0% found this document useful (0 votes)
228 views60 pages5-14magnetic Circuits Chapter 4 - 118-177p PDF
Uploaded by
anunilaCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
228 views60 pages5-14magnetic Circuits Chapter 4 - 118-177p PDF
Uploaded by
anunilaCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
/ 60
4
Magnetic Circuits
BASIC PRINCIPLES OF MAGNETIC CIRCUITS
41, Basic Formulae, The path of the magnetic fiux is culled a magnetic circuit,
‘A magnetic circuit is analogous to an electric circuit. A review of laws of magnetic circuits
is given below.
Let ‘O=magnetic flux, Wb ;
Amarea of the magnetic path, m* ;
Jmlength af maguetic path, m ;
Bemfiux density, Wh/m? ; H=‘at'=pragnetising force, A/m ;
AT=total mmf,
pr wemabsolute permeability of the magnetic material, H/m ;
e=permeability of free space=4n x 10-7 H/m ;
he=relative permeability ; S=reluctance, A/Wb ;
Amp =premeance, Wh/A.
In an electric circuit Ohm's law expremes a relationship betweea current, emf, and
resistance ; while in a magnetic circuits a similar relation exists rélating flux, mmf and
reluctance. This relation is t
eG)
or e(4°2)
Reluctance (43)
‘H=‘at'=mmf per unit length=flux ¥ reluctance per unit length
woxt Lo® LLB wl’)
Tae A ee
For the case of a material of length J, and carrying a uniform flux, the total mmf AT’
AT=HXt='ai' x1 (3)
Tn a series magnetic circuit, the total reluctance is the sum of reluctances of individual
or S=5,+8y+Sy+-..
where. ‘Smtotal reluctance
and Sy Sy, Sy... mereluctances of individual parts,
(46) MAgnaTiC cincutrs a9
The total mmf acting around a complete magnetic circult is:
The mmf AT OSH 915,45, 444.000]
SAT,+AT +A
math +ataly tat, +.
or ar=§ ‘at’
Eqns, 4°7 and 4°8 represent the citcuital law for magnetic cir
etc., are the mmfs pcr metre for individual part and J), ly ty.
connected in series.
In parallel circuits, the same mmf is applied to each of the parallel paths and the total
flux divides between the paths in inverse proportion to their reluctances, as in corresponding
electric circuits or
PaO Fyfe
9)
Flux density B—>
Wb/m*
at
Alm 120 ELECTRICAL MACHINE DESIGN
ing be AT, the applied mmf
aM Oe yO
Par ”
1
or phate
Dat Natbenes as (4°10)
where Se-totcl reluetonce of magnetic circuit and
Si, San 85 chictance of individual parts
A =total permeances of magnetic circuit= 1/5
while Ai, Ag Pars onete, are prorn.eances of individual parts
-
4+
ee
wii |
5 1
fe
7 i
c 4 ; y
t
2 = ae
“
aun 7 war
Alm
Fig. 42, B-—/ curve for Blectrical Steel (non-oriented)
‘Lobys stampings (dynamo grade). ‘MAGNETIC CIRCUITS m1
_ 42, Magaetization Curves (B—H or B—‘at? Curves), The aniogies between
electric and magnetic circuits have certiinly a valu ax aid: to unlerstanding of magnetic
circuits but there are some estential differences between them :
1, The electric current is a true flow of electrons but there is no flow of magnetic
flux and so the term flux is misleading in this context.
2. In an electric circuit, energy is conwumed so long as the current flows while in a
magnetic circuit, energy is expended in creating flux and not in maintaining it.
3. The most important difference b-tween clectric and mignetic circuits arises from
the fact that whereas the resistance of an electric ircuit is not directly dependent upon the
value of current flowing and may be regarJed ats constant for ontinary purposes, the value of
reluctance of a magnetic material is dependent upon the vaiue of flux passing through it.
The relative permeability of ferromagnetic materials may vary between few hundred and
about five hundred thousand. Consequently the mif per anette length ‘af’ has a value that lies
between 2 B to 2000 8.
For this reason (No. 3 above) the actual valnes of permeability aud reluctance are
hardly used in magnetic circuit catculatinns Ty actual practice magnetization (B—H oF
B—at") curves of magnetic materials relating flux density #19 mmf per unit length, Ho af
are used for the rapid determination of nec seary Oxeiation
421, Bia? Curves, Fig. 41 shows the B -H curves of commonly used magnetic
materials Strictly, each specimen of a ma v unique Bat’ relation, but it is a
practice to tse average curves as given in Figs 4'1—i'4. Figs. 4°2, 4°3 and 4°4 show
H curves of dynamo grade stecl, transformer grade ster] and cold rolled grain oriented
steel respectively as supplied by Sankey Prestings Division of M/s Guest Keen Williams Ltd. .
For work on digital computers, the analytic relations between Band H prove more
convenient. Two of the most vsed mathetical relationships are given below :
a Sor (Hl)
and Batot el bagi + wof412)
16 T bE
where a, 5, do, 04, 4, bi, by are constants, Eqn. 4'11 gives reasonable and approximate values
while Eqn. 4°12 is Letter overall fit. Ifalternating magnetization is used, only odd powers of
H must be ured in Eqn. 4°12.
Non-magnetic materials (like air ete.) have a constant value of permeability and 20
the Be‘at’ curve for them is a straight line passing through the origin.
For air or any other non-magnetic material, mmf per metre
a: Bw By ;
a Bm T8000 B oAO13)
where B is in Wb/m*.
43, Magnetic Leakage. It is impossible to confine all the magnetic flux to
ven path (there being no magnetic insulator), and therefore the designer's problem
Eecomé: that of providing a path of low. reluctance so that comparatively little ux
leaks away from. the, path and then supplying somewhat langer mf to, compensate
for the flux which leak, away. This fiux which strays away completes its circuit by
paths which prevent its utilization in the functioning of the apparatus or machinery.
For the operation of electric machinery, some air gaps are necessary in the magnetic
paths Dut these air gaps should be Rept 0 cainhoun of Teogth and maximum of
cros-section so as to reduce their reluctance. A bong air gap of small cross-section would 122 LECTRICAL MACHINE. DESION
require «large mmf resulting ia Large coils of many turns and would also result in a
fandency for flax to wander away frxm its main path. This flux which strays away
from the main path is called the leakage Aux, The leakage flux doss not contribute
to either transfer or conversion of energy.
However, the leakage flux affects the performance of rotating machines and trans:
formers, The leakage flux affects the excitaton demands of saliant pele michines, the leak-
‘age reactance of windings on which
the performance of the a.c. machines
bs primal ave he forces between
especially under short con-
ditions, voltage regulation of a.c.
generators and transformers, com-
Mutation conditions in d.c. machines,
stray load lowes, circulating currents
in transformer tank walls and several
other performance indices of impor-
tance.
For ic circuit calculations,
aterm ‘ co-efficient’ is in-
troduced in order to take into account
the leakage flux. The value of this
leakage co-efficient is defined as
‘Leakage co-efficient,
Flux density B—>
‘Wh/m*
useful flux-+leakage flux
‘useful ux
total flux
"useful fix
Ce
w4'14)
“al
Alm
Fig. 43. BH curve for electrical steel (non orien
Ghades Band 190, Grades 91nd 6 od
rol wom oriented ‘transformer steel) of M/s ce.
MAGNETIC CIRCUIT CALCULATIONS
44, Calculation of Total Mmf, The calculation of total mmf required to
establish the requisite flux in a magnetic circuit involves. the knowledge of dimensions and
configuration of the magnetic circuit. The magnetic circuit is split up into convenient parts
may be connected in series or parallel. The flux density is calculated in every part
and mmf per unit lengtl is found by consulting ‘B.‘at’ curves. The summation of
mms in series gives the total mmf.
‘The method looks quite simple but there are some Parts in the majmetic circuits,
like air gap and tapered teeth which present complex magn‘tic problems. ‘These problems
are solved with special techniques outlined below. ne proiem
441, Mmf for Air Gep,
Let Lelength of core, ly=gap length,
ymslot pitch, Wemwidth of slot,
We=width of tooth, Wemslot opening,
‘se=number of radial duct, and Weewidth of each duct, a fe
fe
ae
s
.
£
Ea.
i"
ae
o
° S05 — 1000 ~~ Ta08
wae
Ayn
Fig. 4-4, B—H curve for Electrical steel (cold rolled grain oriented) Grade 123 (Grade 56 of M/s G.K.W.)
The iron surfaces around the air gap are not smooth and so the calculation of mmf
for the air gap by ordinary methods gives wrong results. The problem is complicated by the
fact that +
1, One or both of the iron surfaces around the air gap may be slotted so that the
flux tends to concentrate on the teeth rather than distributing itself uniformly over the air
P
Po. There are radial ventilating ducts in the machine for cooling purposes which effect
in a similar manner as above.
$8. Insalient pole machines, the gap dimensions are not constant over whole of the
pole pitch.
Consider the iron surfaces on the two sides of the air gap to be smooth as shown in
Fig. As ‘The flux is uniformly spread over the entire slot pitch and goes straight acroas the
air
—
I my — + ha
Fig. 43
tn on ty mn ie rt
be J ofS 124 ELECTRICAL MACHINE DESION
In a slotted armature, however, the effective area of flax path is substantially
decreased resulting in an increase. in reluctance of sit gap. Consider the case of a aloted
armature with a very small gap length as shown in Fig. 46. The flux in this case is only
Confined to the tooth width
effective or contracted slot pitch
ya Menye We 416)
Reluctance of air gap of a sloted armature
ly Ly ‘"
= =-— 41 wll
‘ pelly Wd (in
There is, however, some fringing of flux around the teeth edges in a slotted armature.
The flux penetrates down the slot as shown in 4°7, It is obvious that the reluctance of
air gap in this case is more than that in the case cf a smooth armature (Fig. 4°5) but lesser
than that in the case where the whole flux is assumed. tobe confined over the
tooth width (Fig. 4°6). A simple method to calculate reluctance in this case is to assume
that the airgap Sux is uniformly distributed over the whole of slot pitch exzept for a fraction
of slot width as shown in Fig. 4°8. This fraction depends upon ratio of slot width to air
gap length, ‘Thus the flux of one slot pitch is distributed over Wet 3M
3
g
3
=
Baty
ta
Me
” 4
sto? | roar
Fig. 4-7 Fig. 8
Effective or contracted slot pitch
Web WS Wet Web BW We
where Kesis the Carter's gap co-efficient
which depends upon the ‘ratio slot
p_lengtl e value of
Garner covefitient can be. taken
from Fig. 49. An empirical formula
which gives the value of Ke dirctly
Kum man -#i9), §
Another useful relationship
which can be used for calculation of
Carter's co-efficiemt for parallel sided
open slots is :
Kan? tan? wt
tog TF) i] (4-201
Ratio Sot:
where y= W./2ig Fly. 49. Carter's Air Gap Co-eficlent. ‘MAGNETIC CIRCUITS. 125
Reluctance of air gap with slotted armature
sm
*
Lely (421)
Let ratio of reluctance of air gap of slotied armature to reluctance of air gap of smooth
armature be Kw. Therefore, from Eqns. 4°15 and 21,
Ku=x— ol
eT, (4°22)
where Kyo is called the gap contraction factor for slots, Therclore, the reluctance of air
gap with slotted armature is Ky» times that with smcoth armature and Kes has a value grester
than unity.
The provision of radial ventilating ducts results in contraction of ux in the axial
direction as shown in Fig. 4"10. It is clear that the effective axial length of the machine it
réduced owing to presence of ducts and this resuls in an increase in the reluctance of air gap.
We can derive a similar expression for ventilating ducts by treating stacks of latninatfons as
teeth and the ducts as slots.
‘. Contracted or effective axial length
Dim L— Kea na Wa (4°23)
where Kes is the Carter's co-efficient for duets.
Values of Ke can be taken from Fig. 4% by
using ratio (duct widtb/gap lengthy) in place of ratio
(slot widih/gap length).
Let the ratio of reluctance of airgap with ducts
to reluctance of air gap without ducts be Kas.
A424)
Vouets,Wg wise
Fig. 410. Pffect of radial
‘ventilating ducts,
When two ventilating ducts, one on stator and the other on rotor are exactly opposite to
duct widsh
4 gap length
Ky is called gap contraction factor for ducts,
each other, the Carter's co-efficient must now be based upon ratio ,asto an
approximation we can ascribe only one half gap to each duct.
The effect of both slotting and ventilating ducts can be allowed for in a single ex-
Pression. Considering one slot pitch.
Reluctance of air gap of a smooth armanice without duets
le
pole
Reluctance of air gap of a slotted armature with ducts
iy
hel ye
reluctance of slotied armature with ducts
of smmcoth armacure without ducts
(425)
Ky=Kux Kot on(426) 126 ELECTRICAL MACHINE DESION
‘where Kis the ratio of reluctance of air gap of a slotted armature with ducts to reluctance
of air gap of a smooth armature without ducts and is called total gap contraction factor
for slots and ducts.
For induction motors, with slots on both sides of air gap, it is customary to calculate
‘gp contraction factors for both rotor and stator slots.
Eyu™gap contraction factor for stator slots,
Kye™gap contraction factor for rotor slots,
and Eymtotal gap contraction factor for slots,
Kin Ko. 427)
‘The expremions involving Ke used above
apply only to open parallel sided slow of great
depth. When slots are shallow, Kw is decreased.
‘Also when semi-enclosed slots are used Ke is
increased since now it depends on ratio (slot
opening/gap length) rather than on ratio
(lot width/gep length). (See Fig. 4°11)
Fig. 4°9 also shows values of Carter’s co-
efficient for semi-enclored slots. Eqns, 4°18 and
Fig. 411 4°22 are modified for semi-enclosed slot
Contracted or effective slot pitch ys’=y.—KaWe
Gap contraction factor Ken
‘The reluctance of air gap with slotted armatures is higher than with smooth armatures,
‘The ratio of the two reluctances is equal to Ky, the gap contraction factor. In other words»
the mmf required for the gap with slotted armatures is Ke times the mmf required for gap
with smooth armatures. From Eqn. 4°11,
‘Manf per metre for air gap=800,000 B
‘Manf required for air gap having a length ly metre, with smooth armatures,
AT,=mmf/m x lengtl==800,000 Bly
.. Mamf required for air gap, having a length ly, with slotted armatures,
AT,=800,000 Ke Bly (4°31)
Eqn. 4'$1 can be written as t
AP = 800,000 Kr le
7
(4°30)
where @-=flux per pole, Wb ;
Aymactual area of air gap per pole, m*,
Now 41,000,000 35 ¢-b
=100,000 32-1 (6°92)
where A AslKe (38)
‘This means that there is a contraction in the air gap area and the gap area has com
tracted to a value Asal , = ‘MaoneEnC chacutTs
‘Area of gap per pole
427
Ay=slot per pole x slots pitch x core length
8
=e xy.
peuxh
(4°34)
where S=total number of slots and p=number of poles.
Contracted or effective gap area per pole
Sip\x yo L, Ly
ge 8 Kem at
Bu rent
waxy! x.
pe L
4°33)
‘The mmf required for a smooth armature is :
ATy=800,000 B ly
while for a slotted armature the air gap mmf is :
AT4=800,000 KsBly™800,000 B (Kyle)
The above relation may be interpreted as that the length of air gap is increased X,
times (instead of saying that area is reduced to 1/K times as earlier) due to the provision of
alots and ducts, Thus effective gap length
lw=K
(4°36)
and therefore, K in this case is called “gap expansion factor”.
Effect of Saliency, In the case of salient pole machines, the length of air gap is
‘not constant over the whole pole pitch.
This gives rise to different values of air
density over the pole pitch. ‘Thus to
Erbw the value of reluctance of the air
ap, it in necessary to know the distri-
jion of magnetic field in air gap. The
methods for the determination of flux
density distribution are given later in
this chapter.
4°12 shows a typical flux
diswibuga curve for a salient pole
machine.
Fig. 4°12 (a) shows the flux tubes
passing from field to armature. The
equivalent electric circuit for this
magnetic circuit is a number of resistances
connected in parallel with each resistance
2 flux tube. In parallel electric
circuits, it is sufficient to know the value
of only’ one resistance and the current
flowing through it in order that the volt-
age across the circuit be known.
Similarly by analogy, we have
only to, know” the” reluctance of one fox
tube and the flux flowing through it in
order to find the mmf required for air gap.
Fig. 412. Flax density distribution curve ta.
‘alieat pole machines.
cs Mmmf required for air gap of a salient pole machine
. AT = five in a flax tube X reluctance of flux tube.
‘Let us considee « fun tube atthe contre of the pole {28 ELECTRICAL MACHINE DESIGN
Flux in the flux tube at the centre
= flux density X area of flux tube
=H, x area of flux tube,
Reluctance of flux tube at the centre
__-ffictive length of air gap at the centre
=~" peXarea of flux tube at the centre
he
Xarea of flux tube at the centre
APy= By X acca of flux tube at the centre
x
“nx 10"? Xarea of flux tube at the centre
= 800,000 Bekele (4°37)
aap contraction factor fora gap length atthe centre of the
pole.
‘The flux tube at the centre of pole is chosen because its actual length ix known. The
length of the flux tube at the centre of the pole is exuctly equal to the length of air gap there.
‘Therefore, the value of K, is based upon the air gap length ls.
‘The field form factor Ky ‘s defined as :
verage gap density over the pole pitch Ba
maximum flux density in the gap By
where
Ky=2 (4°38)
fiuxperpole ©
We have Bem crea per pole +L ~ (nD/p\E
where s=pole pitch=nD/p
‘The value of By can be calculated after determining the value of field form factor
Ey. The methods determining Ky are given later in this chapter. Fig. 4°12 (c) shows, an
approximate flux distribution curve for a salient pole machine. From this,
= average flux density
field form factor Er maximum flux density
—pole are '
= pole piteh y +(4°39)
‘The assumption Ky=Wis fairly correct for machines with normal proportions and a
fair aegree of saturation. With long air gaps the pole arc may be considered as extended
by say, 4 lp to take account of fringing.
Example 41, The stator of a machine has a smooth surface but ile rotor has open
type of alote with elot width Weestooth width, We=12 mm. and the length of air gap lem 2 mm.
Find the effective length of air gop if the Carter's co-ficientme 75 1+ There are no rodiat
duct.
Solution, Carter's co-efficient for slots
1 .
Kam paTig 70:
Slot pitch y= Wet We=24 mm,
From Eqn. 4°12, gap contraction for slots
——
fam TR 24—0'545X
Since there are no ducts, gap contraction factor for ducts, Kyl. MAGNETIC CIRCUITS ag
From Eqn. 426, totsl gap contraction factor Ky= Kee Kya t'97 4 1237.
Effective gap length (sce Eqn. 4°36) lw=Kely= 1°37 %2=2'74 mm.
Example 4:2. Calculate the mmy required for the cir gap of a machine having core
length—=0°32 m including 4 ducts of 10 mm each. pole arce= 0°19 m ; alot pitch=66'4 mm ; slot
opening=65 mm ; air gap length=5 mm ; flux per pole=52 mW. Given Carter's coveffcicnt
ta 0°18 for opening/gap=I, and ie 0°28 opening|gap=2.
ing _ 05
a
ws Carter's co-efficient for slots Kam 0"18.
This is a salient pole machine with semi-enclosed slots.
Gap contraction factor for slots Kum ——Btay- (see Eqn, 4°29)
p—KaWe
“e540 Texs @! 014
duct width _ 1
“gap length 05
Carter's co-efficient for ducts Kas=0'20.
Gap contraction factor for ducts
Eu= ikea (See Eqn. 4°24)
Ratio
Total gap contraction factor Ky-=1"014x 1°36 =1°05.
Aux/pol
‘pole arc Xcore length
= 2X 10°
O'19x 0°32
Mnf required for arr gap 47 800,000 Ke Ly ly
= 800,000 x 1°05 x 0°854 x 5 x 10-3587 A.
Example 4:3. Eitimate tht effective gup area per pote of a 10 poie, alsp ring induction
motor with following data :
stator bore = 0°65 m, core lenyth=0'25 m, No. of stator slota=90
stator slot openings=3 mm, rotor siots = 120
rotor slot openings mem, air gap length=0°96 mn
Carter's coefficient for ducta=0'68, Carter's coveficient for slota= 0°46
umber of ebntilating duce=3 each on rolor und slator,
width of each ventilating duct=10 mm,
Sotation. Stator slot pitch=. mx6s0 2227 mm,
Gap contraction factor for stator slots
. . - 27
Kae es Te TIS.
Flux density at the centre of pole By=
=0'854 Wh/mat. 0 HLBCTRICAL MACHINE DESION
Rotor diameter=0°65—2x0'95=0°6481 m
Rotor slot pitch = TAGS 01697 m
‘ ___1697
Gap contraction factor for rotor Hots Keemrgy 9 ml 089
Gap contfaction factor for slots Kum KouX Kea 1065 x 1'089—1°16.
Gap contraction factor for ducts Kate pp
250
“30-Teexax10 1089
Total gap contraction factor Kym Kno Kegm 1°16 1-089—1°26
‘Actual area of air gap per pole y= TOS) x 0-25—51-05x10-* m’.
From Eqn. 4°33, effcctive air gap area per pole
Ay! we Ag] Eem=51°05 X 10-41-26 40°52 x 10-* mat,
Example 44. 4 175 HVA, 90 pole water wheel generator has @ core of length 1°72 m
and a diameter of 6°5 m. The stator elote (open) have © width of 28 mm, the slot pitch being
{64 mm and the air gup length at the centre of the pole ie 80 mm. ‘Thare are 41 radial ventila-
ting ducts each 6 n.m wide. The total mmf per pole ia 37000 A. The mmf required for the ait
gap is 87% of the total mmf per pole. Batimate the average fla density in the air gop if the field
form factor is 0°7.
The Carter's co-efficient ean be calculated from the following relationship
Kom 2 tanty——Liog THF]
where y= Wally for slots
Wal? Ie for ducts
W., We are widths of slot and duct respectively and ly is length of air gap.
Solution,
io _tlet width 2.
Ratio ap lengthy) ~Y 72x30 70 57-
Carter's co-efficient for slots
u2 - y—Liog TH
Ke 2f tan y- i105 4TH]
“2 tan 40367 Geplog 4 THOS 367 |-0176
367
4+ Gap contraction factor of slots Bw=— Mey = Tem 064
fo _-dtuct width
myo
Ratio digap length) "Y "2xS0
ce Carter's co-efficient for duets
27 nto ioe
Form 2f an o1- gt vitor fess.
factor for ducts
L 1720
Kom [Kana Wa TH0—OO ATX 1 007
‘Total gap expansion factor Ke 1"1064X 1007 ==1°114.
‘Manf required for air gap A Ts=0°87 x 27000—23490 A,
But AT y= 800,000 Ky By be
<. Gap contract MAGNETIC CIRCUITS 131
Hence maximum flux density in air gap
a. 23490 . .
Be 6 000 Ke Te BOO RGTRT TiaxoR Ie 708 Wel
Average flux'density in the air gap
Bov= Ky By=0°7 X0°8786=0'615 Wh/m*.
442, Net Length of Iron, The cores of magnetic circuits are built up with lami«
nated steel plates wherever required. These laminations or stompirgs ate insulated from
each other by paper, stuck to one side of the lamination, Kaolin clay or enanicl, Moreover in
order to have an effective cooling of the machine, the length of the core is divided
packets of about 40 to 80 mm width separated by vent spacers. These vent spacers iorra
ing ducts through which air is circulated. These ducts are radial as shown in Fig #10
and their width normally varies from 8 to 10 mm.
From above it is clear that whole of the length is not occupied by iron ; rome part of
the length is taken up by ventilating ducts and some part by insulation betwern steel lamina
tions and air spaces ereated by irregularities in thickness of laminations. Tt is usual to define
iron space factor, called stacking factor, as the ratio of actual length of iron in a stack of
assembled core plates to total axial length of stack,
Gress iron length La==length of slot portion conductor
‘=core length—lengih of ventilating ducts
=L—mWe
Net iron length Lam Ki(L—neWe)
where Kemstacking factor for iron which largely depends upon thickness of plates and
the type of insulating material employed. Tle manufacturers specify the
stacking factor fora single lamination. The sacking. factor depends upon
the thicknest of core and the thickness and the type of intulation used for
laminations. The stacking factor for built up cores is smaller and an
average value of 0'9 may be assumed for all practical purposes,
443, Munf for Teeth, ‘The calculation of mmf necessary to maintain the flux in
the teeth is difficult owing to the following complex problems :
1, The teeth are wedge-shaped or tapered when parallel sided slots are used. This
means thut the area presented to the path of flux is not constant and this gives different
values of flux density over the length of teeth
2. The slot provides another parallel path tor the flux, shunting the tooth. The
tooth are normally worked in the saturation region and therefore’ thei, permeability is low,
and as a result an appreciable portion of the flux goes down the depth of the slots. The
Presence of two parallel paths, the reluctance of one part depending upon the degree of
saturation in the other, makes the problem intricate.
Teeth, ‘The mmf required for teeth’ cua be easily calculated whatever
may be their shape, if the ux going down the alot is neglected. ‘The correction, to take
alot flux into aecount, can be incarpprated later on.
Following are the methods usually employed for the calculation of mmf required for
teeth.
i
Grophical Method, The mmf per metre for the whole length of tooth 1s not
10 roel Mtet cot the mie everywhere. ‘Therefore, to obtain correctly
jue of total mea, it i to construct « showing the manner in which
‘over the length ofthe tooth. The mean ato he Eraph sive te equivalent:
the whole of'thé tooth.’ ‘Fhe totat mmf for the teeth is given JH, the ine
‘carried out for the cormplere height of scot; = - :
if
eae
Hi
I 192 ELECTRICAL MACHINE: DESIGN
Therefore, total mmf required for the tooth,
AT =mean ordinate x height of tooth
= almeen X l= atmann X de (4°42)
The height of tooth 1 is equal to ds, the depth of slot.
To determine this atweas, it is necessary to construct first a graph showing the manner
in which the flux density varies. Froin the known value of flux pcr tooth, the flux de
is evaluated for a number of sections along the length of the touth from tip to root (Fig.4.13).
‘The corresponding values of ‘at’ are found from the B—‘at' curve of the material and are
plotted. The value of almen is obtained from the graph, as shown.
-#— ns
t
Flux
1
lyeds |
Distance “rom root Distance fru > root
Fig. 4:13, Bat curve of tapered tooth and Fig. 4-14, Calculation of atmoan f0F tapsced tooth
calculation of afasan using Simgnon's rule
(0) Three ordinate Method (Simpson's rule). This metnod can be applied to teeth of
vary simple form and of a small taper and is bated upon the ion, that the curve
eelating ‘af’ with flue. 'ensity, is. a prrabola. In this metho, values of
equidistant points, the ends of the tooth and its centre, Macnetic Crecurts 133
The mean value of ‘a’ is givin by. :
Otmen = stato =f 4°48)
where at), aly, ats are the values of ‘at’ £. ~ sections shown in Fig 4°14,
(0) Bury Method. This mcthod is applied to teeth of small taper and is based upon
the assumption that value of ‘at’ obtained for flux density at a section 1/3 of teoth height
from the narrow end is the mean of ‘at’ for whole of the tooth. This method is the most
simple of all the methods and results are sufficiently accurate if the teeth are worked at low
saturation.
Let Biya=flux density at 1/3 height from narrow end,
otyjy=value of mmf per metre for Buys as obtained from B— at’ curve
Total mmf for teeth ATr=atyyX ==atys X de (4°44)
444, Real and Apparent Flax Densities, [1 has already been state:] above that the
slot provides an alternative path for the flux to pass, although the Sux entering an armature
from the air gap follows paths principally in iron.
Ifthe teeth density is high, the mmf acting
acrots the teeth is very large and as the slots are
in parallel (Fig. 4°15) with the teeth, this mmf
acts across the slots also. Thus at saturation
densities, the flux passing through the slots be-
comes large cannot be neglected, and any calenla-
tion based upon ‘no slot flux’ leads to wrong results
This means that the real flux passing through
the teeth is always less than the total or apparent
flux. Asa result, the ‘real flux density’ in the
teeth is always ‘less than the ‘apparent flax
density’, Fig. 4-15. Slot fux due to aatsration :a teeth,
‘The apparent flux density is defined
total fux in a slot pi
oy = flat flue ina slot pi
Ba tooth area
h
and the real flux density as:
“Braye 2etval fx in a tooth,
tooth area
pate Inan actual machine, taking the flax over one slot pitch, there are two parallel
1, Irom path,
Area of iron path Ai=tooth width Xnct iron lengtl.=Wex Li
2, Air path,
‘Area of air path Ag=total area—ircn area
m(slot pitch x core length)—(tooth width x net iron length) =ysx L—Wix Ly
If © is the flux over one slut pitch, we have :
= O%4O0
where c=flux passing through iron over a slot pitch,
©,=flux passing through air over a slot pitch.
total flux over a slot ®
+ Bew™“Ton area over wilot a 134 ‘BLECTRICAL MACHINE DESION
%
wit+
A AG
Ae
Boat + GE
Brat + BK (445)
Ba=flux density in air=pgH=4 10-7 alreat
where almei:=mmf per metre across the tooth for tooth density Brut.
ratio At x= tit aca | Lw= 14s .
and K=ratio ‘av ironarea—CLGWe lee)
(4-47)
(4°48)
(4°49),
se Brest Bayy 48 X 10-7 ‘al raat K
= Borp—4nX 10-1 ‘at! (Ke—1)
total area, Lys
where Kyo l +H tt Se oe
(It should be borne in mind that the value ‘at'rat used above is corresponding to the
real flux lonsity Brat). The slot and the tooth form a parallel magnetic circuit and therefore
the mmf across them is the same. Since the length of fux path through slot and tooth is the
same and hence mmf per metre length for both is the same. The mmf per metre in. tooth is
‘rat corresponding to real fs density Brat the maf per meter acrom the dt it alo
‘nal
Eqn. 4°48. cannot be casily solved as there are two unknowns, Brust and ‘at’, the
latter depends upon the value of former. In order to solve this equation for machine’ with
any dimensions it is disirable to have data available which immediately gives the values
of ‘at’regi and Breat for various values of Bapp and Ks. Fig. 416 shows a fini of curves
from which’ Brsot and ‘at’ can readily be obtained. These curves correspond to B—H curve
of ordinary stcel laminations given in Fig. 4"1.
‘Thore is another method available for obtaining the values of Bret aid ‘al’. This
method in comparatively lengthy but useful for individual cases.
Eqn. 4°48, involves two unknowns and so for solution, another equation is needed. MAGNETIC clncUrts iss
‘The second relation is defined by
the B—‘al' curve of magnctic. ma,
terial used for the teeth. B—‘at’
curve is drawn and the straight line
Beoge Bra + 40% 10" ‘ata (Kel)
slaid out asshown in Fig. 417.
‘When ‘st'nei=0, B=Bay. This {
corresponds to point 4. When
Boy
Werking
te
Brat 0, ‘ el FIOM ST) ay
ee
'B- at curve
Sapp Brett 4H 16 eggs,
ten
This corresponds to point 4’,
A straight line is drawn con- bene! Dane
necting points A and 4’. The nara)
intersection of straight line AB with
B—‘at! curve gives the value of Brest Fig. #17. Determination of Brow
and the corresponding value of ‘al'rea. ‘The intersection of the two curves gives the vatue
of Breat and the corresponding value of ‘at'rea!.
Example 4'5. 4 laminited tooth of armature steel in an electrical machine is 30 wm
long and has a taper auch that mazimum width is I'd times the minimum, Eatimate the mmf
required for a man fius density of 1°9 W/m in this tooth. Uae Simpeon'a rule. The Bal?
cuirve for the material of tooth
Tree
== [=== |= [|=
Solution. Let Wn, Wu, and Wi be respectively the maximum, mean and minimum
widths of tooth,
B
Wojmt
41,000
Way t4 Wha (given); Way POEM 1-2 try
Flux density at any section of tooth Br= fux/touth
fiux/tooth
net iron length x width of tooth
cs The flux density at any section of the tooth is inversely proportional to the tooth
width,
Let By, By and Bry be respectively the minimum, mean and maximum flux densities,
“ Bu BagWu/ Wem 1'9x1'2/1'4=1°63 Wh/m* ; By=1°9 Wh/m*
and Big™ BigP ty/ Wig 1'9X1'2=2°28 Wo/m*.
‘The B-at curve is plotted in Fig. 4°18. From this curve the values of mmf per meire
for flux densities By=1°63 Wh/m', Biy=1 9 W/m! and Biy=2'28 Wh/m! are respectively :
at, = 4500 A, aty= 17000 A, ‘@fy= 100,500 A
Applying simpeon’s rule, mean value of mmf per metre
stan (See Eqn. 448)
= 4500 +4 17000+ 100,500 | oq Alm.
A Total mmf required tars X = 28833 X 30 x 10-*=865 A 136 FLECTRICAL MACHINE DESIGN
Example 4-6, The armature of a d.c. machine has a diameter of 0'2m, 20 parallel
slota euch Gown wore ind 2 vam deep ; the core in O15 m long with I duct 10 mm wide, and
the air aay mm ron. tavitariin.om the slampinga sa 10% of th thickness,
Uf the maximum flu devsity wider the pole ia 0'9 Whim, determine the mif required (0
ovrrcome the reluctance of tip and (th. Catir's coveffiient for the clots is 0275 and for the
alote is 0275 and for th: duc! 0.39 The maynetization curve for the ivon ia as follows
a | Ms i | ve 20 | 2 23
whine
aH ] two | sooo | esa |. 19000 | 3000 | 12000
Aim |
Solution, Slot pitch at the gap surface 1 De? meal 67 mm.
Since the insulation on laminations is 10% of the thickness of the laminations, the
stacking factor Kes 0°,
21-67
Gap contraction factor for slots Kee= 1-082
150
Gap contraction factor for ducts Kel= aay FTN IO
=1027 MaGNetIC CIRCUITS 137
Gap contraction factor Kym 082 x 102721111
‘Mnf required for air gap ATy=800,000'Ky By ly
=BC0,0010 x 111 X0°9 x 8 X 10-2400 A.
__ The B-'at’ curve for the ferromagnetic material used is plotted in Fig. 4.18 The mmf
required for the teeth is calculated by using simpson's rule and Mix method.
The flux over one slot pitch passes through one tooth.
Flux ovr a slot pitch Or= Ry ye L=0'9 X 21°67 » 10-* X 0 15=2°925 x 10-* Wh.
Simpson's Rale Top of Tooth, Slot pitch y=21'67 mm.
Tooth width Wrad67—6=15'67 mm.
Net iron length Lam Kd L— nae) =0°0H0' 15—1 X 1X 107*)—=0'1341 m.
. 2925x1078 ‘a0 wr
Mlux density Baroy,“ o-1541x 1567 10-7! 99 Wo/mt
From Fig. 4°18 ‘at,=1790 Alm
Middle of Tooth. Diameter 0025-0175 m.
Slot pitch 2X07 ma18-96 mm.
Tooth width Wig=18'96—6= 12°96 mm.
Flux density Brym Bay» Wei Wegm 99 x FOE
= 1681 Whimt
From Fig. 4°18 at, = 4000 A/m
Root of Tooth. Diameter = 02-2 x 0025=0'15 m.
Slot piteh sa 2O 1625 mm.
Tooth width 16°25—6= 10°25 mm.
Flux density Big= Bu, X Wy[ Weg= 199 X 15°67/10°25— 27125 Wh/m*,
From Fig. 418 uf, =42500
By simpeon’s rule, eth tdaty aly 1790-44 004-4500
=10050 A/m.
Monf required by teeth AT = ahason X dom 10500 X25 x 10-3263 A.
Total mmf required for gap and teeth==ATy+ATi=2400+263=—=2663 A.
Diameter at } height from root of tooth=D--2x2/3 ds
=0°2—2 2/3 X0'025=90'1667 m.
1667
Slot pitch m=18°06 mm.
Tooth width Woyls=18'06—6= 12°06 mm.
Flux density Buly= Be x Woy/ Weyia= 1°39 x 1567/1206
=1°809 Wh/m*
From Fig. 4°18 atyy=11000 A/m
Manf required for teeth AT:=11000 x25 x 10-*=275 A
‘Total mmf required for gap and tecth=2400+275=2675 A.
Example 47. Calculate the apparent flux density at a particular section of a tooth
from the following data :
Tooth width= 18 mm ; slot width=10 mm gro care Length=O 32 m ; _ mumber of
ventilating ducts=4, each 10 mm wide ; real fluz densiiy=3 2 W/m’ ; permeability
corresponding to real flu deneily==B1'4x 10-%.H/m (henry per metre); stacking factor =0°0,
Solation, We have magnetizing force or mmf per metre, H=3/ SUECTRICAL MACHINE DESION
Maat per metre correipoadi
ability w= S14 10 is ¢
22
Saxe!
Net iron length L1=0°9(0'32—4 x 10 x 10-8) =0'252 m.
to real Hux density Breat=2'2 Wb/m! and perme-
‘at = =70,063 Ajm.
Slot pitcn ye Wet We 12+10=22 mo
a Lye _ 09222,
From Eq. 4 48, Kix Tee Sige iy 28
From Eqn. 4.48, apparent flux density
Bory= Brio + 4X 107? at (Ka~1)
22+ 4 x 10-7 x 70,066(2'228—1)
=2°317 Whjmt.
Example 48, Delermine the apparent luz density in the teeth of ade. machine
sehen the real fur density is 215 Whjon-;aiot pitch £8 mm ;elot width 10 mm and the gross
core length 0°35 m. The number of ventilating ducts is 4, each 10 mm wide. The magnetiaing
force for a flux density of 215 Whjm? ix 55000 A/m. ‘The irom stacking factor is 0-9.
Solution, Net iron length La=bi(li~-neWe)=0°9(0°35—4x 10x 10-4)
1279 m
ridth Wi= ye - We=28—10=-18 mm
From Eqn. $49, Ky fine BAM aygs
2°15 Wb/m, the value of “at”
Tooth
Corresponding to Bre
Apparent flux density
Bore= Brat +48 x10" at” (Ka 1)
=:2'15 +4 x 10-7 55000(1°95—1)=—=2°2156 Wh/m*.
Example 49. The armature core of a d.c. machine has a gross length of 0°33
including 3 ducts each 10 mm wide, and the iron space factor ie 0'9. If the slot pitch at a
Particular eection te 25 mm and the slot wiith 14 mm, estimate the true flux dinsity and the
‘umf per metre for teeth at thie section correaponding to an apparent fluz density of 2-3 Wo/m',
The magnetizalion curve data for armature stampinga i
2 1 [os 19 20 | m1 | a2 Jos
‘Whim
3700 10,000 | now [? 41,000 | 70,000 | 109,000
1
Solution, The B ‘at’ curve is plotted in Fig. 4°19,
Net iren lergth Ly=0'910'33- 3 10x 10"
Tooth width
28.<2
A325. 29°18,
” xi
and real flux density Brest Barp—4 x 10"? ‘ai’ (Ke~1)
=23—4nx 10-1 ‘at (2°78—1)
m=2'3 -2'237 x 10"* ‘at’, MAGNETIC cRCUTS
139
‘This is the equation of a straight line. The line can be drawn by locating two points
on it. ‘The two points are located as under.
When ‘at'=0, Brst=2'3 Wh/m*. This is represented by point A.
When ‘at'=70000, Brasi=2'3—2'237X 10 "x 70,000=2°143 Wb/m*
sal T ata
23 Bregt? 23- 2:237x10 atreat
SSS SS
f
1
7
'
+8 i 1
‘Magnetisation q
curve 1
+6
y
v4 i
at
atreal= |
$8000~
121 1 >
ot 2 50 7-80-8000 «HO x10
at’
Alm
Fp
This is reprosented by point B. A straight line is drawn through 1 and B cutting
the maguetisation curve at 0. Point C is the operating point.
From point C,
Brest 2°17 Wh/m* and ‘at’ = 59000 A/m.
9ap=8 mm, slot pitch ot the
, Length of Vength of
cir goptrnface-25 s,s ph a he bottom of laoet tm, Sepik of lte60 mm, wih of
ample (104d. machina hath fllonng dat
machine =O"
slim 12 im.
number of ventilating ducts in armature
teidth of each ventilating duct
fas per pole
aloulate:
(a) the mmf required for air gop,
(0) the mmf required for teath neglecting the
sto fs,
ul the manf required for lath considering the
Don Bat sane for, ordinary et ais
ashen in Fig, 41, Foi the cave of
£9 ould
*s conaflcient.
ere afr dng Coe
=
=10 mm
278x107 WD.
Fig. 420. Details of d.c. machine of
‘Baample 440 (Ail mensions in cm) 140
calling this as Bimen.
SLECTRIGAL MACHINE DBSION
Solution,
Fig. 4°20 gives the details of the machine.
Gap contraction Jactor for slots :
slot width _ 1-2
gap length 08
Carter's co-efficient, froin Fig. 4°9 corresponding to 1°5, for open slots, Ku=0°21.
Kn=gap contraction factor for slots
112.
Ratio
Gap contraction factor for ducts:
Duct width=01 mm and number of ducis=5.
; vo duct width 10
* Ratio “Gap length ~@ 7! 2
From Fig. 4°9 corresponding to 125, for open slots, Kos is 0"18.
Gap contraction factor for ducts
036
jax ox 10 10
“2X 10261141
= 1026
Total gap contraction factor Kem Ku Kea-=
(a) mf required for gap
Assuming a rectangular distribution of flux over pole are.
<. Form factor Ky=¥.
flux per pole
pole arc Xlength of core
667 W/m?
<. Minf required for the air gap==800,000 KeBals
= 800,000 x 1141 x0°667X 8x 10-*=4870 A.
(0) Mnf required for teeth neglecting saturation
Net iron length Ki L—neWa)=0°9 (0'36—5 x 10 10-#)=m0'279 m,
Gap density at the centre of pole Be=
le are 03
No. of teeth in the pole are—
a > Re-enterant
é tops
30
x
2
* — —— --
Time —» himt/metre H—=
@ ®
Fig. 424, Re-eatrant Hysteresis}loops.
| 4°51 and 4°58 give fairl; results in the case of static machines like trans-
armen easton, and ae iccisnegnes in which the flux oscillates along a path that is
ractically fixed. However, the rotors of rotating electromagnetic machines are subjected to
juxes that change their direction on account of rotation. This gives rite to a phenomenon
aalled “rotating hysteresis”. Fig. 4°25 shows a typical case of the variations of hysteresis
oss at different values of flux density for rotating electrical machinery. ‘The rotational
aysteresis loss is greater at low magnetization than the corresponding hysteresis loss due
ic Held, while at high flux densities the rotating hysteresis loss actually
iecreasee beconsing quite low at very high values of ux densities, as ‘MLECTRICAL MACHINE DESIGN
‘There are additional eddy current lowes produced in the rotors of rotating electric
machinery on account of the fact that the
flux can‘“ut everywhere be confined to a
direction parallel to the plane of lamina-
tions. Some flux enters the rotor at the two t
ends and also the sides of the teeth through
the spacer provided for venation, Since &
this flux enters the iron in a i te
mal to the. plane of laminations, it gives rise i
Alternatir
fo tysteresie loss
Rotational
toan appreciable iron loss. It is hysteresis loss
to point out that in the case
armatures of d.c. machines, the variation of
flux density is a combination of *linear and
rotational magnetization, and therefore the
low under these conditions is uncertain.
(i) There are some practical factors . if —
which lead to increased iron loss in_ built up Flux density, 8
cores. The laminatizns may be shorted by Fig. 425. Rotational alternating hysteresis loss,
burns on the edges due to worn dies, by grinding, or by faulty insulation on rivets, clamping
bolts or the laminations themselves, thus giving increased eddy current loss.
_ Mechanical strains vet up by cold working or assembly of core affect the nature of the
material and thus increase the iron loss, sometimes as much as a factor of 2'5.
There are additional iron losses owing to distortion of flux produced by armature reac:
tion on load. This is because the iron losses are proportional to maximum value of flux
5
Wika
Specie iron oss
Flux density, Bry a
W/m?
@ Transformers (©) Rotating machines
-Fig. 4°26. Typical loss curves for transformers and rotating machines,
‘The iron loa mey be ‘masa effect ditingushed manner in which the ux vara
io osc. magneton may be ine, on Tottonal Licey agetinict ane ak Pa
= onthe mame Fotational magnetization means a roughly constant Bel whoes ‘MAQNETIC CIRCUITS 141
density and maximum valuc of flux density increases with load on account of distortion o
field form giving rise to increased losses. ‘The iron losses also increase on account of ageing
effect.
Due to the above mentioned reasons, it is impossible to calculate the iron lowes
in built up cores. In practice, loss curves which represent the specific iron loss as
function of maximum flux density are used. The loss curves are obtained by measurements
done on built up cores and thus take into account some of the factors listed above. Typical
Joss curves for transformers aud rotating machines are shown in Figs. 4'26 (a) and (b).
Lom curves for diffrent. grades of stampings manofactured by Sankey Presings
Division of M/s Guest Keen Williams is given in Figs. 4°27 to 4°30.
1-48 Qealty 2-41 Quality 3-42 Quality
Fig. 427, Los carves of Blectrical sheet steal (Non oriented) 0-35 sim thick, 148 ELECTRICAL MACHINE DESIGN
weg
Specitic tron toss —e
Q at
a a a a a
Om =
WO/m?
148 Quality 241 Qual 342 Quai
4—-Medium Resitance Sspecial Lohye the
Fig. 4:28, Loma curves of Blectrical aheet steel (Non oriented) 0-5 mm thi
+o
‘The loss curves are prepared by the
inanufacturers cf stampings on the basis
of laboratory tests done on prepared speci-
mens,
In fact, the rerults obtained in the
laboratory, by testing prepared specimens,
cannot be relied upor. as the losses in the
built up cores of actual machines are in-
variably more than the losses obtained
laboratory tests. In order to take addi-
tional iron losses into account the losses a8
obtained from iron loss curves must be 1
multiplied by a factor which is 1°4 to 1°6
for d.c. machines, 1" to-1°4 for induction
motors, 1°4 to 1°6 for salient pole synchro-
nous machines and 1°15,f0 1°25 for turbo... tyes
alternator. ;
‘As pointed out earlier the two lomes Fig "439" “tate Creed $—Grade 96 4 Grade 92
é.e, byteresis and eddycurrent losses may " ‘qrieated) 0°35 mm thick,
wikg
3
‘Specitic won toss = ‘MAGNETIC CIRCUITS 149
be taken as proportional to Bet. (See Eyn. 4°57). But even fur same flux density the
gpecific iron loss greater for teeth than core. At 50 Hz the specific iron loss may
PimaBnt Wikg
++(4°58)
25 |
{>
Bo 2.
83
2 4
Bo ,
ol |
0807 OSS
Bm—e
woJm?
1-Grade 48 2-Grade 51 3-Grade 56 4-Grade 62
Fig. 430. Loss curves of Electrical sheet steel (oriented) 0-33 mm thick
when ‘a’ is a constant whose values are given below :
Machine Port
AC. Core
Teeth
De. Core
Teeth
[At flux densities above 1°6 W/m the losses increase rapidly above the values calcula-
ted by Eqn. 4°58.
46, Pulsation Losses, In a rotating electrical machine, the armatures are slotted and
aa a result when the rotor moves (hare are rapid changes of local gap relurtanos. This chan
ie“ Gluctance gives rise to flux pulsations #.e. changes in the air gap flux a condition which
in aces additional losses called pulsation losses jn the teeth and pole faces. This effect is
Pomsiderably aggravated if the length of air gap is small as compared with slot openings. Slotting
ao produces harmonic fields which cause high frequency lostes near the gap surface. These
Tomes may sometimes be considerable (as in the case of induction motors) although they arg
difficult to calculate, 130 ELECTRICAL MACHDIE DESIGN
Example 411, Calculate the specific iron loss in a apecimen of alloy steal fur a maxi
swum flu density on 3:2 Wh/m* and a frequency of 50 Hz, using O'S mm thick sheets. The
resistivity of alloy steel is 0°3% 10-* Om. The density ia 78x 108 *. Hysteresis lose in
tach oyele 6 00°) m’- ve ‘alms
Solution, From Eqn. 4°53, eddy current loss
pL FEE gine = SLE Wie
m? x 508 X 1°28 (0'5 x 10-9)",
FIX 1OX TEX 107 0°83 Wike
Hysteresis loss pr PX S0 = 2'564 Wiig at 50 He.
Tot. Iron loss Pe=2'564-+0°633—63°2 Wiig.
le 4:12. A specimen of cold rolled grain oriented #'3 mm thick stampings has
a resistiolty of O'Sy10-* Om. The published hysteresis loop ie easentially rectangular in forms,
with a eovrcive force of 1- Alm forall alues of peak fius denatice upto 1°6 Wola. The
manufacturer quotra the iron lose in the material as 12 W/kg, with a sinusoidal flue density 10
Wom? (peak) at 100 Hz. Caloulate the loss in the material from ite properties and compare
it with the quoted value. The mass density is 7650 kg/m,
Solution, The material ix subjected to a peak flux density of 1Wb/m* and therefore
the value of co-ercive force is 12 A/m for all values of fiux densities to which the material is
subjected. The hysteresis loop is thus a rectangle with sides 2X12=24 A/m and 2x1
=2Wb/m*.
Loss per cycle=area of hysteresis loop==24 x 2=48 W-s/m*.
Hypteresis loss at 100 Hz,
48 x10.
7650
° From Eqn, 4°53 eddy current loss,
ye wif Batt*
EERE ime ELE Wee
at*X(100}* x (1°0)8 (03 X 10-2)" ga
Gx "5x 10x 7650 =0'987 Wiks
Total specific iron loss pi=0°627-+0°387=1°014 W/kg.
‘The calculated iron loss is smaller than the quoted,
—
‘=0°627 W/kg
Example 4-13, The hysteresis lone in a sample of iron was fowhd to be £9 Wig at a
seruency of f= 60 Hs and ata masinum fas density of Bo = 1W0/m?_(a) Coleultetherefiom
the ie in the expression, jele=m “ fem specific gravit; ir M
foie the Toes. per by at e288 He and a flux density Bux 1'8 Wo/m*, 'y af irom ia 76,
Selution. Specific gravity of iron=7°5,
o Density of iron=7500 lkg/m* (MAGNETIC CIRCUITS 13
Energy loss duc to hysteresis By!'? J/m*.
‘Hysteresis iron loss (power) /kg pac=n Bn?” X 7500 wi)
Itis given that the hysteresis iron loss ps at f-=50 We and Bus Wh/m? is 4°9 W/kg.
o 4-9=m x (50) x (1) 7500
or = 1307 1078,
Hence, hytercsis loss per kg at 25 Hz and a flux density of 1°8 Wb/m? is :
Pan {Bre 7X7500= 1307 X 10-# X25 x (1°8)F x 7500
= 6°66 Wikg.
Example 414. A laminated iron cylinder ie rotated in a magnetic fee
loss is 250 W at 600r p.m. and 312 W at 712 ¢.p.m. Find the loss if the la
twice aa thick, the induction density increased by 20% and the speed were 720 r-p.m.
Steinmetz co-efficient ax 1°6.
Solytion, The frequency of pulsations is proportional to rotational speed.
Let WV be the speed in r.p.m. and Vs be the volume of iron.
Rep Bmtt?
From Eqn. 4.53, eddy current loss Poel be ve
_RAIN'B att
6p
=KJN* whore f= AN and A is a constant since the frequency
of flux pulsations is directly proportional to the speed of
rotation.
%
From Eqn. 4'51, hysteresis loss
Pa= Kif Bab Ve= KaAN Lye!V = Kx
where Ke’ and Ka’ are constants if By and ¢ remain constant,
J. Total iron loss Pe=Ke'N*-4+K'N
Substituting the values of Ps and N in the above expression
500=K,' (480)*+Kx’ (480) oli)
312=K. (600)*-+Ka’ (600)
From relations (i) and (i)
E,'=0'00104, Ky’ =0 5425
‘The hysteresis loss increases in direct proportion frequency (which is proportional to
speed in this case), lux density raised to power 1°6 and is independent of the thickness of lami-
The eddy current loss on the other hand increases in proportion to flux density
squared, frequency squared and square of the thickness of laminations. y
‘+, Total loss Bn increased by 20%, thickness increased to twice the original value and
N=720 r-p.m.
P¢=0°00104 (720)* x (1°2)# X224-0°5425 x (720) x (1°2)!" "= 3622 W,
MAGNETIC LEAKAGE CALCULATIONS
47, Effects of Leakage Flux, Tt has already been stated that it is impossible to
confine all the flux to useful paths ; there being always a leakge flux. 152 ELECTRICAL MACHINE DESIGN
If the leakage flux alternates, it will induce voltage in any winding with which it links.
This is known as the leakage reactance voltage. The reactance corresponding ta, this voltage
plays an important part in the performance of a.c. machines,
In d.c. machines, the leakage flux passing in non-useful paths affects the field excitation
of the machines. ‘The excitation has to be increased to compensate for loss of ux, Although
the leakage flux is constant with time but during commutation it is reversed when the coil
currents are reversed, giving rise to reactance voltage. This reactance voltage opposes the
change in current and makes the commutation difficult.
‘The estimation of leakage flux is difficult owing to the complex geometry leakage paths.
imposible to get very accurate results as no amount of mathematics can define com-
plexity of Icakage flux. Normally the results obtained are checked against experimental
data.
48, Specific Permeance. The leakage flux can be assumed to consist of flux
tubes of lensth y and a constant width 8z along the effective depth or length J. of the field as
shown in Fig. 4°31 (a).
8x
() in two dimensions
Permeance of a tube 3A =p ;
(In this case only reluctance of the path through air has been considered and the reluc-
tance of flux path through iron path has been neglected).
‘+. Permeance of whole field +f L sn Ln fe = (4°59)
Specific permeance is defined as permeance per unit length or depth of field.
<. From Eqn, 4°59,
Specific permeance A= sepy (4°60)
If the lergth of tubes is constant over a height h as in Fig. 4°31 (6)
a.
do} .
Specie permeance rmme [ane (4°61)
a ‘MAGNETIC CIRCUITS 153
Eqn. 4°61 is based upon the assumption that the, mmf is constant aver all flux tubes
as the integration is carried out,
The above expressions for specific permcance have to be modified if whole of the
leakage flux does not link with all the turns.” The analysis for this case is done below :
Let T=total number of turns,
Je=current in cach conducter,
=effcctive flux; this fux is a hypothetical flux which is
assumed to link with all the 7 turns producing same number
of flux linkages as are produced by ‘actual arrangement of
winding and field
effective fux, ©
total mmf
__7o total flux linkages
(PNT) Total turns x total mmf 4°62)
The flux linkages of the actual arrangement can be found by dividing the flux into
infitesimal parts ds, each represented by a line or tube of force linking with T's turns.
Flux linkages of 7 turns=7's d®e
Total flux linkages of winding=[ 740.
Now A meffective permeance=
but, d0,=mmf producing this fux x permeance of infinitesimal part
de
=HTaxpol =.
xm
de
Total flux linkages Tex btaxpal . ena | tet = :
By denfition of effective permeance (Eqn. 4°62)
: a
jm total ux linkages pte my
total turns X total mmf TT)
Pe \' de |
il F)S wey
A Te \* de
