o = -. c Clarifying the task In his formulation of demands and wishes, for instance with the help of th.' checklist (Figure 4.5), the designermust pay careful attention to the clarificatiorr of the various tasks to be performed by the product. A characteristic demancl ol the specification of a modular system is that it must fulfil several overrrll functions. Of particular importance for the economic analysis and application of modul..' are data about the market expectationsof particular variants. Friedewald [7. | /l speaksof the quantification of function variants for the technical and econorrrit optimisation of modules. Whenever the implementation of rarely demandr',1 variants increasesthe overall costs of the modular system, an attempt must lrc made to remove such variants. The more searchingthese analysesare before tlrc actual development is begun, the greater are the chances of arriving at ;r cost-effective solution. However, the reduction of types by the removal trl infrequently demanded and costly variants cannot be finalised until the elabo. rated solution concept or even the embodiment design provides reliablc information about the cost of the different variants and also about the influencc of every individual variant on the cost of the modular system as a whole. t @ \ I \ Lr- s' E6 c 92 6E E 6- S€ > a- il, l'r d l I ,l c.l FO ,7 I E stab lishing functio n structure s o bo The establishmentof function structuresis of particular importance in llr€ development of modular systems. With the function structure-that 0) is tl splitting up of the requiredoverall functioninto sub-functions-the structurc( the system is already laid down, at least in principle. From the outset, tl designer must try to subdivide the overall function variants into a mininru number of similar and recurring sub-functions(basic,auxiliary,speciali adaptive functions, see Figure 7.20). The function structures of the ovcr function variants must be logically and physically compatible, and the su functions determined by them must be interchangeable.To that end, it is usclr if, depending on the particular task, the overall function can be achieved 'essential' functions and by additional task-specific'possible'functions. Figure 7.21 shows the function structure for the modular bearing systc discussed in [7.3, 7.23]. The most frequently demanded overall functio namely 'non-locating bearing', 'locating bearing' and 'hydrostatic locati bearing', together with the appropriate basic, special, auxiliary and adapt functions, are represented.By means of the sub-function 'seal betwecn rotilti and stationary systems', we can show that it is often more cost-cffectivc lrl combine several functi<lnsinto onc complcx l'unction; thus in thc prcscnt cirsc, the sealing function was comhincd with tn ilclaptivc function to sltisl'y vitrioitr conditiorts.Thc productitlrt mrxlulc. which pcrforms this colrrplcx function. wlr (.) o t- \ B o c c o z b6 .l (.) o.l F- 9=F =:.='6:iE oa/5= LU: bo Il 6 348 7 D e v e l o p i n gs i z e r a n g e sa n d m o d u l a r p r r r r l , , , , . 7.2 Modular products accordingly specified as an unfinished one that could be completed dr.rrirrg production as a simple line seal, as a line seal with an additional labyrinth or ir\ 0 seal with an additional coupling adapter (see Figure 1.22). Ir should also b6 stressed that there are special functions (special modules) that occur in at lerrrl 'transfer axial force Fa from rotltirrr'trt one overall function variant (here: stationary system'), others that represent possible modules for all overrrll 'set and measureoil pressure'),and yet others that orrly function variants (here: 'feed high pressure oil'). become necessaryat a certain size (here: In the setting up of function structures the following objectives shoultl 116 borne in mind: - Aim for the implementation of the required overall functions by combination of the minimum number of easily implementable basic tions. -Try to divide the overall functions into basic functions and if necessarr irr auxiliary, special and adaptive functions in accordancewith Figure 7.2(l such a way that variants in high demand are predominantly built-up with lrrrr functions, and more rarely demanded variants with additional special rr adaptive functions. For very rarely demanded function variants, tnr systemswith additional functions (non-modules) are often more cost-cllr tive. -Try to combine several sub-functions in a single module if this incrci cost-effectiveness. Such combinations are particularly recommended for implementation of adaptive functions. Searchingfor solution principles and concept vanants The next step is to find solution principlesfor the implementation of the varir sub-functions. To that end. the desisner should, above all. look for s principles as provide variants without changes in working principle and I design. As a rule, it is advantageousto stipulate similar types of energy ,= 6 F r') L! ---1-i ci F-- ao tl 'r' T\ -+\ I I -J I 1'hc plain bctring systcnlnlusl l'rctlcsigncdduringthc conccptunlphlsc b{] I similarphysicalworkingprinciplesfor the individualfunctionmodules.Thusi more cost effective and technicallv advantageous, in the combinatiort sub-solutions into overall solutions (concept variants), to implement vari drive functions with a single type of energy rather than provide a single mt system with separate electrical, hydraulic and mechanical drives. A satisfactoryproduction solution is also ensured by the implementation severalfunctions by a single unfinished module that can be completed in vitr ways depending on the requirements. However, so complex are the technicaland economicfactors involved thirt impossible to lay down hard and fast rules. Thus, in the case of the system (Figure 7 .22) ft seems technically and economically advantagcout provide the bearing shell with lateral locating surfacesfor taking up sntall rr forces. With larger axial forces, hclwcvcr, rolling lrcarings must l'lc llrov i n s t e a d ;i t w o u l d b e a m i s t a k c t o t r y , f i l r p u r c l y t h c o r c t i c a lr c a s o n s ,t o l r i l l t h e r a d i a l a n d a x i a l f i r r c c so v c r t h c c n t i r c s i z . cr n n g c b y n l c o n s< l l ' p l l i n b c i r r i t t ; T-l tl € q I I I 1--- I -l I I I I I I I -l I I .l _., l1 F o o = u 6 J N cl a- !l ll L_ J bo 350 7 D e v e l o p i n g s i z e r a n g e sa n d m o d u l a r p r o r l r r r l t 7 2 Modular products 351 two alternative lubrication systems (free ring or fixed ring) because tht respective advantages and disadvantagescan only be determined by lrrt tions than to impose such adaptations on the whole modular system. An alternative is the use of mixed systems. experimentsl7 .23].The designof the ultimatelychosenbearingsystemis sht in Fisure 7 .22. Prep aring dimens io ned lay o uts Selecting and evaluating If several concept variants have been found during the previous steps,each rrrrr now be evaluated with the helo of technical and economic criteria so that t most favourable solution concept can be selected.Experience has shown llrrrl since the properties of any one variant are not yet sufficiently clear at this strrle such selections are very difficult to make. Thus, in the case of the bearing system, preliminary evaluationshave tt, made even in the conceptual phase, for instance as to whether the axial lorr should be taken up by plain or rolling bearings. However, the final choitt'r lubricating system can only be taken after the building of prototypes :r experimentation with them. Apart from the determination of the technical rating of individual con..r variants, economic factors are of crucial importance in the design of morlrr systems.To come to grips with them, the designermust estimatethe productr costs of the individual modules and their relative effect on the cost ol modular system as a whole. To that end, he will first of all determinc 'function expected costs' of the sub-functionsor of the modules fulfilling tlrc At the lower level of embodiment characteristicof the conceptual phasc. cannot usually hope to come up with more than very rough estimates. Si basic modules appear in all sorts of variants, he will select such soltrl principles as provide the most cost-effective basic modules. Special and adi modules take second place in the minimisation of costs. For minimisins the costs of a modular svstem. not onlv the mod themselvesbut also their interaction must be taken into account; in partic the influence of special. auxiliary and adaptive modules on the cost of thc modules. The influence of the cost of every overall function variant on thc of the modular system as a whole must be fully determined. This may pr( complex task. Thus, in the bearing system we have been considering. functionvariant 'cool oil internallv' would sreatlvinfluencethe costof thc module 'bearing housing', becausethe dimensionsof the specialmodule 'w cooler' determine the dimensions of the housins and hence the overall costs. there is only a small demand for this variant then it is certainly rr cost-effectiveto fit the oil cooler to the outside of the housing and to put ul) the extra cost of an oil pump. ln short, the layout of the basicmoclulesmust be adaptedto the cxpc demand.To that end, the influcnccof the rcmainingmodulcsis ol' importance.If it is impossiblc to prrlviclc a markctablcaclaptation ol thc conccpt.thc lcastcost-cft'cctivc functirtnvnritntsshouldbc clirninatccllrorrr modularsystcm.lt will ol'tcn bc ntorc economicullo rcplaccurrr,rsuirl virri which rcndcr lhc ovcrlll syritcnrIn()rccxpendvc.by mrkirrgirrdivirlull Once the solution concept has been selected, the individual modules must be designed, in accordance both with their functions and also with the production requirements. In the design of modular systems, manufacturing and assembly considerationsare of paramount economic importance. By paying heed to the embodiment design guidelines laid down in 6.5.6 and 6.5.7, the designermust try to provide basic, auxiliary, special and adaptive modules with the maximum number of similar and recurring parts and the minimum number of unfinished parts and manufacturing processes. When selectingstep sizes,the designershould aim at the optimum divisibility of modules, and to that end he may well adopt the differential construction approach. The determination of the optimum number of modules is, however, a complex task, for it is influenced by the following factors: - Requirements and quality must be maintained and the propagation of errors must be taken into account. Thus the greater the number of individual components, the greater the number of fits, and this may have untoward repercussions on the function, for instance on the vibration of the machine. Overall function variants must be created by simple assemblyof modules. Modules may only be broken down to the extent that functions and costs allow. In modular products marketed as overall systems,variants of which the client can assemble himself by combinations of the modules 17.291,the most common modules must be designed for equal wear and tear and for easy replacement. In determining the most cost-efficient modularity, the designer must pay special heed to the cost, not only of the design itself, but also of overall schedulingand of manufacturing processesincluding assembly,handling and distribution. Figure L22 showsthe scalelayout of the bearingsystemwe have been ussing. In Figure 7 .23 the structure of the overall function variants is shown the form of a family tree. In both thesefigures,only the most important blies and individualparts of the bearingsystemhave been entered;the ual modularityis greater.If the functionstructureis comparedwith the final ular structure, it becomes clear that in the given modular system several nctionsare fulfilled bv a sinelemodule or its variants.Table 7.6 showsthe ules used and their assisnedfunctions. 'p aring pro duction documents uction documents must be prepared in such a way that the execution of rs can be based on the simple, and if possiblecomputer-aided,combination nd l'urthcr cluboration of modules for the required overall function variants. I ) r i r w i n g s r c q u i r c a r r a p p r o p r i a t e p a r t - n u m b e r i n gs y s t e m a n d c l a s s i f i c a t i o n , 352 7 D e v e l o p i n g s i z e r a n g e s a n d m o d u l a r l ) r o ( i r Lr,r 2 Modular products O -ol al 2l ! al I 5. I F I c..t c\ t- =! bo -dFSRR tl. mmmmm -sh€ co CAXXXXXXXXXXXX O a I o oE =!l EA 5 E-i^" Er !.f ,^'!-, E; sT y i.\fr=s.e^" g sr-"HE oouIE=is" 7i E= !4!-y t *?Ei +€7! y &--v E! i2'E i 72;.; = d = d e s i $ d E #s " E i = E E s#i5E tE [tg .;ggig J = a] a- >' bo c5 ;6 bo IJ. .= E c (, o r = o o * '6 *O = = d d F J - N L o o 9s 9P E '-0 E =v j, EP =oe 'tr! cq >E oaNe6vJ ca cy', > U 0 5 >E< -c) u)o = m6m ::S!*<(laI 3 S 3 : : :: : I : g' g i f @ g' E o I : z =u = ! E.=c= I L d .= o o : E r , $ e E -v Y=r -Er ' -e- - E r : a 6-Y;=-= f- "Ts:I+==.;3H3EagE F dz 1H*.o oo-F:J E ! A*G a 354 D e v e l o p i n g s i z e r a n g e s a n d m o d u l a r p r o r l r r rr r two prerequisitesof the optimum combination of modules (individual parts an(l assemblies). The combination of individual modules into product variants must be rr.. corded in the parts list. To build up a parts list, the designer can refer to rlr(, so-calledvariant parts list [7.13] which is based on the structure of the proclrrr't and in which a distinction is made between'essential'and'possible'modules. Particularly suited to the numeration of drawings and parts lists is the methorl of parallel encoding, which assignsidentification numbers for the unequivot rrl and unmistakeabledescription of componentsand assembliesand classificatrorr numbers for the function-orientatedrecording and retrieval of thesecomponcnl\ and assemblies.The classification number is of particular importance in rlrr. modular product system, because it helps to detect functional and otlr,.l similarities of comoonents. \ / 7.2.3 Advantages and limitations of modular systems For the manufacturer, modularsystemsprovideadvantages in nearlyall arr.rrs: -ready documentation for tenders,project planningand design;designinrlrr done once and for all, thoughit may be more costlyfor that very reason: - additional design effort is needed for unforeseeableorders only; -combinations with non-modules are possible; -overall schedulingis simplified and delivery dates may be improved; - the execution of orders by the design and production departmentscan bc c short through the production of modules in parallel; in addition parts can supplied quickly; - computer-aided execution of orders is greatly facilitated; - calculationsare simplified; -modules can be manufactured for stock with consequentsavings; - more appropriate subdivision of assembliesensures favourable asscrrr conditions; and - modular product technology can be applied at successivestagesof protl development, for example, in product planning, in the peparationof drawi and parts lists, in the purchase of raw materials and semi-finishedmatcrilt in the manufacture of parts, in assemblywork, and also in marketing. For the user Ihere are the following advantages: - s h o r t d e l i v e r yt i m e s ; - b e t t e r e x c h a n g ep o s s i b i l i t i e sa n d e a s i e rm a i n t e n a n c e ; - b e t t e r s p a r ep a r t s s e r v i c e : - p o s s i b l e c h a n g e so f f u n c t i o n s a n d e x t e n s i o n so f t h c r a n g c ; a n d - a l m o s t t o t a l e l i m i n a t i o no f c r r o r s t h a n k s t o w c l l - d c v c l o p c dp r o c l u c t s . For the munul'actur(rthc limit of a moclularproduct systcm is rcachcclwhcrrc t h c s u b c l i v i s i o ni l t t o t r r o d u l c sl c a d s t < l t c c h n i c a l s h o r t c o n r i l r g sa n d c c o n r Iosscs: 7.2 Modularproducts 355 - Adaptations to special customer's wishes are not as easily made as they are with individual designs (loss of flexibility and market orientation). - Once the system has been adopted, working drawings are made on receipt of orders only, with the result that the stock of drawings may be inadequate. - Product changes can only be considered at long intervals because once-andfor-all development costs are high. - The technical features are more strongly influenced by the design of modules and the modularity than they would be by individual designs. - Increased manufacturing costs, for instance of locating surfaces; manufacturing quality must be higher because re-machining is impossible. -Increased assemblyoutlay is likely. - Since the user's as well as the manufacturer's interests have to be taken into consideration, the determination of an optimal modular system may prove very difficult. - Rare combinations needed to implement unusual requirements may prove much costlier than tailor-made designs. For the user there are such disadvantages as: - special wishes cannot be met easily; - certain qualitative characteristics may be less satisfactory than they would be with special-purposedesigns;and weightsand structuralvolumesof modularproductsare usuallygreaterthan and foundathoseof speciallydesignedproducts,and so spacerequirements tion costsmay increase. Experiencehas shown that, while modular production helps to reduce general staff costsin particular),it may leadto an ads (includingadministrative s e i n s h o p f l o o r w a g e s a n d e s p e c i a l l yi n t h e c o s t o f m a t e r i a l sb c c a u s e ,a s ntioned earlier, it tends to involve greater weights and volumes. Only if a lar systemis developedwith the expressintention of renderingevery nction variant more cost-effectivethan a speciallydesignedproduct can there a significant reduction in overall costs. .2.4 Examples rrc motor systems lar systems are particuarly cost-effective in the production of such versal drive svstems as electric motors. Figure 7.24 showsa modular high-outputelectricmotor system17.34].Of the ifiable modules, Numbers 1,2,5 and Il are fixed basic modules, Numbers 6, 7, 8, 9 and l0 are basic modules with possible adaptations for specific uirements (for example Numbers 4, 6, 7 and 8 for adaptation to various ages or Number 9 for adaptation to the coupling dimensions) and Numbers 3 I2 are special modules for meeting certain safety provisions. This particular modular systemis at one and the same time a size range, every m o d u l c b c i n q a v a i l a b l ei n d i f f e r e n t d i m e n s i o n s . 356 7 D e v e l o p i n g s i z e r a n g e sa n d m o d u l a r p r o r l r r r F i g t r e 7 . 2 4M . o d u l a r s y s t e m f o r t h r e e - p h a s e m o t ohrisgohfo u t p u ta, f t e r [ 73 4 1, s h r the mostimportantmodules 1 Base frame; 2 Support frame of outer housing;3 Cover platesand ventilation grid; J Terminal box; 5 Stator housing; 6 Stator laminations;7 Stator winding; 8 Winding ct 9 Shaft and rotor; 10 Rotor laminations and winding; 11 Bearing; 12 Cover plate A smaller degree of modularity is found in the next example-an elc motor systemof medium output 17.341.Figure 7.25 showsits structurc contrastto the motor in Figure 1.24, the statorhousingis combinedwith stator laminations and with the windins into a sinsle and undivided module. As a result, the following production sequencehas to be obsc manufacture stator housins as a welded construction-assemble and welcl laminations-add windings. In other words, the housing and the laminrt cannot be manufactured in parallel as they were in our first example. ['lcrc. I modular product technique bears mainly on the attachmcnt and constructi()rl t h e c o o l e r a n d t e r m i n a l b o x e s . o n t h c i n s t a l l i r t i < lonf v a r i o u sp r o t c c t i < l nd c and on the bearings(rollingbcarings,plain bcarings),Figurc7.2(rshowsI 'l'his possiblcarrangcrncnts l'rlr coolcrsancltcrntinul brlxcs. nrodullr prr systcm,lrlrl, is cornbincdwilh l sizcrungc$y$tcm. 7.2 Modular products lP 23 SllstJmIPR System lP 44 u t0 Svstem IPW 24 Figure 7.25. Modular systemfor three-phasemotors of medium output, after [7 .34], howing the most important modules Squirrel cagerotor;2 Winding head cover;3 Rolling bearingassembly;4 Radial plain ring;5 Bearing cover; 6 Stator housingwith laminationsand winding; 7 Cover grid; 8 eatherprotection;9Airfilter; l0Terminal box; ll Coverplate l2 Coolingsystem; irlwater cooler Another modularproductsystemfor electricmotorsis describedin [7.1].It is on the size range shown in Figure 7.77 and also containsthe plain bearing em shown in Fipure 7 .22. are another familiar example of modular systems. They involve a multiplicity of market-determined function variants (for instance, the attachmcnt of different input and output devices,various shaft positions and different gcar ratios). For the user it is a great advantage if he can build up gearbox to suit his particular requirements or if, in the caseof damage,he conl'igurati<lns sinrplc itncl spccdy replacements.For the maufacturer the modular nrakc Cun 3s8 7 D e v e l o p i n g s i z e r a n g e s a n d m o d u l a r p r o r l r r ,t r 3s9 2 Modular products elements ol Cooler water or bothsides, onrtght Cooler front or connectionconnection left,water back frontor back withcoonrt Layout elements onbothr,'1'', ondemand onlyIPR44 rl-llts+ TJll r# 6r:Ih lll,+ll 4l-lql il.t | ilt Layout wrthsidefixed cooler on demand Layout withcootttrt elements onboltt',tlr'1 0noeman0 Figure 7.26 Possiblearrangementsof coolersand terminal boxesfor the motor sho* rr Figure7.25 system provides a comprehensive gearbox system with only a few housings, w h e e l s ,s h a f t sa n d b e a r i n g s . Figure 7.27 givesan example: the housing of a modular gearbox allowing different input shaft positions and also for different gear arrangements (si and multiple stages,spur and bevel gears) 17.2ll. The reader will noticc undivided housing 1, which is closed on the output shaft side with an oval 2, and on the other side, for easier gear assembly, with alarge, circular cr underneath a smaller oval cover 4. While the circular cover 3 locatcs bearings of the various input shafts, the oval cover 4 merely covers the ut:l|l IVIililI |re ll-lT- iltF L_r__.]____-u Y ruil1 ilI II r e 7. 2 7. ' H a n s e n - P a t e ngte' a r b o x[ 7 2 1 1 r aperturesto various spur gearcombinations,by the inclusionof a special al cover for the flange-mounting of a drive motor, and by the provision for a vel gear input stage. The disadvantage of this highly modular design is that it mands a very accurate location of the various covers for proper alignment of shafts and seals. Moreover, the housing is not fully utilised unless all the gear are built in. A further development,namely a split housing[7.221, iates these problems (Figure 7.28). In that case the small individual covers t remain only carry the seals or only cover the bearing apertures. aperturewhen the input is throughbevelgears;but if the input is through gears, it contains a shaft seal. The slow running output stage IV is alwitvs through a bearing in the opposite side of the housing, with its shaft seal ltc the oval cover 2. The complete housing is therefore broken down into sc function/production modules, the central housing section and the circulitr serving as basic modules, the two oval covers serving as modulcs aclitpltd particular arrangements and dimcnsions of thc shal'ts,and thc bcitring ol b e v e l p i n i o n a n d a r c c t a n g u l a rb l i n d c t l v c r s c r v i n g a s s p c c i a l m o d u l c s . arr4ngcmcnthas thc advantitgcof providing.with u singlcccntrill lduptntiottol lltc r vnrinnlsby uppropriutc scction,a multiplicityol'possiblc IV x l2T1 ; I ; i g u r c7 . 2 1 ,I |l.o u s i n gf o, r ' l I a n s c n - P a t e n t ' g e a r b o. 2 thc housingis syntntctrical 7 D e v e l o p i n g s i z e r a n g e sa n d m o d u l a r p r o t l t r ,I 360 Another modular gearbox design is shown in Figure 7.29 [7.31)- It $:rr developed for the expresspurpose of providing a broad gear system (in resl'rt,t of shaft positions and gear ratios) with the fewest possiblemodules. In contrrrrl to the example shown in Figure 7.27, severalhousing variants were providctl to match more closely different gearbox variants and hence to reduce weight rrrrtl volume. Another modular gearbox, varied, is describedin l7 .7a1. which and outputs in particular can 7 .29.WGW gearboxl7.371 Fig:ure Endcap withbullet lrolltly F i g u r c7 3 0 . O p c l t t t 6 t l t t l l t\tV s t c t l l ( } l c r t l l r t ' V r t t(sl ) c r r r 1 p ,l ) r r i s [ r r r g ) , ( i r) n t o t l t r l c s( h ) s p et ' i t r t er t P l i t t t 7 2 Modular products 361 Further examples Further examples taken from hydraulics, pneumatics and machine tool construction can be found in the literature17.2,7.79,1.351. Mo dular conveyor system While all the systems discussedabove are examples of 'closed' modular systems, Figure 7.30 shows the modules and a specimen plan of an 'open' modular system. 363 8 Summary 8 Summarv Steps E 6 ; lvlethods and ards E s (, O main suppo(ing C = ') = ! After examining the historical background and the fundamentals, t h i s b o o k l r ; r r described a step-by-step approach to design, starting with product planning :rrrrl the clarification of the task and proceeding to the conceptual and e m b o d i r r rrtt l design phases. As special aids to cost-effective design it has also discussetltho development of size ranges and of modular products. Conceptual design and embodiment design are the two crucial phases in t creation of technical products or systems. Their respective steps are showrr Figures 8.1 and 8.2, where the various methods and aids are correlated rrr their main or supporting applications. The figures also chart the progr of the designer's work and the timing of the various steps. The propr correlations are not, of course, absolutely fixed, because tasks and problt differ from product to product, and because some of the steps may have to omitted. In other words, the various methods should only be applied when they required. Work should never be done for the sake of systematicsor for peclir reasons.Depending on his inclinations,experienceand skill, the designersh< choose the appropriate method for a particular step. The ability to abstract, to work systematically and to think logically creatively complement the designer's professional knowledge. In the va design steps, these abilities are demanded to varying degrees.Abstractiotr particularly needed for the identification of essentialproblems, for setting function structures, for determining the characteristics of classification schc and for the application of the principles and rules of embodiment dcsi Systematicand logical thinking help in the elaboration of function structurcs, setting up classificationschemes,in the analysisof systemsand processesirrrtl the combination of elements. Creative ability helps in the variation of funct structures, in the search for solutions with the help of intuitive methods, ilr combination of elements with the help of classification schemes rtr tlesi c a t a l o g u e sa n d i n t h e a p p l i c a t i o no f t h e b a s i c r u l e s , p r i n c i p l e sa n c l g u i c l c l i P r o f e s s i o n akl n o w l e d g ei s n e e d e d p a r t i c u l a r l yf o r t h c d r a w i n g u p o f s p c c i l r r t i o n s , f o r t h e s e a r c hf o r w e a k l i r r k s .l i l r s c l c c t i o t ta n d c v a l u i t t i o na n d l i r r c l t t b a s e do n t h e v a r i o u s c h c c k l i s t sa r r d o r r l i r u l t t r n c i n g , Figurc [3.3c<lmbincs thc chccklistsuppropriatelo thc variousdcsignplr Thc listsarc in accorcllnccwitlr thc gcncruldlrcctivcsgivcnin 2. 1.(r,lrrrl c Trend studies Market analysis Specification t.2. Abstraction 5.2. E natural systems 5.4.',| known solutrons 4.1 4.1 Tests, measurements 5 . 4 .I Brainstorming SVnectics 5.4,2. processes Syslematic study0f physical 5.4.3. frcalion Class schemes 5.4.3. Design catalogues 5 . 4 .3 Sketches Intuitive improvements 5.5. procedures Seleclion 5.6. Evaluation methods 5.8. Value analysis 6.5.6 E O E E E @ o o o o o o o o o o t . 4 .I physical mathematical relationships a o o o o Black boxrepresentatlon Funct on structure Literature search E = a OE 8.1 The systematicapproach d 6 o o o oo o o o o o o o o o o o o o o o o o o o o o o o 8.1. Correlation of methods and aids with the various steps of the conceptual ign phase (numbers refer to chapters and sections) t the technical function is implemented economically and safely. The i n g s o f t h e c h e c k l i s t sa r e a d a p t e dt o t h e v a r i o u sd e s i g np h a s e s . Before a specification can be drawn up, the requirements must be known in ail so that the function and important constraintscan be identified. For that r n , i n t h c c h c c k l i s t u s e d d u r i n g t h e c l a r i f i c a t i o no f t h e t a s k , t h e h e a d i n g : t i t r t ' n r a k c s w l y f o r t h c a s s o c i a t c dh c a d i n g s ' g e o m e t r y ' , ' k i n e m a t i c s ' , 364 8 1 The systematic appro;r,h I Summary rof task E 6 = o y)" E -f, Ehi 6o =- > = 6 igure 4 5) oe -E 4.2. o q o o o Functlon slructure 5 S o l u t j omne t h o d u s rn 0 phase conceptual 5 C h e ci ks t o.z. o ry o Pr*d" l..o Force kansmiss on I oo o oo ooo c ooooooooo o o ooo _ l _o , I " Faulttreeanalys s R i s rke d u o c tn Evaluation methods (Figure 5 44) w ththe Compatible overal I task Function Embodying Function ooo o o c o c o 6.6 o Evaluati ng Function design Layoul Embodiment Layoutandform 0esrgn ility Durab Deformation Stability Resonance Expansion Corrosion Wear Form design Satety Safety Safety Ergonomics Ergonomics Ergonomics Production Production Production Quality control Ouality contro Ouality control Prelerred bydesigner's Assembly c0mpany Assembly Assemb,, Transport Transport Transport 0peration Operation 0peration Mantenance Maintenance Maintenance Costs Costs Costs Schedules Schedu les isable Real i np r i n c i p l e Incorporates direcl measures salely Durabi ity(Stress) Deformation ity Stabil Resonance Expansion Creep Reaxation Corrosion Wear Ergonomics Standards Product on Assembly contro Ouality Transport 0peration Maintenance procedures Selection Discovering ldentifying optimumChecking concept embodiment optimum embodiment Determining layout, l0rms andmaterials (Figure (Figure 5 60) (Figure 6 2) 6 136) \J ol tasks 0ivision Seif-help andp anned Stability itv instabi b u r 0Ien e s ldentilying thebest combination ol prnctples Fullils demands ol thespecilication o B a s ircu l e s :i m p l i c i t y , | 6 . 3 . clarity, safety Evaluating principle Working principle Working o ooo Solutonconcept >t Embodiment design Selecti ng a= = Specilicat on Z 6 s!? s E Conceptual design 8.3. Summary of checklists with main headings o o o Figure 8.2. Correlation of mcthocls anrl airls with thc stcps ol' thc cmtr<lclimcntdcsiytt phase (numbcrs ref'crto chaplcrs an(l scclion$) )es', renergy', 'material' and 'signals',all of which facilitate the identification description of th'e overall function. Similarly, in the embodiment phase, the 'cmbodiment' ing is replaced with the appropriate 'layout and form design' r i r c t c r i s t i c s .S i r n i l i r r h c a d i n g s a p p l y t o e v a l u a t i o n ; t h e y h a v e a w e l c o m e t u r t c l r r r rw c yh i c h c n s u r c s t h i r l t h c y c < l v c ra l l c o n t i n g c n c i e s . S o t t t c o l l l t c t t t c l l t o d si l n ( l i r i ( l sw c h i l v c c x a n l i n c d i l r c a p p l i c a b l ea t d i f f e r e n t 8.2 Time commltment :ls of embodiment and can therefore be used repeatedly. This is particularly casewith the documentation (for example, specification,function structure, ction and evaluation charts). Moreover, it has been found that systematically )orated documents for certain product groups have a wider application in : they can be used for other products, thus reducing the overall cost of thc ematic approach. I Introduction I Time commitment meet the objection that systematicproceduresare very time-consumlng,w(' e broken down the man-hours spent on each step of the conceptual phast' gure 8.4). These percentagesare basedon practical experienceto date at thc Steps En9 Clarfyinqthetask oroblens essential lo 'derlrly Absl.actrng 10v, function structures Establishing Intuitive e g bralnstormlng Searchtno lor solutions drscursrve y ngquaitative principles andselect solution Combining P r e l i m i ncaar iyc u l a t t 0 n s F i r m i nr roni n l o varlants Prelim concept narylayouts References 15% 1.1 Bach,,C., 1880(12th.ed 7920).Die Machinenelemente. Stuttgart:Arnold Bergstrdsser Verlagsbuchhandlung. 7.2 Bar, S., 1970.Aufgabe und Stellungdes Konstrukteursbei der Schwerindustrie, Konstruktion22. 1-5. 1.3 Beitz, w., 1971. .systemtechnikim Ingenieurbereich,vDl-Berichte 174 (with _ further bibliog.refs.). Diisseldorf:VDI-Verlag. 1.4 Beitz,w., 1970.systemtechnik in der Konstruklion, DIN-Mitteitungen 49,295-302. 1.5 Beitz, W., W. Eversheimand G. Pahl, 1974.Rechnerunterstritztes Entwickeln und Konstruieren. im Maschinenbau, Forschungshefte;Forschungkuratoriunt Maschinenbal,Vol. 28, Frankfurt:Maschinenbau-Verlig. 1.6 Bischoff,W. and F. Hansen,1953.Rationelles KonstruiZren. Berlin: VEB-Verlag Technik. 1.7 Bock, A., 1955.Konstruktionssystematik-die Methodeder ordnendenGesichtspunkte, Feingercitetechnik 4. 3Y. )E o/ ?q o/ varlants concePl Evaluating r00% phase(estinlirtt' of breakdownof man-hoursspenton the conceptual ,ure8 4. Percentage on experience) ;ed iversity and in industry. The greatest percentageis devoted to conventitlrtltl .ivities, that is to firming up into concept variants which involves prelimirtrrty culations and layouts. A systematic approach during the conceptual rvides, with very little extra time expenditure, a broader over-vlew il tmises a greater chance of arriving at an optimum solution. Things arc ttr ry different in the embodiment phase. By consulting the checklists lowing the basic rules, principles and guidelines,the designerwill usually I le to save time and effort. Checking with the help of fault-identificirtir : t h o d s , m o r e o v e r , h e l p s t o i m p r o v c q u a l i t y a n d o n l y b e c o m e su n a c c c p l i r l . v a l u i t t i o n sd o n o t c o n s u t l l cl t r < t r i o u si f i t i s n o t c o n f i n e d t o t h c c s s c n t i a l sE r c h t i m c w h c n o n c c o n s i c l c r st l t c i t t l i l r n t i t t i o nt h c y y i c l d , c s p c c i a l l yi t t t l t d l r c S l i l r w c a k l i n k s .A n y o n c l l r r r r i l i i rwr i t l r t h c n l c t l r ( x l sc i l n . i t t a n y c i t s c .i l l ' r l v o S r t t l d c r a n d b c t t c r r c s u l t s i r t i r r c l i t l i v c l y s h o r t t i t t t c . l r t p i t r t i c t r l i r r ,t l t d i l c r r y t t i c u p p r o i t c h h o k l s o r r l i t B r c n l c t p r ( ) m i l i co l l t v r t i t l i r t gt i t t t c - c o t t s t t t l t t t t l r o r s t l t r c t t t i t l i t c k o l ' i t t l o r t t t i r l i ( l l tl l l l t t l r a l l t , 1 . 1 2E r k e n s ,A . , 1 9 2 8 B . e i t r r i g ez u r K o n s t r u k t i o n s e r z i e h uZn.gv, D I 7 2 , l 7 - z r . 1.13 Eversheim,W., 1969 E,ineanalytische Betrachtungv"onKonstruktionsaufgaben, Industrieanzeiger 97, Vol. 87. 1.14 Federn,K., 1970 wandel in der konstruktivenGestalttng,Chem.-Ing-Tec.hn.42, 729 731. 1 . 1 9K e s s e l r i n gF,, 7 9 4 2 D i e s t a r k eK o n s t r u k t i o nz,. v D I 1 1 6 , 3 2 l - 3 3 0 j 4, 9 - 7 5 2 . 1.20 Kesselring, F., 1954.Technische Kompositionslehre. Berljn,Gottingen,Heidelberg: Springer. l.2l Kesselring,F., 1951.. Bewertungvon Konstruktionen. Diisseldorf:VDl-verlag. 1.22 Koller. R.' 1973.Eine algorithmisch-physikalisch orientierteKonstruktionsmethodik. Z. vt)t lt5. 147_152,309_317, 843_847,1078_1085. 1 23 Ktrllcr, I{.. 1976.Krtnslruktionsmethctde ftir den Maschinen-,Geriite-untl Apparate/ r a r r .l l c r l i n , I l c i t l c l t r c r gN, c w Y o r k : S p r i n g e r . l . l ' l L i r r r t l i c r rK, . . l 9 l l , M u : c l t i t t t ' t t t , ! u n r , nI -i < . c' .i y . r z D i gr: M a x J u n e c k ev e r l a g s b u c I rl rl r n rIrl rr r t , R e f e r e n c .' Vols. 1-6 technrca-senc: Leyer, A., 1963-197I. Maschinenkonstruktionslehre, Basel,Stuttgart:Birkhauser Martyrer, E., 1960.Der Ingenieurund dasKonstruieren,Konstruktion72, 1-4. desallgemeinenMaschinenbutrs Matousek, R., 1957(reprint). Konstruktionslehre Berlin, Heidelberg,New York: Springer Heuristtk, SchriftenzLtr .\(l Muller, J., 1,970.Grundlagen der systematischen Wirtschaftsftihrung. Berlin: Dietz. Yol 1 Berlin: Gottingen,Heidelbcrl Niemann, G., 1950.Maschinenelemente, Springer(2nd ed. 1965,3rd ed. (with M. Hirt) 1975). Bcrlrrr einesGenies,Wiesbaden, N. N., 1977.Leonardoda Vinci, Das Lebensbild Vollmer, 439-505. Opitz, H. et al., 1977.Die Konstruktion-ein Schwerpunktder Rationalisienrrrl Industrie-Anzeiger93, 1491-1503. die rn,,,l unterstiitzen und Konstruktionslehre Pahl, G., 1967.Entwurfsingenieur erne Konstruktionsarbeit,Konstruktion19, 337-344 Konstruktiort'l Pahl, G. and W. Beitz, L912-74 Fiir die Konstruktionspraxis, (1972),2s (1973) and 26 (1e74). 46. .l I I Penny, R. K., 1970, Principlesof engineeringdesign, Postgraduate 349. Redtenbacher,F., 1852. Prinzipien der Mechanik und des M a s c h i n e n l t t t t t ' 257-290 Mannheim:Basserman. Reuleaux, F. and C. Moll, 1854 Konstrukstionslehre fiir den M a s c h i n e r t l t , t r r Brunswick: Vieweg. Richter, A , 1974. Nichtlineare Optimierung signalverarbeitenderGerate, I,/t/ Berichte 219 (with further bibliog. refs ) Diisseldorf: VDI-Verlag. Richter, A. and G. Kranz, 1974. Ein Beitrag zur nichtlinearen Optimierung trtttl dynamischenProgrammierungin der rechnerunterstritztenKonstruktion, Konstr trl'' tion 26. 36I-367. Riedler, A , 1913. Das Maschinenzeichnen.Berlin: Springer. Rodenacker, W. G. 1,970. Methodisches Konstruieren. Berlin, Heidelberg. Ncw York: Springer (2nd ed. 1976.) Rodenacker, W. G. and U. Claussen, 1973-74. Regeln des Methodischen K, struierens Mainz: Krausskopf. Rotscher, F., 192'7. Die Maschinenelemente.Berlin: Springer. Roth, K., 1974. Aufbau und Handhabung von Konstruktionskatalogen, l'l) Berichte 219. Diisseldorf: VDI-Verlas. rI Roth, K., 1968. Gliederung und Rahmen einer neuen Maschinen-Gerate-Konst stionslehre, Feinw erktechnik 72, 521-528. Roth, K., 1970. Systematik der Maschinen und ihrer mechanischenelemcttlrrt Funktionen, Feinwerktechnik 74, 453460 Roth, K. , f974. Methodisches Ermitteln von Funktionsstruktur und (icstrr VDI-Berichte 219 Diisseldorf: VDI-Verlas. Roth. K., H. J Franke and R Simonek, 1972.Die Allgemeine F u n k t i o n s srtr rk I ein wesentliches Hilfsmittel zum methodischen Konstruieren, K o n s t r u k t i r r .nl 277-282. I { o t h , K . , H . J . F r a n k e a n d R S i r n o r r c k ,l 9 7 l A l g o r i t h m i s c h e sA u s w a h l v c r l l r l r r 75, 337-345 zur Konstruktion mit Katalogctr, I"cittv,arktccltrtik I { p t [ . K . . I I . J . F r a n k c a n d I { S i n r o n c k , 1 9 7 2 .A u t ] r i r u u t t c l V c r w c t t t l t t t r gr t K i r t r r l r r e c Ir 'rr i rd l s r l c t h o t l i s c h cK o t t s t t t t i c t c t t .K t t n : ; l n t k l i t t t2t 4 , 1 4 9 4 5 t j S i r l i n g ,K . - 1 1 . . 1 9 7 0 ,I ' r i n z i p -r r r r t lV ; r r i r r r t l c r t k o r t s l r t t k ti ri ot t ltc r A t t l t r t t g s r r l r r\ r. r , I r r r r g V o n r u s s c t z u r ' t l l c rnr r r r l ( i r r t r t r l l r r g c r rl ,' l U - l l r ' r i L l t t t 1 5 2 , I ) i i s s c l t l t r r l :\ ' l t l ' Vcll;rg. S i r r r o r r c kl.t , l ( ) 7 1 l ) i t . k r r r r r l l r r k l r v cl , t t n k l t o t tu t t t l i l t r c l ; o t t t t t t l i e r t t t t tgt t r l , r t r ( ' ( ' l l n ( t' r( ' \ l t t l z t r ' K o t t r l tl l t ( ' l r l t , I t t n x e t A t a ht n t l - 7 s . l ' l \ t ' l ( , References 369 1 52 Sorensen,E., 1958.Konstruieren-schopferische Ingenieurarbeit, Z. vDI 100, 1123_1128. 1'53 Tschochner,H. , 7954.Konstruieren und GestaltenE,ssen: Girardet. 1.54 vDI-FACHGRUPPE Konstruktion(ADKI), 1967.EngpassKonstruktion Kon_ struktion19,192-795. 1'55 VDI-Richtlinie2222,1973.Sheet1 Konzipierentechnischer Produkte.Dtisseldorf: VDI-Verlas. 1'56 VDl-Richtlinie 2225, 1969.Technisch-wirtschaftliches Konstruieren.Dirsselclorf: VDI-Verlag. 1.57 Wachtler,R , 1969.Die DynamikdesEntwickelns(Konstruierens), Feinwerktechnik 73. 329-333. 1.58 Wachtler,R., 1967.Beitragzur TheoriedesEntwickelns(Kontruierens),Feinwerktechnik7l, 353,357. 1.59 wdchtter, R., 1971Entwickelnund Konstruieren-Tdtigkeiten, die wachsendem Zeitdruck ausgesetztsind, VDI-Nachrichten25. No. 6. 1.60 wogerbauer,H., 1943.Die Technikdes Konsiruierens,2nd ed., Munich, Berlin: Oldenbourg. 7.61 zangemeister, c., 1969.Zur charakteristikder Systemtechnik. TU Berlin (AufbauseminarSystemtechnik). 2 Fundamentals 2' 1 Branlrm p . K., 1975 . Produktivitritssteigerung in clermittelstrindigen IndustrieN RW. VDI-Taschenbuch. Diisseldorf:VDI-Verlae. 2.2 DIN 44 300, 1972. Informationsverarbeiting-Begriffe.Berrin, cologne: BeuthVertrieb. 2.3 DIN 44 301. 1973. Informationstheorie-Begrtffe( Normentwurf).Berlin, cologne: Beuth-Vertrieb. 2.4 DIN 69 910. 1973. wertanatyse,Begriffe,Methode. Berlin, cologne: BeuthVertrieb. 2.11 Klaus, G., 197r.w_drterbuch clerKybernetlk.Frankfurt,Hamburg:Fischcr. 2'12 Koller, R., 1973.Eine algorithmisc-h-physikalisch orientierteKonitruktionsr.ncthodlk. z. vDI ll5. 2'13 Koller' R', 1974.Kann der Konstruktionsprozess in Algorithmengetasstuncldcnr ^ .. S..hl.r tibertragenwerden?vDI-Berichie219.Dtisseliorf:vDl-Verlag. 2 14 Krumhauer,P., 1974.Rechnerunterstiitzung fiir die Konzeptphase der Konstrukt i o n . ( D i s s e r t a t i o nT)U D 8 3 . B e r l i n :T U . 2 l-5 Mcwes, D.' 1973.Der Informationsbedarfim konstruktivenMaschinenbau.yDlTlschcnbuchT 49 Diisseldorf:VDI-Verl"as. 2 l(r Miillcr' .1.' 1970. Grundla{'n der system"atischcn Heuristik. Schrilten zLtr soz. Wi rt.\( ltuI t.\| t:ihrrrrrg Berlin: Dietz. (;.. 1967 work.sv,srcnts l l7 Nrrtflcr'. Da,signTheItlealsConceptHomewood,Illinois: I t i t ' l r ; r l It )l l l r v i r r I. r r c , Referenccs Nadler, G., 1963 Work Design. Homewood, Illinois: Richard D lrwin, Inc. n.n. (anon.), L958. Lexikon;der Neue Brockhaus. wiesbaden: F. A. Brockhaus. fiir Schnittwerte, wt. Z. industriellc Opitz, H., 1970. INFoS-Informationszentrum Fertigung 13. Opiti, H., K. Brankamp and H.-P. Wiendahl, 1970.Aufbau und Anwendung eines funktionsorientierten Baugruppenklassifizierungssystems, Industrie-Anzeiger 9) (1e7031). Petzold, H -J., J. Pohlmann and W. Haag, 1974. Modelle der innerbetrieblichert Informationsversorgung.BetriebstechnischeSchriftenreihe RKW/REFA. Berlirt Cologne, Frankfurt: Beuth-Verlag GmbH. Rodenacker, W. G., 1970 (2nd ed. 1976). Methodisches Konstruieren. Konstruk tionsbiicher Yol 27. Berlin, Heidelberg, New York: Springer. Rodenacker, W. G. and U Claussen, 1973 and 1975. Regeln des Methodiscltttt Konstruierens,Vols. I and II. Mainz: Krausskopf Roth, K., H.-J. Frank and R. Simonek, 1972. Die allgemeine Funktionsstruktrrr ein wesentliches Hilfsmittel zum methodischen Konstruieren, Konstruktion 2J 277-282. Schon, F., \973. Wirtschaftlicher Konstruieren durch bessereInformation, v/)/ Berichte 191. Drisseldorf: VDI-Verlag. VDl-Richtlinie 2222. 1973. Sheet 1. Konstruktionsmethodik, Konzipieren tctlr nischer Produkte. Disseldorf: VDI-Verlag. Wertanalyse. Dtisseldorf: VI) I VDl-Richtlinie 2801, 197017I, Sheets l-3 Verlag. Voigt, C.-D.,1974 (3rd ed.). Systematik und Einsatz der Wertanalyse.Munitlr Siemens-Verlag. Weizsacker von, C F., 1971. Die Einheit der Natur-Studien. Munich: Ilanser Wolf, S., 1970. Informationssystemfirr Maschinerlbau-Unternehmenmit E,inzcl und Serienfertigung, wt.-Z. industrielle Fertigung 60, 26-39. Zimmermann, D., 1972. Produktionsfaktor Information, Yol. 72 of: Wirtschu.lrt filhrung- Kybernetik- Daten verarbeitung. Neuwied: Luchterhand. Zimmermann, D., 7971. StrukturgerechteDatenorganisation,Vot. 10 of: Witr schafts-fiihr ung- Ky b ernatik- D atenvera rb eit u ng. Neuwied : Luchterhand' Zimmermann , o., pel . ZAFO-Eine allgemeine Formenordnung fur werk.stilt Lt Stuttgart: Grossmann. re design process l e e l i c h . K . H . a n d H . H . S c h w e d e , l 9 7 4 .L e r n - u n d A r b e i t s t e c h n i kw. i i r z b u r g : /ogel. <riit, V., 1969.An Introduction to Engineering and Engineering Design, Second ldition, New York, London, Sidney, Toronto: Wiley & Sons Inc' _eyer,A., 1973.Zv Frage der AufsAtze irber Maschinenkonstruktionin der technika', technika 26, 2495-2498. ) a h l . G . . l g T 2 . D i e A r b e i t s s c h r i t t e b e i m K o n s t r u i e r e n ,K o n s t r u k t i o n 2 4 , 1 4 9 - 1 5 . \ ' 4C 344-.1'l() )cnny, R. K., 1970.Principlesof Engineering Design, PostSraduu.te.J. . tisscltlorl r ' D l - R i c h t l i n ie 2 2 2 2 , 1 9 7 3 ,S h e c t l , K o n z i p i c r c nt e c h r r i s c h ePr r o c l u k t e D y ' l )l - V c r l a u . y l ) l - l { i c h t i i n i c 2 2 2 3 . 1 9 6 9 I } c g r i l l c r r n t ll } c z c i c l r r t u r t g ci nt tt K o n s t l u k t i o n s b c r c i c l t ) t i s s c l t l o l:l V I ) I - V c t l r t g . c r l l s l l t r k t i o l r( A l ) K I ) , l ( ) ( r ( rl .: r t t P l c l t l t t t t g t " t t A u s d c r A r . b c i td c l V I ) i - l r r r c h g l r r l ) l )K 'tir l l c g r i l t c r r r r t l l c z c i c h r r r r nigr r r . ( o r r s l t t t k l i o t t r l l c r c i c lKt .t t t t . t t r u k t i t tl tf t{ ' . 1 9 ( }. 1 ( ) l Ncttr|tt'tl Wirlrl,"M, lt., l()(r(),(;,ilnll(t!.'n rinrt lllunugerrrfnt'lnltrtrrttlri{,rrr'f}'rl('rtl('r' l l c r l i r r :I i r c l l t c l l t i r t t ( 1 , References 371 4 Product planning and clarification of the task 4.1 Arlt, J., 1971. Dynamische Produktionsprogrammplanung. Dissertation TH Aachen. 4.2 Brankamp, K., I974. Produktplanung-Instrument der Zukunftssicherungim Unternehmen, Konstruktio n 26, 379-327. 4.3 Franke, H.-J , 1975Methodische Schritte beim Kldren konstruktiver Aufgabenstellungen, Konstruktion 27, 395-402 4.4 Geyer, E., 1972. Marktgerechte Produktplanung und Produktentwicklung. Part l: Produkt und Markt Part II: Produkt und Betrieb. RKW-Schriftenreihe 18 & 26 Heidelberg: Gehlsen. 4.5 Hansen, F., 1966. Konstruktionssystematik.Berlin: VEB-Verlag Technik 4.6 Kehrmann, H., 1912 Die Entwicklung von Produktstrategien Dissertation. TH Aachen. 4.7 Kehrmann, H., 1973.Systematikzum Finden und Bewerten neuer Produkte, wt.-Z ind. Fertigung 63, 607-612. 4.8 Kesselring, F. and E Arn, 1971. MethodischesPlanen, Entwickeln und Gestalten technischerProdukte, Konstruktion 23, 212-218. 4.9 Kramer, F., 1975. Produktinnovations- und Produkteinftihrungssystemeines mittleren Industriebetriebes,Konstruktion 27. I-7. 4 . 1 0 P a h l , G . , 1 9 7 4 . K l d r e n d e r A u f g a b e n s t e l l u n gu n d E r a r b e i t u n gd e r A n f o r d c r u n g s liste. Konstuktion 24,195 199. 411 Pahl, G.,1970. Wege zur Losungsfindung,Industrielle Organisation39. 1-5-5-161. 4.12 Rodenacker, W G., 1970, 2nd ed. L976 MethodischesKonstruieren. Konstruk' tionsbilcherYol. 27. Berlin, Heidelberg, New York: Springer. 4.13 Roth, K., H Birkhofer and M. E,rsoy, 1975. Methodisches Konstruieren neuer Z. -V Dl 117, 613-618 Sicherheitsschlosser, 4.14 Schmitz, H., 1972. Produktplanung. VDI-Taschenbuch T 32 Diisseldorf: VDIVerlag. 4.15 Stotko, C , 1973. Produktplanung-massgebend fi.ir den Unternehmenserfolg, Industrielle Organisation 42, 2ll-214. und Aufgabe. 229, 1976. Produktinnovation-Herausforderung 4.16 VDI-Beichte Diisseldorf: VDI-Verlag. 4 17 VDl-Taschenbuch T 46,1975. SystematischeProduktplanung-ein Mittel zur Unternehmenssicherung.Dirsseldorf: VDI-Verlag. 5 Conceptual design 5. 1 Baatz, lJ., 1971. BildschirmunsterstirtztesKonstruieren, Diss., RWTFI Aachcn. 5.2 Beitz. W., 1968. MesstechnischeHilfsmittel des Konstrukteurs, gezcigt an llcis' nbau, Antriebstechni k 7' 3-5u-362' oielen aus dem Elektrogrossmaschine 5 . 3 b e i t z , W . , 1 9 7 3 . M e t h o d i s c h e s K o n z i p i e r e n t e c h n i s c h e rS y s t c m c . g c z c i g t a r t t Beispiel einer Kartoffel-Vollerntemaschine.Konstuktiott 25' 65-1 | 5.4 Bengisu, O., 1SZO.ElektrohydraulischeAnalogie, Olhydruulik untl Prtautttutik14. 122 t27 5.5 Boesch, E ,1954. Die Organisation industrieller Forschung.Industrielle Organisation 23.335-344. ,5.(rBrader. C. and G. Hohl, 1974. Rechnereinsatzin der Konzeptphasedes KonstrukV DI-Berichle 219. Diisseldorf: VDI-Verlag tionspr<rzesses, ,5.7 ('hladck, W., 1971.TechnologischeVoraussagen,Z' VDI Il3, 1277-1225. 5.il ('lirusscn. LJ., 197I Konstruieren mit Rechnern, Konstruktionsbticher29. Berlin, I l c i d c l t r c r g .N c w Y o r k : S p r i n g e r . l p p l i c a t i o no f t h e D e l p h i 5 . ( ) t ) r r l k c v . N . I ) . r r n t l( ) . I l c l n t c r . l 9 6 l A n I l x p e r i r n e n t a A ( il( t t ( c V o l 9 . N o , 3 ( A p r i l 1 9 6 3 ) M c t l r . r r lt o t l t c I l s t l o f l : x l r c r l s , M t t r r t t g t ' r t r <, \' n 3tz Refe rc n cr't 5.10 DIN 40 7000, 1,963 Schaltzeichen-Digitale Informationsverarbeitung. Berlin, Col ogne: Beuth-Vertrieb. 5 . 1 1 D r e i b h o l z , D . , l 9 7 5 . O r d n u n g s s c h e m a tbae i d e r S u c h ev o n L o s u n g e n .K o n s t r u k t i o r t 21.233-240. 512 E,wald, O.,1975. Losungssammlungenfiir das methodischeKonstruieren Disscl dorf: VDI-Verlag. -5 13 Feldmann, K., 1,974.Beitrag zur Konstruktionsoptimierung von automatischt'rr Drehmaschinen. Diss., TU Berlin 5.14 Findeisen, D., 1974. Dynamisches System Schwingpriifmaschine Fortschritr-lJ, richte der VDI-Zeitschriften, Ser 11, No. l8 Dusseldorf: VDI-Verlag. 5.15 Follinger, O. and W Weber, 1.967.Methoden der SchalttrlgebraMunich: C)lclerr bourg. 5.16 Gerber, H., 1973.Ein Konstruktionsverfahrenfur Gerdte mit logischerFunktiorr. weise, Konslruktion 25, 13 l1 5 . 1 7 G e s c h k a ,H . a n d H . W i g g e r t , 1 9 6 8 S u c h em i t S y s t e m ,D e r V o l k s w i r t 3 6 3 T 5.18 Gordon, W J. J., i961 Synectics,the developmentof Ueative capacity.New Yorl' Harper. 5 . 1 9 G u n t h e r , W . , 1 9 7 1 .D i e G r u n d l a g e nd e r W e r t a n a l y s e ,Z . V D I 1 1 3 , 2 3 8 - 2 4 1 5.20 Herrmann, J , 1970. Beitrag zur optimalen Arbeitsraumgestaltungan numcri:, lr gesteuertenDrehmaschinen, Diss., TU Bertin. 5.21 Hertel, H., 1960. Biologie und Technik Struktur, Form, Bewegang. Mirrrr; Krauskopf. 5.22 Hertel, H., 1960 Leichtbau. Berlin, Gottingen, Heidelberg: Springer. 5 23 Keller, K., 1971 Entwicklung eines Fertigungsautomaten zur Fertigung r,'rr Filigrantragern, unpublished thesis, TU Berlin 5.24 Kesselring, F., 7951 Bewertung von Konstruktionen, ein Minel zur Stetterurtt!t,'rt Ko nstruktionsarbeit. Drisseldorf: VDI-Verlag. 5 2 5 K o l l e r , R . , 1 9 7 3 .E i n e a l g o r i t h m i s c h - p h y s i k a l i s oc rhi e n t i e r t eK o n s t r u k t i o n s m c l l t o dik. z. vDl ll5, 147-r52,309-31'7,843-847, 1078-108-5. 5.26 Koller, R., 1974. Kann der Konstruktionsprozessin Algorithmen gefasstuncl tl, ttr Rechner iibertragen werden VDI-Berichte 219. Drisseldorf: VDI-Verlag. .5.27 Koller, R., 1976 Konstruktionsmethodefur den Maschinen-,Gerrite-und Ayslturttr, bau. Berlin, Heidelberg, New York: Springer 5 . 2 1 1K r a m e r , F . , 1 9 7 5 .P r o d u k t i n n o v a t i o n su- n d P r o d u k t e i n f u h r u n g s s y s t eemi n c s n ) r t t leren Industriebetriebes,Konstruktion 27, | 7 -5.29Krick, E V, 1969. An Introduction to Engineering and Engineering Desigtt. -'ttl edn., New York: Wiley & Sons,Inc. 5 30 Krumhauer, P I974. Rechnerunterstiitzungfiir die Konzeptphase der Kortsttrrl tion, Diss., TU Berlin. 5 . 3 1 L o w k a , D . , 1 9 7 5 M e t h o d e n z u r E n t s c h e i d u n g s f i n d u nigm K o n s t r u k t i o n s l . r t t z t ' r r , Feinwerktechnikund Messtechnik38. 19-21. 5 . 3 2 M i l i t z e r , O . , 1 9 7 5 .R e c h e n m o d e l lf u r d i e A u s l e g u n gv o n W e l l e n - N a b e n - P a s s l t ' r h ' r v e r b i n d u n g e n ,D i s s . . T U B e r l i n . n , M i l t t , t ' 1t , 5 3 3 n . n . ( a n o n . ) , 1 9 7 2 ,K r e a t i v i t a t D o k u m e n t a t i o nd e r 4 3 K r e a t i v - M e t h o d e Mugazin l l, -51--57 .5.34Osborn, A. F.. 1957.Applicd Imagination-['rincipla.s und Procctlurcso.l ( t(tttt\( 'l'hinking Ncw York: Scribner . t r l s t c l l c nv o r t I i t t t ) k l i r r t r o 5 . 1 5 l , i r h l .G , 1 9 7 2 A n a l y s cu n d A b s t r i r k t i o rrtl c sI ' t o b l c n t s A s t r t t k t t t r c t tK. o t t . s l r t t k l i r2t r. 1 r ,2 3 5 l l l ) , 5 , . 1 ( lr, r r l r l (, i . , I 9 7 4 M c t h o r l i s c l r cKs ( ) n \ l t u i ( ' t ( ' r l /, l ) l " l l u i t l t t r l l ( ) , l ) t i s s c l t l t r r l\ :' l r l V c rl r r g 5 , 1 7 l , i r l r l ,( ; , l 9 7 l l { i i c k h l i c kz r r r l { c i l r t " l t t t r l t c K o t t r l t t t h t i o t t s I r t \ i s '1, r , , 1 1 1 1 1 1 1 1 ' 1 1 " t r l(1,.l()l .l()5, t l t { I t t r t z ,I l ( ) 7 \ I ) i ! : r t r t l t lrt' trt r l , |. ) t r r r r . l r hl r f V l ) l V c r l ; r I References JIJ 5.39 Richter, A. and H.-J. Aschoff, 1971. Problemstellungenbei der funktionsorientierten KonstruktionssynthesesignalverarbeitenderGerite aus der Sicht der Systemdynamik. Feinwerktechnik 75. 37+379. 5.40 Richter. A. and H.-J. Aschoff ,7911. Uber die funktionsorientierte konstruktive Gestaltung von signalverarbeitendenGerdten nach statistischenund dynamischen Gesichtspunkten, Feinwerktechnik 75, 443-416. 5.41 Rodenacker, W. G., 1972. Festlegung der Funktionsstruktur von Maschinen, Apparaten und Geraten, Konstruktion 24, 335-340 5.42 Rodenacker, W. G., 1970. MethodischesKonstruieren, KonstruktionsbilcherYol 27. Berlin, Heidelberg, New York: Springer 5.43 Rodenacker, W. G., 7976. MethodischesKonstruieren, KonstruktionsbilcherYol 27.2nd edition. Berlin, Heidelberg, New York: Springer. 5.44 Rodenacker, W. G. and U. Claussen, 1973 Regeln des Methodischen Konstruierens.Mainz: Krausskoof. 5.45 Rohrbach, B., 1969. Kreativ nach Regeln-Methode 635, eine neue Technik zum Losen von Problemen Absatzwirtschaft 12, 13-75 5.46 Roth, K., 1974. Aufbau und Handhabung von Konstruktionskatalogen, VD1Berichte 219. Diisseldorf: VDI-Verlag. 5.47 Roth, K., 1970. Systematik der Maschinen und ihrer mechanischenelementaren Funktionen Feinwerktechnik 74. 453460. 5 48 Roth, K., 1974 Systematik mechanischer Flip-Flops und ihre Bedeutung ftir die Konstruktion von Schaltelementen, Feinwerktechnik und Messrcchnik 82, 384-392 5 . 4 9 R o t h , K , H . B i r k h o f e r a n d M . E r s o y , 1 9 7 5 . M e t h o d i s c h e sK o n s t r u i e r e n n e u e r Z VDI ll7, 613-618. Sicherheitsgurtschlosser, 5.50 Roth, K., H.-J. Franke and R. Simonek,19'72 Die Allegemeine Funktionsstruktur, ein wesentliches Hilfsmittel zum methodischen Konstruieren, Konstruktion 24, 217-282 5.51 Roth, K., H.-J. Franke and R. Simonek, 1972 Aufbau und Verwendung von Katalogen fiir das methodische Konstruieren. Konstruktion 24, 449-458. 5 . 5 2 S c h l o s s e rW , . M . J . a n d W . F . T . C . O l d e r a a n , 1 9 6 1 .E i n e A n a l o g o n t h e o r i ed e r Antriebe mit rotierender Bewegung. Othydrautik und Pneumatik 5,413-418 5.53 Schmidt, H. G., 1973. Entwicklung von Konstruktionsprinzipien fiir einen Stosspriifstand mit Hilfe konstruktionssystematischerMethoden. Study Project, TU Berlin. 5.54 Simonek,R.,I974. EinVerfahrenzurErmittlungderSpeziellenFunktionsstruktur mit Hilfe der EDV, Feinwerktechnikund Micronic 78, lTl7. 5.55 Stabe, H and E Gerhard, 1974.Anregungen zur Bewertung technischerKonstruktionen. Feinwerktechnik und Messtechnik82, 378-383 (with further bibliog. rcl'crences). 5.56 Stahl, U.,1916. Uberlegungen zum E,influssder Gewichtung bei dcr Bcwcrtung von Alternativen, Kons truktion 28, 2'73-214. 5.57 Steuer, K., 1968. Theorie des Konstruierensin der Ingenieurausbiltlung.Lcipzig: VEB Fachbuchverlag. -5.58 VDl-Taschenbuch 135, t972 Wertanalyse-ldee, Methode, System 4. Di.isseldorf: VDI-Verlag. .5.59 VDI-Richtlinie 2222, Sheet 1,. 1975 Konzipieren technischerProdukte. Diisseldorf: VDI-Verlag. 5.fi) VDI-Richtlinie 2225, 1969. Technisch-wirtschaftliches Konstruieren. Diisseldorf: VDI-Verlag. 5.6 1 VDI-Richtlinie 2801, 1970 Wertanalyse.Diisseldorf: VDI-Verlag. 5.62 Voigt, (1.-D., l95U Systematik und Einsatz der Wertanalyse.Berlin, Munich: S i cr r r cr r s N4c.w Y o r k : I l c i n l r o l d 5 0 1 W i t l r i r r g (. ' , , l 9 5 l t , ( ' t t u t i v t ' I ' l r i r t k i r t 1 References 374 5.64 Zangemeister, C., 7970. Nutzwertanalvse in der Systemtechnik. Munich: Wittemannsche Buchhandlung. Entdecken, Erfinden, Forschen im Morphologischen Welt5.65 Zwicky, F., 1976119'71. blld Munich. Zwich: Droemer-Knaur 6 Embodiment design 6.1 AEG-Telefunken, 1971. Biegen. Werknormblar 5 N 8410. 6 2 Altmann . F. G 1962. Antriebstechnik. Z. VDI 104, 965-976 6 3 Andresen , U.,1915 Die Rationalisierungder Montage beginnt im Konstruktionsbiiro, Konstruktion 27, 478-484. Abr from: Ein Beitrag zum methodischen Konstruieren bei der montagegerechtenGestaltung von Teilen der Gross-serienfertigung, Diss. 1975, TU Braunschwerg. 6.4 Banninger, E., 1910 Normung-Zwangsjacke oder unentbehrlicher Helfer des Konstrukteurs? technica 21, 1947-19'72and 22, 2171-2117. 6.5 Bartel, A., 1963 Die Entstehung von Passungsrostbzw Reiboxydation, Ma,rchinenschaden 36 on 28.11.195U. 6 . 6 B a u m a n n , K . , 1 9 5 8 .M a n n e s m a n n - L e c t u r e 6.7 Bautz, W. and A. Thum, 1935. Die Gestaltsfestigkeit,Stahl und Eisen 55. 1025-1029 & Schweizer. Bauzeitung 106, 25-30. 6.8 Beckstroem, J., 1965 Eigenspannungenund Betriebsverhaltenvon Schweisskon struktionen. Konstruktion 17. l0-15. 6 . 9 B e e l i c h , K . H . , 1 9 7 3 .K r i e c h - u n d r e l a x a t i o n s g e r e c hKt ,o n s t r u k t i o n 2 5 , 4 l 5 4 2 1 6.10 Behnisch, H., 1973. Thermisches Trennen in der Metallbearbeitung-wirtschaft' lich und genau, ZwF 68,337-340. 6.ll Beitz, W., 1971. Normung und Systemtechnik-Grundlage fiir ganzheitlichc Betrachtungsweisein Konstruktion und Fertigung, DIN-Mitteilungen 50,378 381 6.12 Beitz, W., 1969. Moderne Konstruktionstechnik im Elektromaschinenbau,Kr.rrr struktion 21, 461468. 6.13 Beitz, W., 1973. Die normgerechte Konstruktion, Konstruktion25,319-327. 6.14 Beitz, W ., 1973. Fertigungs- und montagegerecht, Konstruktion 25, 489-497 6.15 Beitz, W. and H. Staudinger,1969. Guss im E,lektromaschinenbau,Konstruktiott 2t. t25 130. 6.16 Berg, S., 1969 Die besondere Eignung der Normzahlen ftir die Grossenstufurrg. D IN- Mitteilungen 48, 222 226. 6 17 Berg, S., 1965. Konstruieren in Grossenreihenmit Normzahlen, Konstruktion l7. 15-21. VDI-Verlag. 6.18 Berg, S., 1952. Gestaltfestigkeit: 6 . 1 9 B i e z e n o , C . B . a n d R G r a m m e l , 1 9 5 3 , T e c h n i s c h eD y n a m i k , V o l s 1 & 2 . l r r , l edn., Berlin, Gottingen, Heidelberg: Springer 1959. 6.20 Bowden, E. and D. Tabor, 1959. Reibung und Schmierungfester Korper Bcrlrrr, Gottingen, Heidelberg; Springer. Konstruieren.SchweizerDruck- trrt,l 6.21 Brandenberger, H., n.d Fertigungsgerechtes Verlagshaus 6 2 2 B r o i c h h a u s e n ,J . , 1 9 7 0 B e e i n f l u s s u n gd e r D a u e r h a l t b a r k e i tv o n K o n s t r u k t i o r t ' werkstoffen und Werkstoffverbindungcn durch konstruktive Kerben, Obe rl r i,c sl . N , ' l i i c h e n k e r b e nu n d m e t a l l u r g i s c h cK c r b c n . V D I - F o r t . s c ' h r i t t . s - B e r i cShet e 2 0 . D i i s s e l d o r f :V D I - V e r l a g . ( r . 2 3 [ J r o r r r r c rA, . , l 9 6 i 3 .W c r t l n u l v s c l l s i n t c B r i c t t cl { i r t i o n a l i s i c r t n t g , W c r k s t u l t \ t ( ( l t t t t L 5 U .N r I 6 , 2 . 1I l u r g r k r r ' 1 .M . . l t ) 7 3 , I ; l i c s s F r ' r c t ' l t t(ci c r l i t l l t t t t g v o t t W c r k s t ( i c k c t t .u ' t - Z n t r l l:crtig 63 . lti7 .192. 'l'tcrtttlttttclt ( r , 1 5 ( ' l l r r r s r r r r , v c rl,l , . l t ) ( r t ) , K r i t r r r ' h c t S P r u t t t t t t t g r l t t . , t l t ttrtlt t t l t t t t t rr r r l 1 l 5 ( , l l, t t t c l t r r t ' l t s i g t 'ltl t ' l t t t r l trt t tl t t t t t p ,A o l r l r l l l l f t References 315 Gestalten. Thesis, TH Aachen 6.26 Compes, P., 1970 Sicherheitstechnisches 6.27 Cornelius, E. A. and J Marlinghaus, 1967. Gestaltungvon Hartlotkonstruktionen hoher Tragfdhigkeit. Konstruktion 19, 321-327. 6.28 Cornu, O., 1973. Ultraschallschweissen,Z. TechnischeRunschau 37, 25-27. 6.29 Czerwenka, G. and W. Schnell, 1967. Einftihrung in die Rechenmethoden des Mannheim: BibliographisLeichtbaus I und II, Hochschultaschenbuch1.241124a. ches Institut. 6.30 Dangl, K., K. Baumann and W. Ruttmann, 1969.E,rfahrungenmit austenitischen Armaturen und Formstricken, Special issue: VG B-Werkstofftagung 1969, 98 6.31 Dhen, K., 1959 Aufgaben und Nutzen innerbetrieblicherNormungsarbeit, D1NMitteilungen 38, 7-11. 6.32 Dilling, H.-J. and T. Rauschenback,1975 Rationalisierungund Automatisierung der Montage, Diisseldorf: BDl-Verlag. 6.33 DIN, Krieg, K. G. (ed), 1975.Nationale und internationaleNormung, Handbuch der Normung, Vol. 1, 3rd edn., Berlin, Cologne: Beuth-Vertrieb. 6.34 DIN, 1975. Verzeichnis der Normen und Norm-Entwiirfe. Berlin, Cologne: Beuth-Vertrieb. 6.35 DIN 820, 1969. Sheet 2. Gestaltung von Normbliitten Berlin, Cologne: BeuthVertrieb. 6.36 DIN 82Q, 1975, Sheet 3. Normungsarbeit-Begriffe, Berlin, Cologne: BeuthVertrieb. 6 37 DIN 752f-7527. 7971 74. Schmiedesttickeaus Stahl. 6 38 DIN 8577,1974.Fertigungsverfahren;Ubersicirl.Berlin, Cologne: Beuth-Vertrieb 6.39 DIN 8580, 1974. Fertigungsverfahren.Einteih.rng.Berlin, Cologne: Beuth-Vertrieb. 6.40 DIN 8588, 1966. Fertigungsverfahren. Verteilen-Einordnung, Unterteilung, Begriffe Berlin, Cologne: Beuth-Vertrieb. 6.41 DIN 8593,1961 Fertigungsverfahren Ftigen-Einordnung, Unterteilung, Begriffe. Berlin, Cologne: Beuth-Vertrieb. 4. G esenkschmiedestuckeaus Magnesium-Knetlegierungen Berlin, 6.42 DIN 9005, 19731 Cologne: Beuth-Vertrieb 6.43 DIN 31 000, 1977. Sicherheitsgerechtes Gestalten technischer Erzeugnisse Allgemeine Leitsritze Berlin, Cologne: Beuth-Vertrieb 6 . 4 4 D I N 3 1 0 0 1 , 1 9 7 4 , S h e e t s 1 , 2 a n d 1 0 . S c h u t z e i n r i c h t u n g e nB.e r l i n , C o l o g n e : Beuth-Vertrieb. 6.45 DIN 31051, 1974, Sheet l. Instandhaltung: Begriffe. Berlin, Cologne: BeuthVertrieb. 6.46 DIN 33400. 1974. Entwurf: Gestalten nach arbeitswissenschaJtlichenErkennttrissen, Berlin, Cologne: Beuth-Vertrieb. 6.47 DIN33401, 1974.Entwurf. Stellteile.DIN33 402,1914.Entwurl: K0r1tcrnm.t.tt"r'on Erwachsenen Berlin, Cologne: Beuth-Vertrieb. 6.48 DIN 40041, 1967 Zuverldssigkeit elektrischer Bauelemantc: IlegrilJL'. Ilcrlin. Cologne: Beuth-Vertrieb. 6.49 DIN 50960, 1963. Korrosionsschutz, galvanische Uberztiga. Berlin, Ctllogrlc: Beuth-Vertrieb 6.50 DNA, 1970. Die Zusammenfiihrung der elktrotechnischenNormen- und Vorschriftenarbeit in Deutschland, DIN-Mineilungen 49, 401-471' 6.51 DNA, Krieg, K. G., 1914. Innerbetriebliche Normungsarbeit. Handbuch der Normung, Vol. 2. Berlin, Cologne: Beuth-Vertrieb. 6.52 DNA-Grundnormen fiir die mechanischeTechnik. DlN-Taschenbuch I. Berlin, ( l o l o g r r e: B c u t h - V e r t r i e b . 651 I)Nn, Kricg. K. G. (ed), 1972: Normung als Instrument der Unternehr r r c r r s l c i t r r n gl l u t t t l l t u t l r d c r N o r r n u n g , V o l 3 B e r l i n , C o l o g n e : B e u t h Vcrtrit'h, References 376 Festlegungen Berlin, 6 54 DNA, 1971. Normenverzeichnis mit sicherheitstechnischen Beuth-Vertrieb. : Cologne vielseitiges Ferti-'-- von bobeneck, O., l9lZ. Die Elektronenstrahltechnik-ein 6.55 gungru"rfahren, Feinwerktechnik und Micronic 77 ' 98-106' . Asthetik Munich: Moderne 6.56 bor"fles, G., 1964. Gute Industrieform und ihre Industrte. 6'5TEhrlenspiel,K,lg6gMehrweggetriebefiirTurbomaschinen,Z.-vDI111'218 dung. Berlin, Cologne: Beuth-Vertrieb' Backenbremsen, Konstruktitttt 6.62 Falli, K., lg6i. ThEorie und Auslegung einfacher 19,268-27r. 663Feldmann,HD.,lg5gKonstruktionsrichtlinienfiirKaltfliesspressteileausStahl. Konstruktion 11, 82-89. der Warmfestigkeit' Arclr 6 64 Florin, C. and H. i.g;unO, 1970.Uber die Grundlagen E isenhiittenw esen 41. 7'77-7 78' fiir den KonstrukO.OSFotter, D., 1975.MaschinenakustischeBerechnungsgrundlagen teur, VDI-Berichte 239 Diisseldorf: VDI-Verlag' Bestandteil einer Firmen6.66 Friedewald, H. J., 7970. Normung-integrieren{er References 6.80 Hartmann, A., 1959. Die Druckgefahrdung von Absperrschiebernbei Erwarmung des geschlossenenSchiebergehduses,VG B- Mitteilung, 1959, 303-307. 6.81 Hartmann, A., 1959 Schaden am Gehdusedeckeleines 20-atii-Dampfschiebers, VGB Mitteilung (1959), 315-316. 6.82 HAusler, N., 1974. Der Mechanismusder Biegemomentiibertragungin Schrumpfverbindungen. Diss , TH Farmstadt 6.83 Heckl, M., 1975. Minderung der Korperschallentstehungund Korperschallfortleitung bei Maschinen und Maschinenelementen,VDI-Berichte 239. Diisseldorf: VDI-Verlag. 6.84 Heckl, M. 1975. Konstruktive Moglichkeiten zur Minderung der Luftschallabstrahlung, VDI-Berichte 239. Diisseldorf: VDI-Verlag 6.85 Heesch, H. and O Kienzle, 7963. FltichenschlussBuchreihe Wissenschaftliche Normung. Berlin, Gottingen, Heidelberg: Springer. 6.86 Hertel, H , 1969. ErmiidLrngsfestigkeit der Konstruktionen. Berlin, Heidelberg, New York: Springer. 6.87 Hertel, H., 1960. Leichtbau Berlin, Gottingen, Heidelberg: Springer 6.88 Herzke, I.,197 1. Technologie und Wirtschaftlichkeit des Plasma-Abtragens,ZwF 66.284-291. 6.t39 Homig, H., \971. Metall und Wasser-Eine kleine Korrosionskunde E,ssen: Vulkan. 6.90 Hirskes, H. and W. Schmidt, 1972 Unterschiede im Kriechverhalten bei Raunrtemperatur von Stdhlen mit und ohne ausgepriigterStreckgrenze, DEW-I-echn Berichte 12,29 34. 6.91 Illgner, K -H , 1972. Werkstoffauswahl im Hinblick auf wirtschaftliche Fertigungen, Z.-VDI 114, 837-841, 992 995 6 92 Jaeger,T. A , 1974. Zur SicherheitsproblematiktechnologischerEntwicklungen, QZ 19.r-9. M t r c l t i n c t r b u r ,M r t n i c l t : I l l r r t s c r ' zttt 6 , 7 ( ) I l r r i l r i r c h ,l i , . n . t l . M o t l i l i z i c r t c l r n c | | r c s c l r | t d c n r f i k k r r r r r u l i l t i o r t s - l l y p < t l h c . s c Scltritligtttrp l ( t l l s c h r c i t c t t t l c r n l l l l ) : r r r e r l e s t r ; t k c r t | | r h l n l l r r l c s Itcriickrictrtigrrrrg 'l'cr'hrr M i t t c i l r r t t g t ' rItM 5 l l l o I i r l r r r r i r l l r r r r rtl , , l l l . , t r i . l , r . l c r t r g l . t : rll) n r m r l r u h : 371 6.93 Jakob, E,. and H. Scholz, 1962.Beleuchungim Betrieb. RKW-Series, Arbeitsphysiologie-ArbeitspsychologieBerlin, Cologne: Beuth-Vertrieb. 6.94 Johnson, Kenneth, Lester: Grundlagen der Netzplantechnik YDl-Taschenbircher T.53. Diisseldorf: VDI-Verlag. 6.95 Jung, A., 1959 SchmiedetechnischeUberlegungen fur die Konstruktion von Gesenkschmiedestiickenaus Stahl. Konstruktion 11. 90-98. 6.96 Kaminsky, G. and H. E Pilz, 1963. Gestaltungvon Arbeitsplatzund Arbeitsmittel, RKW Series, Arbeitsphysiologie-Arbeitspsychologie Berlin, Cologne: BeuthVertrieb. 6 . 9 7 K e i l , E , E O . M i i l l e r a n d P . B e t t z i e h e , 1 9 7 1 .Z e i t a b h d n g i g k e idt e r F e s t i g k c i t s und Verformbarkeitswerte von Stahlen im Temperaturbereichunter 4(X)'C,,4rr'lr. Eisenthuttenwesen 43. 757-1 62. 6 98 Kesselring, F., 1954. TechnischeKompositionslehre Berlin, Gdttingcn, Ilcitlclberg: Springer. 6 . 9 9 K i e n z l e , O . , 1 9 6 1 N o r m e n u n d K o n s t r u i e r e n ,K o n s t r u k t i o n1 9 . l 2 l - 1 2 5 6.100 Kienzle, O., 1943. Normung and Wissenschaft,Schweiz. Techn Z ,533-539 6 . 1 0 1 K l e i n , M , 1 9 7 0 .E i n f u h r u n g i n d i e D I N - N o r m e n , 6 t h e d n . , S t u t t g a r t :T e u b n e r 6.102 Klein, H. C , 1959. Hochwertige Schraubenverbindungen:Gestaltungsprinzipien und Neuentwicklungen, Konstruktion ll, 259-264 6 103 Ktocker, J.,1,967. ZeitgemasseForm Mrnich: S0ddeutscherVerlag. 6. 104 Kloss, G., 1964. Einige ribergeordnete Kor.rstruktionshinweisezur Erzeilung echter Kostensenkung, VDl-Fortschrittsberichte,Ser. 1, No. l. Drisseldorf: VDIVcrlag. (r.105 Kkros, K II.. 1973. Werkstoffoberfldchenund Verschleissverhaltenin Fertigung r r r r t lk o r r s l r r r c t i v cA r t w c n d : : r ' r gV, l ) I - I l t r i t h t c 1 9 4 D t i s s e l d o r f :V D I - V e r l a g . r/chrc. ( r .1 0 6 K l r r t t c r ,K , , l 9 5 l r r n t l l ( ) 6 0 ,' l l c l t r t i r c l r r . \ < / r n ' i r t 1 3 t o t g , V o l s . I . Q2 . 2 n d e d n . l l e r l i r r ,( i ( ) t t i n t l ( ' r It .l e i t l c l l r c r gS: p r i r t g c r References 378 von Kunstoffen,Z.-vDI 1'1L, 6.107Knappe, w., 1969. ThermischeEigenschaften 746-752. RKW-SeriesArbeitsphysiologie6.108Koch, H., 1963. Ltiftung des Arbeitsraumes, Berlin, Cologne:Beuth-Vertrieb' Arbeitspsychologie. Parts 1 & 2. Stuttgart: Maschinenteile 6.109Kohler,^G. and H. Rognitz, 1,961.. Teubner. zwischenKonstruktionund rationallerFerti6110 Koenig,W., 1953.Wechselwirkung gwg, Z.'VD195, 89G903. Frictionand Wear,London: Butterworth' 6.111kragelskii,I. V., 1'965. Kurbelnund Pedalenwissenmuss 6.112Kro"emer.K. H., \961. Wasman von Schaltern, Berlin, Cologne:Beuth-Vertrieb. Losungenin der ma6.113Kiihnpast, {., 1968. Das Systemder selbsthelfenden Konstruktion.Diss.,TH Darmastadt' schinenbaulichen 6.114Kurtze, G., H. Schmidt and w. westphal, 1.915.Physik und Technik der Heidelberg:SPringer. -Allgemeine Nr. 2: Maschinenkonstrukriont Gestaltungslehre, 6.118Leyer, A,.",1664. 2) Basel, Stuttgart:Birkhauser' lehre (technica-series technica12. 6.119Leyer, A., 1963.Grenzenund Wandlungim Produktionsprozess, 191-208 technicu l\4aschinenba]us, 6.120 Leyer,A., 1913-4.Kraft- und BewegungselgTellg.dgl 26'(1973) 2498-2510,2501-2520;technica5 (1914) 319-324;technica6 (1974\ 435 440 Nos. 1-6 (technica-series) Maschinenkonstruktionsle/rre 6.121 Leyer, A., 1963-1971. Birkhduser' Basel,Stuttgart: dcs Darstellungsart 6.122 Lipertz, tt., lgl+. Neue zeichnerische _zurRationalisierung Diss . Vorgehensweisen. methodischen bei vornehmlich Konstruktionsprozesses TH Darmstadt. 6 123 Maduschka, L., 1936. Beanspruchung von Schraubenverbindungenund zweck mdssige Gestaltung der Gewindetrager. Forsch. Ing Wes' 7,299-305' K., 1969. Schwingungen.Stuttgart: Teubner. Magnus, 6.124 6 t25 Magyar, J., n.d., From unpublished lectures,TU Budapest' 6.126Mahle-Kolbenkunde, 2nd edn. , 1964, 18 and 45. 6 . 1 2 7Martin, P. and M Mathey, L975. Zwerlassigkeitsbetrachtungenbeim Ttrr binenschutz. VGB Kraftwerkstechnik 55, 574-580' 655-660' 6 . 1 2 8Matousek, R., 1957 (repr. 1974) Konstruktionslehre des allgemeinen Mttt chinenbaus.Berlin, Gottingen, Heidelberg: Springer. 6 . t 2 9 M a t t i n e . A . a n d K . U l m t r , 1 9 6 3 . S p a n n u n g s v e r t e i l u n ign M e t a l l k l e b v e r b i r r dungen,Z.-VDI 105, 1449 1457. 6 . 1 3 0 M a J z . W . a n d H . K i e s . 1 9 7 3 .F u n k e n e r o s i v eus n d e l e c t r o c h e m i s c h eSse n k e n ,Z r t / 68.418-422. Tcnrptrtttttr inlrtlga.sluliontirer 6.131 Mclirn. E arrcltl. Parkus. 1953.Wiirrtrcsl)dtuuut!:(tt S p r i r t g c r . V i c r r n l t : -'l'irProggc. /r'lrir'r. l ( ) 7 0 , l ) c r r k c r ri t t V c r t o r t t t t r r t g ccrlll c i c h l c t t t l ; r t 6 1 . 1 2M c r r g c s ,( i i r n t l I { . ( ) 1 76 2 9 ' I ) i r r r t ' . s i < t r r i c rvc(l)l l l K t r r t s t s t . l l t c i l c 1/ .1 ," 1 ' l ' l I l l , 1 { I l ' l ( ) , vott Wcllctt-N:rlrt'tr ( ) r l r c A t t s l c p t r r t g l I r M , , 1 9 7 5 ,l { t , t ' l r t . r r r r r o r c l l O . t 1 . lM i l i t z c r . l ' r t s s t t ' t l c t v t ' t l r i t t t l t t t t l . l )t i' tr tr , I I ! l l e t l r l t r r I t . t l \ , f r r l r l t . rI . l ( ) ( r 5 ,i ) t r I r t r l l t t t t , ! r t l t t g r n t e n r tu u l l t r . , l r l r r i t t t i r ' l t , ' r h r ' l t . ' I ttl't, l r r l l t ' t l t t t :\ ' r ' t l ; r l I t t l r t t t t t ' Refercnces 6.135 Miiller, E. A., 1962. Klima im Arbeitsraum, RKW-Series Arbeitsphrsrul,rgrc. Arbeitspsychologie. Berlin, Cologne: Beuth-Vertrieb. 6.136 Mitler, K., 1970. Schrauben aus thermoplastischenKunststoffen, Wcrkstttlltltttt 514 & 515.Munich: Hanser. 6.137 Muller, K., 1966. Schrauben aus thermoplastischenKunstoffen. Kunst.\t().ll( 1t\. 241-250.422429. 6 1 3 8 M i i l l n e r , E . , 1 9 6 0 . E n t w i c k l u n g s t e n d e n z einm B a u v o n T u r b o g e n e r a t o r e l ./ l / l ( N achrichten ( 1960), 279-286. 6.139 Munz, D., K. Schwalbe and P. Mavr, lgTL Dauerschwingverhaltenmetalli.sclrt,r Werkstoffe. Braunschweig: Vieweg. 6.140 Murrell,K.,1971. GundlagenundPraxisderGestaltungoptimalerArbeitsprozt,.ssc. Diisseldorf. Vienna: E,con. 6.141 Neuber, H., 1958. Kerbspannungslehre. Berlin, Gottingen, Heiclclberg: Springer 6 . 1 4 2 N e u b e r , H . , 1 9 6 8 . U b e r d i e B e r i i c k s i c h t i g u n gd e r S p a n n u n g s k o n z e n t r a t i obnc i Festigkeitsberechnungen, Konstruktion 20: 245_251. 6.l43 Neumann, J. and K.-P. Timpe, 1970.Arbeitsgestaltung.Psychologi.st'hc f,rpltlt'rttt' b_eiUberwachungs- und Steuerungstiitigkeiten.Berlin: VEg Oeutriher Ve rllg tlcr. Wissenschaften. 6 . 1 4 4 N i e m a n n , G . , 1 9 6 3 . M a s c h i n e n e l e m e n t ev,o l . l , | & 2 t s e r l i n . G ( i t t i n g c r r . Heidelberg: Springer. 6 145 Niemann, G., 1975 Maschinenelemente.vol 1 Berlin, Fleidelberg. New york: Springer. 6'146 n.n. (anon.), 1969 ErgebnissedeutscherZeitstandversuche lanper Dauer. Dusseldorf: Stahleisen. 6.747 n.n.. (31on.;, 1958 Nickelhaltige Werkstoffe mit besonderer Wdrmeausehnung, Nickel- Berichre D 16. 79-83. 6.14u oehler, G. and A. weber, 1972. Steife Blech- unt) Kunststofl'konstruktionen, Konstruktionsbricher Vol. 30. Berlin, Heidelberg, New york: Springer. 6. 149 Pahl, G., 1973. Ausdehnungsgerecht.Konsrruktion 25, 367-373. 6.1-50Pahi, G.,1972. Bewdhrung und EntwicklungsstandgrosserGetriebe in Kraftwerken, Mitteilungen der VGB 52, Kraftwerkstechnik, 404_415. 6.1-51Pahl, G., 1967. Entwurfsingenieur und Konstruktionslehreunterstiitzentlie moderne Konstruktionsarbeit, Konstruktion 19, 337-344. 6.152 Pahl, G., 1973. Grundregel frir die Gestaltung von Machinen und Apparaten, Konstruktion 25. 271-277. 6 1 5 3 P a h l , G . , 1 9 7 5 . I n t e n s i v e r eS i c h e r h e i t s b e t r a c h t u n g ednu r c h m e t h o c l i s c h c K s prrstruieren, Chemie-lngenieur-Technik47, 457-464. 6 . 1 5 4 P a h l . G , 1 9 6 3 . K o n s t r u k t i o n s t e c h n i ki m t h e r m i s c h e nM a s c h i n c r r b l u .K o r r . r / r r r A tion 15.97-98. 6 . 1 5 5 P a h l , G . , 1 9 7 3 .P r i n z i p d e r A u f g a b e n t e i l u n g ,K o n . y t r u k t i o t2t5 , l 9 l _ 1 9 6 . 6 . 1 5 6 P a h l , G . , 1 9 7 3 .P r i n z i p i e nd e r K r a f t l e i t u n g , K o n . s t r u k r i o n2 5 , I - 5l _ 1 5 6 . 6. 157 Pahl, G. , 1973. Das Prinzip der Selbsthilfe, Konstruktion 25, 231 23i . 6 . 1 5 8 P a h l , G . , 1 9 7 1 . S i c h e r h e i t s t e c h n i ka u s K o n s t r u k t i v e r S i c h t . K o n s r r u k r i o n 2 3 . 207_208. 6.159 Pahl, G. and E. Schmidt, 1973 wie sieht die wissenschaft die Zukunft der Sicherheitstechnik? Sich erh e itsin g enieur g, 404-419. (r.l(r0 Paland, E. G., 1960. Untersuchungen iiber die Sicherungseigenschaften von Schraubenverbindungenbei dynamischer Belastung. Diss., TH Aanover. 6 l6l I'lltigcr, A.. 1964. Stabilittitsprobleme der Elastostatik.Berlin: Gcittinsen. HeidelItcrg:SIrrirrg,cr. 6, 1(rl ltrrlrirr.wicz. I: ,l()65 l;ricriortrttrd wcor o.f Materiul.t.New york: wiley & Sons. Int'. ( r l t r l l { l r r n r . l r l r o l t ,l rl , l l . l ( ) 7 { } S , i r . l r t . r l r c i t s g c r ( . ( (. li lt l.(s, t\ u l l e rvr o r r M i r s c h i r r cunn l e r References 3g0 besonderer Beriicksichtigung des "Gesetzes i.rber technische Arbeitsmittel", Z.:VDI 1.12.55-5'7& 109 114. 6.164 REFA,19'75. Methodenlehredes Arbeitsstudiums,4th edn. Munich: Hanser. 6.165 Reinhardt, K.-G., 1969.Verbindungskombinationenund Stand ihrer Anwendung, Schweisstechnik 79. Heft 4. 6.166 Reuter, H., 1958. Die Flanschverbindungim Dampfturbinenbau, BBC-Nachrichten 40.335-365. 6.167 Reuter, H., 1958. Stabile und labile Vorgiinge in Dampfturbtnen, BBC'Nachrichten 40.391-398. 6.168 Rixmann, W. , 7962. Etn neuer Ford-Taunus 12M, ATZ 64,30G311. 6.169 Rodenacker, W. G., \976. MethodischesKonstruieren. Berlin, Heidelberg, New York: Springer, 2nd edn. 6.170 Rognitz, H. and G. Kohler, 7959, FertigungsgerechtesGestaltenim Maschinen- untl Gertitebau. Stuttgart: Teubner. 6.171 Rohmert, W., 1966. Maximalkrbfte von Mdnnern im Bewegungsraumder Arme und Beine Report 1616 by N. Rhenish-Westphalian Government. Cologne. Opladen: WestdeutscherVerlag. 6.172 Rohmert, W , 1973. Psycho-physischeBelastung und Beanspruchung von Fluglorsen REFA-schriftenreihe Arbeitswissenschaftund Praxis 30. Berlin, Cologne: Beuth-Vertrieb 6.173 Rohmert, W, and T. Hettinger, 1963. Korperkrafte im Bewegungsraun, RKWSeries Arbeitsphysiologie-Arbeitspsychologie.Berlin, Cologne: Beuth-Vertrieb. 6.174 Rohmert, W. and P. Jenik, L912. Maximalkrdfte von Frauen im Bewegungsrautrt der Arme und Beine, REFA-Schriftenreihe Arbeitswissenschaft und Praxis 22 Berlin, Cologne: Beuth-Vertrieb 6.175 Rohmert, W., W. Laurig and P. Jenik, 1974 Ergonomie und Arbeitsgestaltung, Dargestellt am Beispiel des Bahnpostbegleitdienstes.REFA-Schriftenreihe Arbeitswissenschaftund Praxis 31. Berlin, Cologne: Beuth-Vertrieb. Reibsystc6.176 Roth, K., 1960 Die Kennlinie von einfachenund zusammengesetzten men. Feinwerktechnik64. B5-142. 6.177 Rubo, F.., 1966. Der chemische Angriff auf Werkstoffe aus der Sichl d c s Konstrukteurs. Der Maschinenschaden, 65-'74 6178 Rubo, E., 1972. Die Wirkung der Erosion bei der Stromungskorrosion. ( - Chemie-Technik l. l7-7-179. 6.179 Rubo, E., 1960. Hohere Sicherheit chemisch beanspruchter Bauteile dtrrclr konstruktive Korrosionsbewertung am Beispiel von Druckapparaten, KonstrttA tion 12. 490-498. 6.180 Salm. M and W. Endres, 1958 Anfahren und Lastdnderungvon Dampfturbitrcrt, nic 76.326-330. 6 . 1 u 4 S c h m a l e . H . . 1 9 6 5 .D a s S e h e n b e i d e r A r b e i t , R K W - S e r i e s . B e r l i n , C o l o g r t t ' Bcuth-Vertireb 6 1 f t 5 S c h m i c l ,E , . 1 9 6 9 T h e o r e t i s c h cu n r l c x p c r i r n c n t c l l cU n t c r s u c h u n gc l e sM c c l t r r t t t ' r r r u s c l c r I ) r c l t n r o r l c n t i i b c r t r i r g t r n gv o t t K c g c l - [ ' r c s s - V c r b i n i l u l r g c l rV I ) l ' l ' " r r r c l t r i t l . r b r ' r i t lSr tccr i c sl . N o l ( r l ) i i r s c l l o r l :V l ) l - V c r l r r g ( r ,l f i 6 S c l t r r r i r l t .K [ ) . i r n r l I ) , . l S r ' l r l o r l c"r 1 9 7 5 , K o t t s t t t t k t i v c M i i g l i c l l k c i t c r r r t r r M i r r r l c r . r r r rt lger ' ( i c l i i r r s r . h ( . i l t \ltlei l n l l . I ' l l l . l h ' r u ' h t t l l t t V l ) l - V c r l r r g . t r , p i 7 S c l r r r r i t l t ,h .l il , , l , t 7 t I . . t r t y t t , , t t tI. ' t ) r u n i l l t y r n n t t n v l t l i l l t r r , ' l r l r r i tt t t r r ll r r t trrrrr irlrtttit'ltl llrttrt't References 6.188 Schmidtke, H , 19i3. Ersonomie 2l r . r r g .M u n i c h : H a n s e r . 6 . 1 8 9 S c h m i d r k e .H . . l 9 6 t t . I J h e r w a c h u n p s - . S e r i e s .B e r l i n , C o l o g n e : B e u t h - V e r t r i t 6 . 1 9 0 S c h n e i d e rP. . , l g 6 T . V e r b i l l i e r e Konsrr men. DIN-Mirrelilunpen 4b. I4l- l4-s. 6 . 1 9 1 S c h i l t z ,W . a n d H . Z e n n e r , 1 9 7 3 .S c h a d e n s a k k u m u l a t i o n s h y p o t h e s e n zlr l(.lr(.r\ d_auervorhersage bei schwingenderBeanspruchung-Ein tiiiiscner U6cr 5lir L, / Werkstofftechnik 4, 25-33 & 97-102. 6'192 von BauelementendesDurtrltll,r,ttrl lchwaigerer, S', 1.970.Festigkeitsberechnung Behtilter-und Rohrleitungsbaus,2nd edn. Eerlin, Heidelberg, New yori: S'ri' ger. 6 ' 1 9 3 S c h w e n k ,W . , 1 9 6 9 S t a n dd e r K e n n t n i s s er i b e r d i e K o r r o s i o nv o n S t a h l , S r a l tt!u r t l Eisen 89. 535-547. 6 . 1 9 4 S h r e i r , L . L . , 1 9 6 3 & 1 9 6 5 .C o * o s i o n . L o n d o n : G e o r g e N e w n e s . 6.195 Skiba, R , 1975. TaschenbuchArbeits.sicherheit,2ndei'n. Bielefeld: Schnridr 6 . 1 9 6 S p i i h n , H . a n d K . F a s s l e r ,1 9 6 6 .K o n t a k t k o r r o s i o n :G r u n d l a g e n - A u s w r r k u l * Verhiitung, Werkstofle und Korrosion 17, 321_331 6 197 Spiihn H. and K. Fdssler, 1972. Zur konstruktiven Gestaltung kurrosionsbcrlrlqluclter Apparate in der chemischen Industrie, Konstrttktiin 24.2,19-l.5li & Jl tL--tt). 6 1 9 8 S p e h n 'H . , E . R u b o a n d G . P a h l , - l 9 7 3 .K o r r o s i o n s g e r e c h tGee s t a l t u n g . Korr,rrrrrlition 25. 455-459. 6 1 9 9 S t e i n a c k ,K . a n d F v e e n h o f f , 1 9 6 0 . D i e E , n t w i c k l u n gd e r H o c h t e m p e r a r u r t u r binen der AF.G, AEG Mitt. 50.433-453. 6 . 2 0 0 S t o f e r l e , T . , H . - J . D i l l i n g a n d r . R a u s c h e n b a c h ,r 9 7 5 . R a t i o n e i l e MontageH e r a u s f o r d e r u n ga n d e n I n g e n i e u r ,Z . _ V D I | l j , : ' 1 5 7 1 g . 6 . 2 0 1 S z a b o ,r . , 1 9 7 2 .H o h e r e T e c h n i s c h M e echanikl & II, 5th edn.,Berrin,Heidelberg, t Stahl und G ussei,sen_W erks toffv erh alten. ryen. Leipzig: VEB_Fachbuchverla,r. Maschinenelemente, repr.. Berlin, Heitlcl_ ris der Netzplantechnik.Munich: Motlcrrrc 6 2 0 5 T h u m , A ' , 1 9 4 4 . D i e E n t w i c k l u n gv o n d e r L e h r e d e r G e s t a l r f c s t i g k c i r , Z.-Vt)l uu. 609-615. 6206 Tietz, H., 1953. lil-lgqttttt".peratur-Kraftwerk mit eiucr frrischdurnp1tcnrpcratur von 610"C, Z -VDI 96.802_909 6 207 Tochtermann' W and F. Bodenstein, 196u & 1969. Kotrstrttktitttt.y<,!t,trtt,tttt, tlt,r , ^ . M a s c h i n e n b a u e sP, t s . 1 & 2 , u t h e d n . B e r l i n , H e i d e l b c r g .N c w y . r k : s p r i n g c r , 6 . 2 0 8 T o d t , F . , 1 9 5 8 . M e t a l l k o r r o s i o n , 2 n de d n . B e r l i n : d e G r ' u y t c r . 6'209 Trapp, H.-J.' 1975..Beitrag zum rechnerischen Betriebsfestigkeitsnachweis fijr Bauteile in Kranhubwerken, Konstruktion 27. 1i2_14g. 6.210 Tschochner, H., 1954. Konstruierenund Gestarten.E,ssen:Girarder. 6.211 ruffentsammer, K., 1975. Ldrmarm Konstruieren-Ein Beitrag zur Humani_ ^. - :i".rung des Arbeitslebens, vDI-Berichte 239. Diisseldorf: VDl-v6rlag 9 ?l? Vlhig, H H . 7970 Korrosion und Korrosionsschutz. Berlin: Akacleirie-Verlag. 6 213 Vcit, H.-J and H. Scheermann,7972. Schweissgerechtes Konstrureren. Schweis.stt'<hni k 32. Diisseldorf: Deutscher verlag fi.ir Sihweisstechnik. (r,211 vDl/ADB.Ausschuss Schmieden, 1g75: Schmiedestilcke-Gestaltung,Anwen_ t l t u r 1 4I l, c i r l t i t ' l r ' .l l i r g c r r :I n i o r m a t i o n s s t c l l e Schmiedestiick-Verwendun ! Indusim t t i c v er b i r n r lI ) c r r t s c h cS l chnricrlcrr, ( r l l 5 \ ' l ) l l t r . r i c l r t cl l ( ) . l ( X r f i ,K o l t l t t o l t l r t r t t , . l ) t i r s c k l o r l 'V: I ) l _ V c r l r r g , R e f er e n c e s Lecturesto VDI6.216VDl-Berichte 214, 1911. Werkstoffeuntl BauteiLfestigkeit, VDI-Verlag Conference,Dusseldorf.Diisseldorf: Dussel6 2lj VDI-Berichte 23g,1g'75Beispieleftir li)rmarmeMaschinenkonstruktionen. dorf: VDI-Verlag Kon6.218 VDl-Richtlinine 2224, lg12 Formgebung technischerErzeugnissefiir den strukteur. Dusseldorf: VDI-Verlag Konstruieren Dusselclorf: 6.219 VDI-Richtlinie 2225. 1969. Technisch-wirstschaftliches VDI-Verlag. 6 220 VDI-Richtl]nie 2226, 1965 Emplehtung fl)r die Festigkeitsberechnungmetallischcr dt'r[: VDI-Verlag. Diisseldorf: VDI 6.224 VDl-Richtlinie 3239,1966 Sinnbikler ftir Zubringefttnktionen Verlag. Grundlag,en Dtts 6.225 VDI-F.ichtlinie 3720.1g'.'5.Li)rmarm Konstruieren-Allgemeine seldorf: VDI-Verlag. Diisseldorf: VDI 6.226 VDl-Richlinie 400tr, ;512, Sheet 2 Uberlebenskenngrbssen. Siemens-Verlag. von Thet'rrto 6.231 Vorath, B.-J.. 1972 B"ittug zur Elmittlung der Ermiidungsfestigkeit plasten. ZwF 61. 412-418. Engirttt'r 6.232 Wachter, A., 1960.Proper design avoidsequipmentcorrosion. Chemical l n g , F e b . 1 9 6 0 ,1 6 2 - 1 6 6 . . u t E i n t e i l u n gv o n V e r s c h l e i s s v o r g i i n g c t t , ., B. Pfeiland G. Keil, 1975Z 6 . 2 3 3 W " a g n e rK Schmierungstechnik 6. 299 302 Vl)l 6.234 Wahl, W. 1975. Abrasive Verschleissschiidenund ihre Verminderung Diisselclorf: v I ) | Berichte 243: "Methoclik der Schaclensuntersuchung". Verlag. 6.235 Wank-e, K., 1963. Wassergeki.ihlteTurbogeneratolen. In AEG-DampJntrbirt''rr' Turbogeneratoren.159-168.Berlin: Verlag AEG --Mllirtz 6.236Warnecke. H. J., I{.-G. Lohr and w. Kiener, 1975 Montagetechnik KrausskoPf. mit tl.tctttto 6 . 2 3 7 W e b e r , A . , t q o + . w e r k s t o f f - u n d f e r t i g u n g s g e r e c h t eKso n s t r u i e r e n ' 16, 2-11 Konstruktion plastischenKunststoffen. D1N-MittL'ilturtl'tt 6 . 2 3 t 3W c b e r , H . , 1 9 6 8 ,B e d e u t u n g u n d A u f b a u d e r W e r k n o r m u n g ' 18.41-16. r t . fl i r t g i r t "' 6 , 2 3 9 W c l c h . B . , 1 9 6 1 .T h e r r . n aIln s t a b i l i t yi n I I i g h - S p c e dG e a r i n g ,l o u r n a l R e f er e n c e s 6.244 Wiegand, H. and K. H. Illgner, 1962.Berechnungund Gestaltungvon Schraubenverbindungen Berlin, Gottingen, Heidelberg: Springer. 6 2 4 , 5 W i e g a n d , H . , K . H . I l l g n e r a n d K . H B e e l i c h ,1 9 6 3 . E i n f l u s s d e r F e d e r k o n s t a n t e n und der Anzugsbedingungenauf die Vorspannung von Schraubenverbindungen, Konstruktion 20. 130 137. 6.246 Winkel, A. and E,. Walger, 1964.Staub am Arbeitsplalz RKW-Series Arbeitsphysiologie-ArbeitspsychologieBerlin, Cologne: Beuth-Vertrieb. 6.217 ZGV-Lehrtafeln, 1966. Erfahrungen, Untersuchungen, Erkenntnisse fiir das Konstruieren von Bauteilen au.sGusswerkstoJJen Dusseldorf: Giesserei-Verlag. 6.248 ZGV-Mitteilungen, n d. FertigungsgerechteGestaltung von Gusskonstttktionen. Diisseldorf: biesserei-Verlag 6 249 ZGV, n.d. Konstruieren und Giessen.Diisseldorf: Giesserei-Verlag 6.250 Zienkiewicz, O. G., 1975. Die Methode der finiten Elemente in der Ingenieurwissenschaft.Munich, Vienna: Hanser. 6.251 Zinkler. B , 1962 Gesichtspunktedtir das Gestalten von Gesenkschmiedeteilen. Konstruktion 14. 274-280. 7 Developing size ranges and modular products 7 1 AEG-Telefunken, n.d.. Hochspannung,s-Asynchron-Nornmoloren,Bauku\lctt\\'r' tem, 160 kW3150 ftll Druckschrift E41.01.0210370. 7 2 A c h e n b a c h , H . - P . , 1 9 1 5 . E i n B a u k a s t e n s y s t e mf i i r p n e u m a t i s c h eW c g e v c n t i l e tvt-Z ind. Fertigung 65, 13-I7. 7.3 Beitz, W and W. Keusch, 1973.Die Durchfuhrung von Gleitlager-Variantenkonstruktionen rnit Hitfe electronischerDatenverarbeitungsanlagenVDI-Berichte 196. Diisseldorf: VDI-Verlag. 7 . 4 B e i t z , W . a n d G . P a h l , 1 0 7 4 B a u k a s t e n k o n s t r u k t i o n c nK o n s t r u k t i o n2 6 , 1 5 3 1 6 0 7 . 5 B e r g . S . . 1 9 4 9 .A n g e w a n d t eN o r m z a h l B e r l i n , F r a n k f u r t : B e u t h - V e r t r i e b . 7 (r Berg. S., 1969. Die besonclereEignung der Normzahlen fiir die Grossenstufung. D I N -M ineilungen 48, 222 226 7 7 Berg, S., 1965. Konstruieren in Grrjssenreihenmit Norrnzahlen. Konstruktion 17. 1 - 52 1 . 7.8 Berg, S., 1959 Die NZ, das allgemeineOrdnungsnrillel North-Rhine Westphalia G o v e r n m e n t P u b l i c a t i o nN o . 4 . 7.9 Berg, S , 1958: Theorie der NZ und ihre praktischeAnwendung bei der PlutLurgtt Gestaltungsowie in der Fertignttg North-Rhine Westphalia Gttvernment Publicirtion No. 3-5. me r T e r l t n i k B c l l i n . ( i ( r t t i r t g c r t . 710 Borowski, K.-H., 1961 Das Baukastensy.tted I l e i d e l b e r g :S p r i n g e r 7 . 1 1 B r a n k a m p , K a n d J . H e r r m a n n , 1 9 6 9 . B a u k a s t e n s y s t c r l a t i k - ( i r u n r l l i t g c tut t t t l A n w e n d u n g i n T e c h n i k u O r g a n i s a t i o n .I n d - A n z e i g e r9 l , n o s . 3I & . 5 0 7.12 DIN 323,1914. Sheet 2. Normzahlen u Normzahlreihen. Berlin. ('olognc: BcuthVertrieb 7 l3 Evershein.r.W. and H -P Wiendaht, 1971 Rationelle Aufrragsult*'icklung irtt Konstuklionsburo. Essen: Girardet 7.14 Flender. 1,972.FirmenprospektNo. K. 21731D Bocholt. 7 l5 Franzmann. K , 1975. Interner Entwicklungsberichtder Fa. Borsig. Berlin. 7 l6 F'ricricwald. H -J, 1912. Normzahlen Grundlage eines h)irtschaftlichenErzeugnisl)t ()grunun.r tscrlin. Cologne: Beuth-Vertrieb. 7 . 1 7 l : r i c r l c w i r l d .. l - . 1 1 9 7 0 . N o r r n u n g i n t e g r i e r e n d e r B e s t a n d t e i l e i n e r F i r m e n k o n z c l r l i o n ,l ) l N - M i t t c i l t u r , r . l.c1r9r .I I I 7 , l S ( i er l r l r r t l 1. , . l t ) 6 t ) A , l r r r l i c h k c i t s g e s cbt zcci n rI : n t w r r r cl l c k t r o r r c c h a n i s c h(ei er r i i t e . ,/ t l,l il 1, l0ll l0l(), 384 References 65. in derHydraulik,wt.-Z-ind' Fertigung 7.19Glaser,F.-J.,1975.Baukastensysteme 19-20. 7.20 Gregorig, R., 1967.Zur Thermodynamik der existenzfAhigenDampfblase an einem aktiven Verdampfun g skeim, V erfa hr enstech nik (1967), 389. 7 21 Hansen TransmissionsInternational, 1976. ProspectusNo. 6102-62/D Antwerp 7.22 Hansen TransmissionsInternational,1976. ProspectusNo. 202lD. Antwerp. 7 23 Keusch, W.,1972. Entwicklung einer Gleitlagerreihe im Baukastenprinzip.Diss.. TU Berlin. 7 . 2 4 K i e n z l e , O , 1 9 3 9 .D i e N Z u n d i h r e A n w e n d u n g , Z . - V D [ 8 3 , 1 1 7 . 7 25 Kienzle, O., 1950 Normungszahlen. Berlin, Gottingen, Heidelberg: Springer. 7 2 6 K i e s o w , H . , H . M i h m a n d R . R o s e n b u s c h ,1 9 7 0 .A u t o m a t i s i e r u n gv o n E n t w u r l . Konstruktion und Auftragsbearbeitungim Anlagenbau, dargestelltam Beispiel des Warmetauscherbaus. IBM'Nachrichten 20, 147-153. 7 27 Lang, K. and G. Voigtldnder,1971 Neue Reihe von DrehstrommaschinengIOSScl Leistung in Bauform B3, Siemens-Zeitschrift 45, 33-37 . 7 28 Lehmann, T., 1959. Die Grundlagen der Ahnlichkeitsmechanik und Beispiclc ftir ihre Anwendung beim E,ntwerfen von Werkzeugmaschinen der mechanischcrr Umformtechnlk. Konstruktion II, 465-473. 7 .29 Maier, K., 1975.Konstruktionsbaukastenirr der Industrie, wt.-2. ind Fertigung65 21-24. in der Verfahrenstechrtil, 7.30 Matz, W., 19,54 Die Anwendung des Ahnlichkeitsgesetzes B e r l i n , G o t t i n g e n ; H e i d e l b e r g :S p r i n g e r . 7 3 l P a h l . G a n d W B e i t z , 1 9 7 4 . B a u r e i h e n e n t w i c k l u n gK, o n s t r u k t i o n2 6 , 7 1 7 9 l ' . 113-118. Fltt 7.32 Pawlowsk:i,J., IgTL Die Ahnlichkeitstheoriein der physikalisch-technischen schung. Berlin, Heidelberg, New York: Springer. 7 . 3 3 R e u t h e , W . , 1 9 5 8 . G r o s s e n s t u f u n gu n d A h n l i c h k e i t s m e c h a n i kb e i M a s c h i n c r r elementen, Bearbeitungseinnheitenu. Werkzeugmaschinen.Konstruktion 19, 465 476. 7.34 Siemens, 1971 Drehstrommotoren fiir Hochspannung, Druckschrift M 2 and s u p p l e m e n (t l q 7 5 ) . 7.35 Schwarz,W, 1975. Universal-Werkzeugfrhs-und -bohrmaschinennach Grundprizi pien des Baukastensystems,wt.-2. ind. Fertingung 65,9-12 7.36 Weber, M., 1930. Das allgerneineAhnlichkeitsprinzip der Physik und sein Zttsrtttt menhang mit der Dirnensionslehreund der Modellwissenschaft.Jahrb. der Sclrillt bautechn. Ges 31. 27+354 7.37 WestdeutscheGetriebewerke, 1975. Prospectus Bochum English Bibliography English texts referred to by Pahl and Beitz are included in this list. Their original reference numbers are given in square brackets. Adams, J. L.,1974 Conceptual Blockbusting, Freeman, San Francisco. Alger, J. R. M. and Hays, C. V., 1964 Creative Synthesis in Design, Prentice-Hall, Englewood Cliffs, N. J. Andreasen, M. M., 1983 Kahler, S. and Lund, T.,1983 Designfor Assembly, Springer, Berlin Andreasen, M. M. and Hein, L., 1987Integrated Product Development,IFS Publications/ Springer-Verlag, Bedford/Berlin, Heidelberg, New York, Tokyo. Archer, L.8., 197I Technological Innovation - A Methodology, Inforlink, Frimley, Surrey. Asimow, M.,1962 Introduction to Design, Prentice-Hall, Englewood Cliffs, N. J. Bailey, R. L., 1978 Disciplined Creativityfor Engineers, Ann Arbor Science,Ann Arbor, Mich. Besant, C. B., 1980 Computer Aided Design and Manufacture, Horwood, Chichester, Sussex. Dalkey, N. D. and Helmer, O., 3 April 1963An Experimental Application of the Delphi Method to the use of Experts, ManagementScience,9,3. [5.9] de Bono, E.,l97I The Use of Lateral Thinking, Penguin, Harmondsworth. -,1976 Teaching Thinking, Temple Smith, London. Dieter, G., 1983 Engineering Design - A Materials and Processing Approach, McGraw-Hill, Tokyo. Burr, A. H., 1981 Mechanical Analysis and Design, Elsevier, Amsterdam, Oxford & New York. Deutschman, A. D., Michels, W. J. and Wilsom, C. E.,1975 Machine Design, Macmillan, London. Dixon, J. R., 1966 Design Engineering: Inventiveness,Analysis and Decision Making, McGraw-Hill, New York. [1.11] Eder, W. E. and Gosling, W., 1965Mechanical SystemDesign, Pergamon, Oxford. Faires, V. M., 7972Design of Machine Elements,Macmillan, London. Flursheim, C. H.,1977 Engineering Design Interfaces,Design Council, London. French, M. J., 1985 Conceptual Design for Engineers, Design Council/Springer-Verlag, London/Berlin, Heidelberg, New York, Tokyo. 7988 Invention and Evolution: Design in Nature and Engineering, C.U.P., Cambridge. Furman, T T. , 1981 Approximate Methods in Engineering Design, Academic Press,New York. (ilcgg. G. L., 1969The Design of Design,C.U.P., Cambridge -, 1912'l'hcSelection of Design, C.U.P., Cambridge -. 1973'l'haScicn<'e of Design, C.U.P., Cambridge. - , l 9 t i l ' l ' l r t I ) c v c l o l t n u ' not . fD e . s i g nC, . U . P . , C a m b r i d g e ( i o r r l o n , W . l . 1 . , . \ r ' r r r ' r i cI sI :.r r p c r & I \ r w . N c w Y o r k . [5.llt] 386 English BibliograPhY d' Butterworth, London' h' London. ttton focused on System Management' Paper tndrstry Safety Conference, Goddard Space English Bibliography Wachter, A., February 1960 Proper Design Avoids Equipment Corrosion, Chem. Engng. 162,166. [6.2321 Wallace, P . J . , 1952 The Techniquesof Design, Pitman , London. Welch, B., 1961 Thermal Instability in High-Speed Gearing, J. Engng. for Power,9lff. [6.23e] 'tPirtut 6102- 621D, AntwerP' -,1976 Prospectus202D, Antwerp [7'21] A Philosophy of Design' Brooks/Cole' Harrisberger, L., 1966'irgin,"'iontiip, Belmont, Ca LondonHubka, V.; 1982 Principles of Engineering Design, Butter-worth' Elements' Wilev' New York' Mechanical ol design optimim tsso n. c., l"h;;;;, 'Mechai.nical -, 1SZS Disign Synthesil, Krieger, Huntington' Nly Ygt-k . Seedsof Hu^o, Futures,wiley, New York. Design"Meihocls: C.,1970 Jones, J. London [6 111] iiiu!"f .tii, i. v ., tsos"Friction and W ear, Butterworth, Engineering Design, Wiley, New and to Engineering Introduction An 1969 V., Krick, E. Y o r k . [ 3 . 2 ,5 . 2 9 ] Leech.D.J.and.I.urner,B.T.,lg85EngineeringDesignforProfit,EllisHorwood. Chichester. Leyer, A., l9'74Machine Design, Blackie' I ondon' Eng. Des., London. fr4i.pfti,'O. L., 1960 The Decisions of Engin.eerinbgDesign,Inst. Design, Inst. Eng' Dcs ' Engineering of Evol"ution Controlled The 1963 E., tUatcneti. London. Matousek,R.,l963EngineeringDesign:ASystemaricApproach'Blackie'London' London' Muyuff , W. H., 1979Pinciptes in Design, Design Council' Boston' Mass' & Bacon' Allyn Design hngineering 1969 H., W. MidOenOott, New York' InventinS'Dekker' 'q'tout rcnoi -, fSSt What Every Enginei Sho'd York' New McGraw-Hill' Design, Engineering D.,1969 Morrison, Homewood, Ill [2. 17| c., tgol worti iysii? o"ri[", The ltlgak Concept, rr-win, N;i;;, -,9eZWork D e s i g n ,l r w i n , H o m e w o o d , I l l ' [ 2 ' 1 8 ] Ne* York'.[5'34] OsUo.n, A. F., 195iA iphed' Imaginarion,Scritnet'i, Methodology' Prentice-Hall' Development and Ptinning, b'rrr;n, lg77 B.', O;it"f.ky, Englewood Cliffs, N. J. New York' Parr, R]E., l97O Principtes of Mechanical Design, McGraw-Hill' 46,pp.344-349 [l 31' Postgraduate (., design, Engineering 1SZOer:iiciptrs of ir""ny, n. 1 5l pitts, Cl, 1973Techniquesin Engineering Design, Butterworth' London' pofui, p., 1976A Bac'kground t6 Engineering Design, Macmillan' London' Rabinowicz,E.,1965 iriirtiona,dtf'a'ofliateriik,Wiley'NewYork'-[6'162] neOto.O, C. O., 1975Mechanical Engineering Design' Macmillan'.London' and Production. Macmillirrt' fi;ir, C. and Koenigsb.ig.r, n, tilO Orstg" foistrength London. Sherwin'K.,lg82EngineeringDesignforPerformance,Horwood,Chichester' New York' S'nEf.V,l. C.,l9ll ltiechaniiat EngineeringDesign,McGraw-Hill' 194] London [6 inr".ir, L. L., 1965Corrosion,Newnes' Dgign. Pergamon,oxlortl iirnnn, H. A.. lg75 A Student,sInrroductbn o Engineering M a c h i n c _ E l c m a n t s , p r c n t i c c-Hall,EnglewoodCliffs.N..l s;.;ia;M.F., lg7gDesignof Pilman, London. Svcnsson,N. L., l976 Iitroiuctbn to l..ngint,Uingl)esign, ;i Ncwncs-lJcttcrworth'I-ondon' lndu.ttriull)<'.ti14rr, i.i"., i.'.,l.)79 Sh,,rtCourscin (iruphic Mttdtlltttli' -. Antlrcirscn. M. M. and Sclrnritlt. 1", 1j.., l9?9 littl4itrctring l . o t t < k l t t Ncwrtcs-lJtrttcrw()rt l' 'lfrrirrg. M . W . : r n t l l . : r i t h w i r i t c . l l ,{ . , l ( ) 7 ?l l t w t t t l n v e n t ,M a c n r i l l i l n . l . o t t d o t t ' , l y r t o r m t u ' A p p n n t h . k t , t l r t ' l ) t , t i 1 4t tt.tl l r r l t t t t t ' t l V I ) 1 , ' i 9 H 7 V l ) l l ) t . t i g : t tl l t t n t l / t o o Ai l l / (icinr.n ctliti.rr). Vctcitt l)errtsr'lrr't S ! , r r t , r , r t t t r t !t , r ' r u l i u l r ( t r i r r 1 r l i l t r ( t r t . l l t ) l l t l I I t t p c r t i e r r rV c c t l l r g 'l ) t t s s c k h r t 387 Withing, C.,1966 CreativeThinking, Reinhold, New York, 1966.[5.63] Woodson, T T., 1966Introduction to Engineering Design, McGraw-Hill, New York. lndex abrasion 254 abstractformulation 77 abstraction 58 and problem formulation 5943 procedure during 60 to identify problems t42,156 accelerationforces 118 adaptations 56 adaptivedesigns 2, 4, 67,78 adaptive functions 3M adhesivejoints I99 adjustments 294 alarmvalve 225,229 algorithmic-physical design method 13-15 algorithmicselectionprocedure 72-13 aluminium-siliconalloypiston 236 ammonia synthesis 209 analysis 32-3,38 of existingtechnicalsystems 83-5 of naturalsystems 83 ofspecification 59 AND-functions 69,72 AND-relationship 298 animal feed bag filling, storing and loading,63 assembly 20, 177,180,194,228,373 assemblydesignaspects 289-296 assemblyevaluation 29+7 assemblyguidelines 291 assemblyoperations 29I-5 assemblytypes 289-90 associatedforces 203 associationofideas 32 automatedassembly 293 auxiliaryflows 75,78 auxiliaryfunctions 24,74, 344 auxiliarymodules 3M Baatz,U. 126 B a c h .C . 6 backwardstcpsmcthod 34 balirncctrl'lilrccs 203-5.212 balancedvalue profile 132 balancingcomponents 2I3 balancing elements 204 basic functions 343 basicmodules 344,350 bearing bracket designvariants 317 bearing-lubricationsystem72 bearingoil pressure 225 Deanng force transmissionin 202 hydrostatic 223 hydrostaticaxial 2I8,2I9,228 bending stressesin 2I9 modular system 346,350 plain 348 rolling 210 taper roller 225 bending moments 208 bendingstiffness 2lI bendingstresses 197,211,279,220 bent parts 277 Berg, S. 3I9 black boxes 26 boilers with membranewall 207 bolted connections 216,247 boltedjoints 2ffi,245,288 Booleanalgebra 69J2,711 bought-outcomponents 284 brainstorming 87-9, 158,160 Brankamp, K. 45 Bredtschneider-Uhde self-sealins cover 2I4 brittle materials I92 BSI (British StandardsInstitution) 259 building-blockconstructionmethod 272 carton assemblymachine 55-6 cast components 272 catastropheanalysis32 Cauchycondition 318,331 Cauchynumber 318,331,332 causc-cffcct relationship 198 cirvitati()n 254 Index 390 CENC (Comit6 Europ6ende Normalisation) 259 CENELEC (Comit6EuroP6ende Normalisation Electotechnique) 259 characteristicnumber 322 checklist,summaryof 362 circuit-breaker 209 steam clampconnectionof superheated pipe 208 clamping 277 clarity l72J ,182,31.5 classificationschemes 68,94-101, 108, 109 classifyingcriteria 94, 95, 98, 104-5 closedmodular systems 345 clutchoverallvaluerating 403 clutchselectionprintout 402 COz-enrichedlye 255 cold extrusion 275 collaboration 334 combinationof principles 29,lO9 companyobjectives 46,47 companypotential 47 companyshortcomings 50 compatibilitymatrix 109,111 compatibilityverification 109 component 20 componentfastening185 componentformdesign 212-81 componentsafety 185 compositeconstructionmethod 2772 gasesstorage 255 compressed computer-aideddraughting 5 computers designapplications 5 potentialbenefitsof 110,112,118 conceptvariants 44, \16-8,348-50 comparing 126,138 determing the rating of 126-9 evaluating 118-19,151,163,350-1 firming up into 146,163 selecting 350-l design 2, 30, 40,57-165,362 conceptual to identifytheessential abstracting problems 5746 comhiningsttlutionprinciples l0U-12 functionstructurcs cstablishing 66-82 c v i t l u i t t i r t gc ( ) n c c l )vt a r i i t n ti t g i t i t t s t tccltnicnl itntl ccttntlntic critcrit I ltt .19 e v i r l r u r l i o trtl t t t i t t t l I l 2 t , c x i t t t t l t l c so l 1 , l t )( r 5 firming-up into concePt variants 11G18 logicalconsiderations 69Jz man-hoursspenton 366 searchingfor solutionprinciplesfor sub-functions 824 selectingsuitablecombinations nz-rc stepsof 57 useofterm 57 concretefoundations I97 connectors 293 constraints 58,59,64,70 fictitious 65 genuine 65 consumercharacteristics136 control equipment 20 conventionalaids 83_6 conveyors,modular systems 36I coolingfan 307 corrosion 92 accompanyingerosion,cavitationand abrasion 254 causesand effectsof 249 contact(bimetallic) 255 crevice 257 local 251-5 transitionzone 255 uniform 249-50 corrosiondamage 249-56 designingagainst 2554 corrosionfatigue 253 costsof components,materials,semifinishedmaterials,and standard and bought-outparts 287 cost reduction 59 costrequirements 194 cost structure 286 costingand costevaluation 284-9 Coulomb'slaw 26 couplings 21, 183,192,222,231 crane drives 202 creativeability 362 creep 24{0_3 abovecriticaltemPerature 243 at room temperature 242 below criticaltemperature 242 designfeatures 246-9 crccpcurvcs 242 c r c c ps t r c n g t h 2 4 1 , 2 4 3 ( ' r i t i c a l ) : r t hA n a l y s i s 3 3 criticrrltcnll)criltttrc l,l(I .l ( ' r i l i e i rwl l t i r l i r r g s p c c t l s3 ( X r ('u\lolttclr1x'cilielttttetiotts 1't'l Index cut-out components 279,280 cyberneticconcepts l7 data banks 83 data collection 36 data sheets 326,328,334,339 preferrednumber decimal-geometic series 3f922 decisionprocess 39 definitive layout 166 deformation 19V203,211., 237,247,248 Delphi method 89-90 designcataloguesI2-I3,101-8 designconcept 1 designcyclewith learningsystemand environment 18 designfactors 8 designfaults 297,306 designfunctions 13 designmethod 4,9-15 designprocess 3&44 flowdiagram 44 phasesinvolved 40 work flow during 4V designscope 14 designtask 45 and activities 1-3 seealso task designtypes 4 designerresponsibility 53 deignerole 2,4,49 detaildesign 3I,424 differentialconstructionmethod 267-9 dimensionalchanges 56 dimensionallayouts351 Institution) 259 DIN (GermanStandards DIN 44 300 37 DIN 44 301 37 DIN 69910 25 discursivebias 92-1.08 discursiveprocedures32 disturbingfactors 297-302 divisionof labour 334 documentation 284,351 dominancematrix 130 dough-shapingmachine74 drawingrequirements 3514 drop fbrging 275 durability 191,221 c l y n a m isci m i l a r i t y 3 1 6 ,3 1 8 , 3 3 6 c c o r l o r r r i c c h i r r a c t c r i s t1i c3s4 ct\lrorrriccritcrilr .ll{ 39r economicevaluation\37 economicfactors 348 economicfeasibility 30 economic rating 128,136, 13'7, 309, 374 elasticcouplings 222 elasticdeformation 792,197 elasticforces 318,330,332,350 elasticpipelinesupports 335 electricmotor end cover 270 electricmotor housing 283, 340 electricmotor magnetsupport 268 electricmotor sizerange 339 electricmotors 1\C332 modular systems 355J electricalinstallations 193 electricalmachines 307 operated electromagnetically clutches 149 eliminationprocedure lI2 embodimentdesign 2,30. 42, 166-374 basicrulesof 172-3 checklist 170_2 corrective steps 166 evaluation of 309-14 guidelinesfor 227-96 principlesof 194-227 stepsof 1.66-iI useofterm 166 energyconversion 21-3,78 energysources,harmfuleffectsassociated with 193 energystorage 99,795 engineeringsystems,fundamentals of 2W37 environmentaldemands18 environmentalsafety 182 epicyclicgearboxes212 equipment 20 ergonomics 177,179,193,228 Erkens,A. 7 erosion 254 errors identification 166,I7I inevitabilityof 32 minimising 32 essentialmodules 354 evaluation 119 comparisonofprocedures 132 conceptvariants 163 during conceptualphase 132-9 embodimentdesign309-14 i n c l i v i c l usatlc p si n 1 3 3 , 27.312 c v i r l u i r t i ocrhr i r r t 1 2 3 1 392 evaluationchecklists 131,135 weightingfactors 135 evaluationcriteria 262,296,310,311 compifingparameters122-3,135 derivationof 120,134 identifying 1192I,134 of approximatelyequal importance I22 weighting 121 evaluationuncertainties130-1,136, 138,144 existingsolution 18 existingtechnicalsystems 83-5 expansioneffects 22840 exponentialequations 333J fail-safeprinciple 183 fatiguesafetyfactor 398 fault-identification 366 fault-treeanalysis 298-301 finite elementmethod 227-8,362 first solutionconcept 78 fits 328 flangedconnection 235 flangedjoints 247 flowlinesof forces 196,201,2045, 22t,230 forceddeformationdiagram 198,216 force transmission 228 in bearings 202 principles of 195-205 path 272,222 forcetransmission principleofdirect andshort 197-8, 205 form design 8, 29, 167,174-8,791-3, 24V311,320 for components 272-81 for jointing 279 for primary shapingprocesses 2724 for secondaryshapingprocesses 274 for separation 276-9 forminterrelationship 28_9 forward stepsmethod 34 Fouriernumber 331 frettingcorrosion 201 l'rictiondrive 2l'7 l r i c t i o n atlh r c a d - l o c k i n g d e v i c9e2 4 I u c lg a u g cd c s i g ns o l u t i o np r o p o s a l s I l 5 l t r r r c t i r rcna r r i c r s l 7 1 , 2 8 6 l u n c t i o nc o s l s 3 . 5 0 .1.1.1 ntotltrlcs 3.1.'1, Irrrtctiort I r r r t c t i o t t - r l r i c t r l isl tyet(t lt l l c s i sI ( t Index function structure 2M,66-82, 157 derivationof 76 developmentof 80 establishing I42,34b practicalusesof 77-82 representationof 82 selectionprocedure 79 simplification 79 studyof 77 variantsof 75 function variants,overall 351 functionalinterrelationship 234 functionalrelationships,logicalanalysis of 69J2 '182 functionalsafety fundamentalsof engineeringsystems 2C_31 fundamentalof systematicapproach JI_/ future development 5l gatevalve 176 gearcouplingsizerange 326-'7 gearcouplingtest rig 114 gearpumps 337 gearboxes 206,207,212 modular systems 35'740 generalconstraints 30 generalobjectives 30 generallyvalid functions 68 geometricscaling 319 geometicseries 322 geometricsimilarity 23I, 315, 316, 319 globalapproach 33 glued connections 199 grinding 279,280 hammerforging 274 Hansen,F. 9 heat transfer coeffient 240 helicalcompressionsprings 393 187, helicopterrotor-bladeattachment 211 Holliger,H. 32 honeycombstructures 84 Hooke'slaw 242 hoopstresses 221 l l t r h k aV, . 2 0 hydraulicprolcclionsystcm l Uu mag,nct Itydro-clcct ric gcrrcrat()r whccl 271 rot()r' gcrtcrirl()t Ityrllo-clccltic rlrnrltueli(ln 170 19.1 Index ideal system 34 identificationof components 292 IEC (InternationalElectrotechnical Commission) 259 'if then' relationship 69 inertiaforces 203,317,318, 330,348 inevitabilityof errors 32 informationcollection 54,166 information conversion 354,38, 42 informationfeedback 18 informationprocessing 22 information quality evaluation 36 information storage 56, 83, 105, 111 information systems 37 initial effect 214 inputs 20,23 inspection 294 inspectioncover 275 integralconstructionmethod 269-71 inter-disciplinarycollaboration 334 intuitive bias 86-92 intuitive throught 31 ISO (InternationalOrganisationfor Standards) 259 joining methods 293 K e s s e l r i n gF,. 1 , 7 1 9 ,f 2 2 , 1 9 4 keyed connections 20I Kienzle,O. 3I9 kinematicsimilarity 316 Koller, R. 13,26 Krumhauer,P. 26 Krihnpast,R. 2I8,222 labyrinth seal 58 laminar flow 332 lathe controls 333 Laudien,K. 7 layouts 167,174-8,791-3,230,311 learningprocess 17-19 Leonardo da Vinci 6 leverlaw 26 L e y e r ,A . 7 , 8 , 7 9 4 lightweightstructures,sandwich constructionfor 84 linear expansioncoefficient 228 finearrelationships392 l i t c r a t u r cs e a r c h 8 3 I o c k i n gm c t h ( ) ( l s2 9 4 krgicirllrrrtcttorts(r() logicalrelationships69J2,78,84 logicalsystems 72 logicalthinking 362 long-termloads 246 machines 20 machiningprocesses276,277 Maduschka.L.2(n.201 Magyar, J. 200 main functioncarriers f67 main functions 24 maintenance 171,181,794,228,313 market analysis 46 market conditions 47 massproduction 3,296 matcheddeformationsprinciple 199-203,205 materials 22 materialsbehaviour 240 materialsconversion 79 materialsselection 236,28I,284 methods 8, 110 2 mathematical M a t o u s e kR , . 7,8 mesurementson existingsystems 84 method 635 89 milling 219 model tests 84.316 modulardevelopment 343 modular productdevelopment 346-54 modular productsystematics 343-5 modular products 34241 modular systems advantages and limitationsof 354-5 developmentof 346 examples 355-61 open 345,360,361 modulecharacterisation 345 module development 345 modulusof elasticity 247,244 monitoring 72,18V9 bearingoil pressure 225 morphologicalmatrix 108 motion forms 29 motion variations 97 motor vehiclefuel garage 61 movementof parts 293 Mtiller. J. 10 multiple gears 213 NAND-functions 69 r r a t u r asl y s t c masn a l y s i s i i 3 r r c g i r l i orrtrr c t h o d 3 4 394 new developments 59 nickel alloys 230 N i e m a n nG , . 7,8 NOR-functions 69 NOT-functions 69,72 objectivestree 120,122 oil scraperrings 320 mixingtap 151,180 one-handed one-offproducts 2 operation 177,181,194,228,313 operatorsafety 182 Opitz,H. 317 8 optimisationcharacteristics OR-functions 69,72 OR relationship 298 2,3 organisationalaspects model 2 organisational originaldesigns 4, 67 outputs 20,23 overalleffects 214 overallfunction 24,29,58,66,73,75, r09,11s overalllayout design 266-72 overalltask 24 overallvalue determination 126,B7-B overlappingjoints 199 overloadprotection 22Vl overridingsimilaritylaws 328,33V2, 340 overridingstandards 328 overridingtaskrequirements332-3 Paland,E. G. 200 pantographconstructions 315 partslists 354 physicafeffects 26,73-:7,92,105 physicalinterrelationship 26-8 physicalprocesses 26,924 picking-upoperation 293 pistonwith unstablecharacteristics223 plannedinstabilityprinciple 225-7 plant 20 plasticdeformation 192 positioningrcquircmcnts 293 possiblcmoclulcs 354 p o t i r t oh i r r v c s t i r r n g a c h i r r c7 6 . I l 0 p o w c rl u n c t i o r t s 3 3 3 PrclcrcnccJrrrtcctlttrcl0() prclcrrctlttunthcrscrics .l l9 1.5,.1.15 prc-krirtlctl cllccts l.l.l tr p r c r s u r er l i s t r i h t r t i o r tl : . 1 Indcx pressurelossfunction 332 primary shapingprocesses 2724 principleofuniformstrength 194,205 problem analysis 32 problemformulation 5946 problem-solving 38-40 asinformationconversion 354 procedure-inherentshortcomings 131 producercharacteristics 136 product definition 49 product ideas 45,47-B product planning 45-9 product selection 48 product structuredata 374 production I77, I79, 794,228,373 productioncosts 264,283 productiondesignaspects 264 87 productiondocuments 284,351 productionmodules 343,346 production-orientatedstructure 364 productionprocedure 265 productionrequirements, overriding 333 productiontimes 264,288-9 professionalknowledge362 progresscheck 39 protectiveequipment I87,193 protectivesystems 187,189 'provideone-wayaxialmotion' function 108 psychologicalaspects 1,3 q u a l i t ya s p e c t s 2 3 , 5 1, 2 6 4 qualitycontrol 717,179,194,228,313 questions method 34 R 10series 319 R 20 series 320 rating diagram I25 rationalapproach 5,6 Redtenbacher, F 6 redundancyprinciple l8+ 7 R e i d l e rA , . 6 relativeexpansionof components 23540 relaxation 2434 r c l i a b i l i t y c s t i m a t i ol n3 l rcpcillparts 377 rcsirlualstrcsscs 202 rcs()nirnccs192,227 l { c u l c n u xl,; , ( r R c u t l r cW , , .11(l l l i r ' h t c r ,A , t t r Index Ringfederconnector2IO risk assessment 190 risk designaspect 303-14 R o d e n a c k e rW , . G. I0,26,92 Rohrback,B 89 Roth, K. 12,26,104,738,227 Rotscher,F. 6 395 signalsflow 82 similarity at constantstress 317-9 similarity concept 316 similaritylaws 315-9 designsbasedon 331 overriding 328,33U2,340 similarityrelationships 317,319,3312 simplicity 172,177-81 ,315, 333 safe-lifeprinciple 183 sinteredcomponents274 safetyaspects 173,l8l-94,228 situationanalysis 46 designingfor 189-94 sizeeffects 196 safetyblow-off valve 298 sizeranges 315 safetychecklist 189 developmentof 342 geometricallysimilar325-8 safetyfactor 191,398 safetyfencecontacts 188,189 semi-similar 32U2 safetyprinciples slidingarrangement325 direct 182-7 slidingcontrol valve 779 indirect 187-9 solderedconnections199 safetyrequirements 30, 193 solution concept 29 safetytechniques 181 2 solution principles 28,29, 32 safetyvalves 225 combinationof methods 912 salesforecast 334 combiningto fulfil overall sandwichbox girder 84 function 108-10,146 sandwichconstructionfbr lightweight for sub-functions 85, 98-101 structures 84 intuitive biasmethods 86 92 schedulerequirements194 searching for 824,143,158,348-50 s e a l sa n ds e a l i n g B f , f 9 2 , 2 0 8 , 2 1 4 , 2 1 8 , selectingsuitable 162 219,239,268,313,322,348 selectionof combinations of 148 searchfield 48 sub-functions 82 secondaryshapingprocesses 274 to fulfil sub-functions 143,158 selectioncharacteristics 105 solutionvariants 83 selectionchart ll2 binary evaluation 130 self-balancingsolutions 218-20 optimisationof 384 seff-damagingeffects2154 rough comparisonof 130 self-helpprinciple 95,2812 Sommerfeldnumber 331 self-limitingeffect 224 specialfunctions 344 self-monitoringsystems188-9 specialmodules 344 self-protectingsolutions 22V2 specification 514,77, 134,151,362-3 self-reinforcingarrangements276 analysisof 59 self-reinforcingbrakes 278 checklist 54 self-reinforcingseals 218 contents 51 self-reinforcingsolutions 276-8 examples 56 self-sealingcover 2f4 format of 52 semi-finishedmaterials 2814 furtherapplications 56 separatingprocedures276 Iistingthe requirements 534 shaft-hubconnections 35,97, f04, 115, method of compiling 534 201 recommendedlayout 52 test rig 139 stability 192,227 shapegenerationofcomponents 397 stabilityprinciple 223-5 shearstresses 99,200 staff exchanges 2 shrinkfits 201,218 standardcomponents284 shut-offdevices 227 standardisation signals 22 asa straitjackct 2-5tj s l 256S s i g r t i r l s c o n v c r s i o2n3 , 7 9 o t rj c c t i v c o Index 396 standards 56,231,25ffi4 developing 2624 typesof 258-60 using 260_2 staticsimilarity 316 235-1 steady-staterelativeexpansion steaminlet pipe 233 steamturbine housing 240 steamvalveoperation 305 steamvalvespindleseals 239 stepcharacteristics322 stepsizes,optimum selectionof 322-5 storagerequirements 291 strengthdiagram 136 strengthproblems 22'7 stressconcentrations 196,199,202,227 stressconditions 227 stresscorrosioncracking 253 stressdetermination22'7 stressparameter 318 stressrequirements227 structureanalysis 33, 84 stuffingbox performance 304 sub-functions 24-8,67-8,73,74,'76,78, 97,286 assignmentof 2054 relationshipsbetween69 solutionprinciplesfor 82 6,98-101' 143,158 subjectiveerrors 130-l sub-objectives 120 sub-systems20, 21,24, 52 sub-values 126,131,132 superheatedsteamPiPes 2ll clampconnectionof 208 supplementaryeffect 214 surfacefinish 96 surfaceforms 28 surfacevariation 97 generatorcoststructure 287 synchronous synchronousgeneratorrotor 26'7 synectics 90-1 s y n t h e s i s3 3 , 3 8 syntheticmaterials230 systemboundary 20 elements 21 system s y s t e m a ta i cp p r o a c h 3 l - 7 , 3 6 2 4 l 0l n ll-10 systcmaticc<tmbinati< dcsign 2 systcnratic ttl I lanscn 9- l() accordirtg ttl Kollcr l3- l5 accorcling to l\ldcttttckcr l(l | | itccordittg lccortlirtgto l{otlt ll ol (r l() tlcvckrpntcttl general comparison and statementof authors'own aims 19 historicalbackground G8 importantcontributions 6-8 nature of I 6 need for 4 6 systematicselectionchart 113 systematicthinking 362 s y s t e m a t r cv a n a t l o n 3 5 systems 20 systemsapproach l5-7, 33 systemstheory 15 task clarification 40,49-55, 151 task division for distinct functions 206-11 task division principle 205-13 taskrequirements, overriding 332-3 task-specificconstraints 30 teaching methods 5 technical artefacts 20 technicalcharacteristics 134 technical composition theory 7 technical criteria 48 technicalevaluation 136 technical factors 348 technical function 30 technical rating 128, 136, 137,309, 3 I 1. -) L+ technologicaldata 83 technologicallimitations328 technologystate 50 temperaturecurves 237 temperaturedistribution23(u-1 temperatureeffects233,24{l_-1 tendencysign 136 tensileforces 208 tensilestresses 279 tensiletestingmachine 73 thermalexpansion 92,228y'.0 thermal processes 331 thermalsimilarity 31.6 thermal strains 232 thermalstresses 246 thermo-stablebehaviour224 time commitment 413 tolerances 328 tooling 27'7 torcluc-limiters328 ttrrsionaloscillations 2Ml stil'llrcss 202,203,21| ttrrsiottal torsiortllslrcsscs 97 t{}ughncss 9 | l r l n s p r l r l 1 7 7 ,l t l 0 . 1 9 1 . l l l ' t , .11. 1 Index T s c h o c h n eH r ,. 7, 8 turbineblades 220 turbinecasings 209,233,248 turbo-machinery 21V24 turning 278 uniform strengthprinciple 796,205 unsteadyrelativeexpansion 237 use-valueanalysis 116-18,130 value analysis 24, 33, 119,2854 value assessment 1244,136 value function 125 value profiles 7312 valuescale I25,130 variantdesigns 4,67,404 V-belt drives 2lI2 YDI Gddeline 2222 19, 43 VDI Guideline2225 719,1212, 124, \28, r33, 136 Velcro fastener 83 vibrations 92,227 391 Wrichtler,R. 77 warningsystems 181,187-8 weak links 33,220 weak spots 131-2,138,I54 wear 192,228 Weber-Fechnerlaw 320 weight effects 330 weightingfactors 1272, 126,r35 welded components 28I weldedconstruction 283 welded joints 202 welding process 279 Wiegand,H. 200 winding device 307 winding machine 269 Wcigerbauer,H. 8 working principles 178 y i e l dp o i n t 1 9 8 . 2 3 7 , 2 4 62 ,5 4 yield strength 276 Zimmermann,D. Zwicky,F. 108 36 lrs*refortir in*reaseti-reci:ait*s:ssf 3L:*e#"s.c f*i' c*"* *ng1in*erilr.g prcd*cts, especiailiv eempiexonesdesigiitieJ l"ryiarg+teernrs of sp*ciailetr:, ti:e ciesig;i prorJsss mustbe carefu!iypiannedanr:i*y*iei"r:;tticnii'y' *xe,;i.rterl. F*r;'thist* be posr::tlls, th* riesi,;l prc*eiq$ f-Iir.ist il# *r*[r+:*d*...4/1, filst ir-rj"* Bha**eii;-'d "iilis th*l: lrrtr ;icps, *ach rs'ithil; ap$i"*i.rri*t*: ri:*rh,teis" editi*ii cf ihe q*s'rd ii':liuei-rtiai br:ckK**'*stra.*k*imra$6c*$lr*: F.{ar*dfu***'f*rSiiueiEue=i: Fraxisu,byFrof*:;s*rsS*riiereiF;:i'ilnnrj\fui:lfgi*ng; Liei*:,iaysd#in*'n .* siruttegy f*i this pificitssa;reihrir';gs tcg*';i"i*r'th* +xte;:sive i:r:iiynf kn*wir:iig*ahcui tiloiiertiaFl;r*achestc;sysl*m*ti':{t*si*n. i",lc; +tll*:-b**k pr+l'icf*s in fii-rillisl'r s,-l*han ineighiintotlei=niarr thinliinge::*l-tiengirieering anei#esign nretl'icC*;cg',r. i'{*nbV{}liac*i:i.il:ir,i,,i. {"Ingiviti,4e,:iii:., wh* sdite#r'h*fi;-*if.:iiEiisn ili:i-rslaii*i:, prr:pnreif tirisn*;ueriii.i*i'r witl:sL:gg;+:stii:rrs iil rksi6rrtaari:ing.He ior iis r-rsr* is Univr;reitli [..eq:lursr in i:l:glin**l'i;'rei L']+sign and ;r Fisiir;u,r'cf Seix'g'r: Ll,;iia#e,eanri:rir.ige, F-ngiand.