US865215A - Art of telephony. - Google Patents

Description

PATENTED S111?.l s, 1907.

D. M. THERRELL. ART 0F TELEPHONY. APPLioATIoN FILED JULY 2s. 1904.

6 SHEETS--SHEET 1.

150.865,215L PATENTED SEPT. 3, 1907.

D. M. THERRBLL.

ART 0F TELEPHONY.

APPLIOATION FILED JULYza. 1904.

e sHEBTswsHmT 2.

fvg 5 w W i ,m

No. 865,215. PATENTED SEPT. 3, 1907. D. M. THERRELL. ART 0E TELEPHONY.

APPLIUATION FILED JULY 2a. 1904.

6 SHEETS-SHEET a.

Ifile )a JJVVI'JYTB.

, mzwfms'm.

No. 865,215. PATENTED SEPT. 3, 1907.

l AD. M. THERRELL.

ART OP TELEPHONY.

APPLIOATION FILED JULYzs. 1904.

` BSHBETS-SHEET 4.

@PPLIOATION FILED JULY Z3. 1904.

I D. M. THERRELL. ART 0E TELEPHONY.

PATENTED SEPT. 3, 1907.

8 SHEETS-SHEET 5.

No. 865,215. PATENTED SEPT. s, 1907.

D. M. THERRELL.

ART 0F TELEPHONY.

APPLIUATIN FILED JULY231904.

6 SHEETS-SHEET @HMM-m UNITED STATE-s" PATENT oFFIGE.

ART on TELEPHONY.

A.To all whom it may-concernai- Be it known-that YI, Darum. MACLACHLIN THER- REriL, acitizen of the United Sta-tes,'residing a't Charles;

ton, in the county of Charleston and' State of South Carolina, have invented new and useful Improvements in the Art of Telephony, .of which the following is a specification. V

My invention consists in an improved method, or system, for the transmission of electrical energy by means of electrical Waves, andparticularly such Waves as are employed in .the electrical transmission of articulate speech' or sonorous sounds, and comprises various improvements to be hereinafter more 'particularly de scribed andclaimed I Y In the transmission of ars'irnple electrical wave over -an electrical conductor :of great length characterized by'high resistance, electrostatic and conductive leakages, the energy lost or dissipated is proportional ,to

, and expressible in' simple terms of the reactive constents of the system, and thereis -no'distortion of the wave form. Such, however,- is n`ot the case for a c omplex electrical wave `composed of a'pluralityof coeiristing waves' such as the waves involved in the-telephonie transmission of the .human voice, or other sonorous sounds,co mpdsed of a' orundamental and an ascending series of 'partials ,corresponding to the overtones, or characteristics of articulation and quality. [n the latter case, each 'constituent `wave of the series is differently andindependently affected," and hence as differently affected, by the electromagnetic and electrostatic constants of the system-though both conspire to discriminate iniavor of the lower, while retarding 4ol attenuatingthe upper .harmonics direcllyas the irequency, resulting in the loss o articulation-andthe natural characteristics of speech which are dependent upon the transmission of the upper harmonics, or overtonos. And while conceding that great losses are proporly chargeable to the dissipation of the main line'con- (luctors, owing to their electrical characteristics as ontlincd above, my researches of the subject, mathematical and experimental, have confirmed me in the opinof average and permanent value is to be obtained by a tuning or readjustmcnt o the line constants; but that' much may be gained by improving the methods and apparatus in use for the electrical transformation and transmission of the primary energy equivalents of 'he voice. That is, by reducing the impedance o. the priniary coil andv obtaining thereby larger primary curi rents: by maintaining a highermutnal inductance and specification of Letters Patent. Application liledll'uly 23.1904. Serial No. 217,846-

"the higher frequencies.

Patented sept. s, V19o?.V l

obtaining thereby a higher efficiency of transformation;

and by so adjusting the secondary system as to generate a larger line current ata higher effective line potential,

,a higher efiiciency of transmission may be effected. l;

In the art of telephony, a primary circuit' consists of a rcircuit including a variable resistance in rela'tion to a diaphragm, asource of electrical energy and the primarywinding bf an induction coil or transformer. The vibrations of the voice acting upon the diaphragm in `operative relation to the variable resistance, causes this resistance to' vary in Vcoilsonance therewith, thereby setting up in the primary circuits' urrent which follows the variations ofthe said resistnce, and therefore the -vibrations of the voice. This cI rrent being of the' variable or pulsating type is capable ci transformation by means of an induction coil or transformer, the primaryo which is included in circuitwith theordinary variableresistane and-source of energy, while the secondary is connected into the main line` ABy this means it is possible, even with the small current to which we are restricted by the transmitter, to so stepup the voltage impressed upon the main lineas to reachover relatively vast distances. But in saying this it must be admitted that such results speak far more for the sensitiveness of the receiver than for the elciency of transformation or transmission. For, though ordinarily we may have flowing in the primary circuit something like .3 of an ampere, thc maximum value of -the current impressed upon the main line is notrmore than .0001 of an ampere, for the lowest frcquency, with a far greatcr loss in transformation-for Right here has been the stumbling blockl in thc way of telephone engineers, up

to the prcscnttime.

\ In order to successfully effect the long distance transmission of specch, the prime rcquisite'mnst be an increased powcr or 'wait efficiency of thc system for each and cvcry frequcncy involved, and particularly for such ireiuencics s correspond to the oycrtoncaof speech 'characteristic-s, if the speech so transmitted is to besatisfactorily intelligible. 'The power factor of the system is first of all dependent upon the amount of energy available in the primary circuit.- 4Unfortunately, we are here restricted to a very small aniperage on account of the 12R loss in the variable resistance portion of the transmitter which causes the transmitter electrodes to heatuand thus directlyaffect the transmission to such an extent as toplacc the available current limit considerably below one ampere, as mentionedabove. This is not the only difficulty Long distance transmission, Ina-kes itneccssary to raise the line potential by transformation. Anything like efiicient transformation presupposes a maximum lmutual induction. This in turn calls for a high mag- ,netic inductance, which is only possible bythe use of x v Y mount reasons'why neither of these requisitesarc adinissible in'tele'phony. `lf we endeavor to increase the magnetic intensity of the coil by the use of an increased amount of iron in the core, we are met with a loss due to an oversaturation of the core. This results .from the direct-,current used in the primary circuit which saturates the core and leaves. no margin for the Variation of the lines 4of force due to the variations in the primary current following the vibrations'of the 'voice, and hence a ,loss in the secondary instead of a gain., In addition. there are hysteresis losses due to frequencies and magnetic density. Again, just proportion to the amount by which we increase the Y, mutual induction ofthe coil by'an increased efficiency Aof the magnetic field, we also increase the self-induction of the primary, and therefore itset'iective impedance, which means a proportional reduction 6l primary current, unless thevoltage bc raised. Even if Afhisbedone, the primary current. is only increased at the expenseof the power factor lor the wattless com- Apollent of the primary will be so great that the point is soon reached-where the heating ot' the transmitter electrodes makes a larger current impracticable, without havingimproved the efficiency of the system.

v This calls for a compromise and also throws light upon l what has been to many for years, the paradox ofothe art. I reicr to the constants and proportions of the standard long distance induction coil, as used bythe American Telephone and TelegraplrCompany. In this coil a minimum o iron is used and the coefficients of self and mutual induction are amazingly low. 4This of co'urse means low cfliciency, but experience has shown, as stated above, that'to attempt a further increase in the efiiciency of transformation .but invites otherl losses which outweighA any apparent gains. Hence the standard cigar coil, and .the attempt to force the problem by attacking the capacity losses upon'the main line.

I have discovered a method of maintaining va high efiiciency of transformation without cntailingnipon the system, either the losses due to oversaturation, hys teresis, primary -or secondary impedance. These discoveries, which are `the result of years of laborious researchcs,. experimental and mathematical, form the basis ofv my invention, which will be better understood by reference to the accompanying drawings which form a part of this specification.

Figure l., is a polar diagram 0i a standard telephonie transformer, oir/induction coil, having the maximum amount of core iron, mutual and 'self induction, allowable in practice, with secondary on short circuit. i Fig. 12.', is au enlarged portion of Fig. 1. with 8,000 ohms in secondary circuit, showing the method 'of neutralizing the primary self-induction b'y a capacity reactance. 55, Fig. 3'. is a setof curves (drawn to rectangular co-ordinat-es), showing the impedance of a primary cir'cuit of ...given constants'for different p. s., also the values of :the impedance for the same circuit wherein the sclf inductance is neutralized by capacities corresponding to the given p. p. s. Fig. 4. is a set of curves showing the values of current for the same circuit, and p. p. s. @shown in Fig. 3, together with the values of theicur- V rienttfvafter anentralization of the self-inductancc .by corresponding to' the given p. p. s.; also curvesuhowing the resultant secondary current values Hernals when two or tlircc such circuits tuned to different p.. p. s.

are connected inl multiple, primary and pecondary..

Figs. 5 and (i, are vector diagrams showing the method of graphically detcrmining the resultant values o f sec.- ondary current and phase relation. under the multiple ,conditions noted in' reference to Fig. 4. Fig. 7. isa diagram illustrating the method of vplacing a capacity in series with the primary coil and transmitter, for the purpose of tuning. Fig. 8. is a diagram illustrating the principle and method of using compensating tra-nsformera, in the mainline for the lpurpose of reducing secondary effective impedance'and increasing the effective line potential. Fig. 9. isa conventional .diagram of my system in theory as applied to local battery systems. Fig. 10. and 11. are vconventional diagrams of the same adapted to common battery or central energy v systems.

Inasmuchfas the principles underlying my invention involve a considerable` departure from the standard practices of the art, it will be necessary here to briefly discuss the laws governing the actions and reactions of the generaltransformer, their 'effects upon tiansflormationand the principles of resonance in rclation thereto, in order that the claims ot" this application may be placed in their true light and thc specifications ,rendered more intelligible.

Let us consider a current fiowing in a single turn o fwire. A A magnetic field is set up consisting of a definite amount of magnetic .fiux, or lines of force, forming closed curves around the given conductor. vThis flux increases and decreases` with increase and decrease of'.

current. Ii the permeability oi the mediumis constant, the magneticr flux is directly proportionalA to the current. I f there are, s, turns `of wire instead of onc, the flux` N, passes through cach turn and consequently there are sN-lines threading, or linked with'the circuit.

The quantity, 8N, may be termed the fluir-turns or inductance of thcrcircuit. v

-If the magnetic induction through any circuit be changed,v duetto any cause whatsoever, an clcctro motive force is developed in thc circuit. proportional to the late of change of the magnetic induction, as first shown experimentally by Faraday. 'lho chungo in the induction to which this inducedclectromolivc force is duc', may be produced by a change in thc con `rent flowing in the circuit itself, in lwhich cus-c thc. clectromotivc force, so induced, is dcpcndcut upon thc rate of change of the current, and is known as the clectromotive force of self induction'. V

Let us now consider two coils in juxtaposition, which we will designate as primary and secondary, cach with a given nuinbcr of turns or convolutionls; let there bc a current of given intensity flowing in the primary coil. 'llhis current in the primary if caused t1; vary, produces a magnetic induction which varies in consonance there. with. Now, this induction not only induces an electromotive force of self induction in the primary'itself,

but also induces an electromotive force in the secondary due to the changing lines of force of the primary and the relative position of the twocoils, which is pro portional to the rate of change of the induction in the primary. This is mutual induction.. This mutual induction may be defined as theratio of electromotive Vforce induced in one circhit to the time ratefof change of the curfew; in the other. producing it. 'It may also be expressed in terms of the indnctionthreading the secondary ,due to the current in the primary. being equal to the rate of change oi the number of lines of force linked with the secondary. The effect greatly increased by the use of iron in the magnetic circuit.. Upon these principles is lmsetltheopei. tion of the devices .used for changing an alternating or varying current from one potential to another. of higher or lower pressure-known as induction `coils or transformers.

1n order to determine thtl action between the primary and secondary circuits ot' a transformer through the medium ol' their common magnetic -field, we must ascertain the amount ot' induction due to the current flowing through the turns of the primary and Iembracing a magnetic circuit of known permeability. This may be readily calculated from the la'ws of the magnetic' circuit. The total magncticlux or induction N is equal t0 the magneto-motive forcedivided by the re luctance.` The secondary electromotive force, in-

duced .by the primary current is proportional to the rate at which the primary. current is changing. and is equal to this time ratef change multiplied by the coefficient of mutual induction.

Conventions used in this specitieation: n=frcquency or periodspersecond. (`=eapacity. l=prin1ary current. :secondaryurrentl El=primary e. m. f. L,- =primaryinductance. L2: secondary inductance. ;1\i='fmutual inductance. total flux of magnetic -field. R=ohmic resistance. Im=impedance. 2=primary phase angle. 92=sec ondary phase angle.- pfp. s.=periods per second. y=transformation ratio. Sbscripts 1 and 2 refer to primary-and secondary circuits respectively, through-` out this'specification.V l

Having briefly outlined the underlying principles of the transformer, we mayprocccd by'wa'y o further exposition, to construct the diagram of u telephonie,

high potential transformer, orrinduetion coil, such as is .instandard use by the American Telephone and Telegraph Compa-ny. Fig. 1. is a polaridiagram of such a c0il. The solid lines indicate secondary on short circuit, the broken lines the sume coill with 8000 ohms in secondary circuit. i y

The constants of the coil are taken as follows: Rl= .5 ohm. R2=85- ohms.' l.1= .`()2l henry. Lgf-21333 henry. M= .0529 henry. n=250 p.. p. s.

Each vector .is assumed to represent maximum values. Let us consider the secondary on short circuit, the primary current as l ampere, flowing through an average resistance,'including the transmitter of, say

i 1() ohms, represented by OA, and (1H, rcspectively.

This linerepresents the electromolive force induced in ilusecondary circuit. by the primary current When the secondary circuit is eloscd'a current ou's which acts induetively both upon the primary and secondary circuits. -This secondarycurrnt lags behind the impressed electmmoti\'e'iorce O. B. in the secondary. due

to self induction. by an angle H2, such that,

` I i l 27Z1LL2- Wc'may represent the secondary effective electro-` niotive force and current bythe vector 0. CfandO. D.

respectively, `wherein 2 1r i E2=R2I2 and I2= in'- I-Il RE 442771112): By the graphical construction for simple circuits', we know that the right triangle 0. C.V B. upon B, ns'an 'hypotenuse, 'represents the secondary electronnitive forces; O. C, representing that necessary to overcome -`resis'tance, =R2 I2, C. B. that nectssary to overcome the self-induction ='27InL,I,. i v

The relation between the primary and secondary circuits isentirely a mutual one. A cut-rent flowing in the secondary induces an electromot'\ t'prce in the primary, jjust the saine as a current Aflowing in the primary induces an clectromotlve force .in the second- ?ary. The electrorioti n.; orcoset up by the secondary current in theprimaryiistermed the baekelclttrof motive force, and is'ninety degrees behind the secondaryv current and equal to 27TnMI2. Therefore, in Fig..

1, we may represent the back electromotive force by the line O. F., 90 behind the'secondary current O. Having assumeda primary current', ,we h'ave as based thereupon, determined the secondary electromotive force, the secondary current, O. D., and the back i-lec- Y tromotive force O. F. 1t 'now remaihs to lird'what` impressed primary;'electrottiotive force isrequired to cause the primary current. to flow. Instead of having but two electromotive forces to overmmeyas would be the' case for a simple circuit, the primary electro motive force must in this case, not only ovcrcotmthe electromotive force of resistanctl and sell-induction in the primary, but also the back electromotire force In such a case-the necessary impressed elwtromotive oreeupouv the primary is O. J.` But, when the sccondary 'current is allowed to 'low las in the case under consideration,there must be a component oi the primary electromotive force to overcome the back elect romotive force due to the secondary current. This electromotive force must be equal and opposite to 01E. and is represented by the line 0. G. equal to -27T'nM I2. Having thus developed our diagram, the primary elrmtromotive force desired is easily` found, since it is the. geometrical sum of the three components, 0. H.,

1I. J. and 0. G.- `The resultant of O. II. and H. L

. ance by bringing the resultant e. m. if O. K.. more in phase with the primary current. wlierehythe primary power factor nigh unto'unity. `Non'` lei. us assume the that the total secondary resistance becomes 8000 ohms ondary e. m. f."rises from 31.45 volts to 80 volts, as

tbei'secondary is also reduced from about 67 to .about telfspbonic:transmission of speech; The diagram of the" 30 l gives O. 1,., and the resultant/oi O. J. and O. G. gives O. K., which represents the required primary impressed telectromotivc force, bcing in the case considcred,.about 24 volts, and leadngthe current by approximately 12".

From Fig:A l, it .is readily seen that since thev com-4 ponent of the primary e. 1n. f. 0. G..necessary to overcome the hack e. m. i. dueto the secondary. is in the dire'ction of the primary current O.l A., the effect of a current in the secondary is to divide the primary e. m. i'. into three components. which. apparently reduces the .self induction of the primary and increases its resist'- current is increased and more power thus obtained. To further illustrate this point, reference is again made to Fig. .1Q In the case'just considered wlrereiutl-.e secondaryofthe coil was assumed to'be-on short circuit` it. ,will be noticed that the e. mf. oiself-induction in the primary was apparently reduced-.bythe back e. m. f. o fthe secondary from H. `ilfto H. X and the resistance increased from O. H. to Xt, whilethe e. m. i. came quite into phase with the current. thus bringing the secondary'to have a line in circuit of suchlresistance I instead of 85 ohms. This is aboutthe equivalent of a 4twothousand mile circuit as ordinarily used for the mme induction coil under these conditions is indicated in dotfiinddash lines in Figs. l a'nd 2, with primeand second indices. The apparent result is that the secshown by 0. (".j while'tlie secondary current falls from 5.37 ampere to .01 ampere; the back' e. m. f. falls {rom- 30.78 voltsto .83 volt as shown by O. F. The lag in l', with an equal shift in the'plnise relation of the back e. In. The result. is shown-inthe position oi 0. G., 'with tb(` resultant of (l. J. and O. Gf., which is 0. K4., as the necessary impressed primary e. 1u. I. equal to about 33. volts. :ind leading the current by about 72 instead of 12, with a power factor proportionatel y red uced.'

The values of the various vectors j nst given are based upon tlie assuniption thatthe primary e. 1u. f. has been raised so as to maintain a power current ol l ampere in the primary, and that the power factor oi the-primary remained the same as when the secondary was on short circuiti In practice, we have a decreasing, or at best a constant primary e. in. f. to deal with. Underthese circumstances, let us consider the primary e. m. t. as 24 volts. Now, with vthe voltage remaining constant, the. effect of increasing thesrcondary resistance as aforesaid is as follows. AThe secomlary-currcnt is reduced pro portionally to the seond-.iry resistance, and swings nearly into phase with the secondary e. 1n. i.; the back,l e. n1. f. which is dependent upon thc value of the sec' vondary cuire-nt decreases with the secondary current andV is advanced by the same'phase angle. The component of the primary, necessary to overcome this back e. n1. f., is proportionately reduced in value and advaut-ed in phase. Now asthe back e. m. f. decreases in value and advances in phase angle, the primaryim- I i'ednuce increases in value, and the impressedp'rimary e. ni.' f. advances or leads over the primary current., by a like angle. ln the'casegiven, the primary impedance is increased from 24 ohms on short-circuitcd secondary to 33. ohms with the external secondary circuit as given, and what is more important the phase angle between the primary e. 1n. f. and tllchprimary current, is incrcascd from 12b in the tirst instance, to 72 in the latter.` With the given primary voltage the primary current is reduced from one ampere to .72 ampere,

which gives us for the secondary e. m. f. according to 75 the equation'59.8 volts instead `ol' the apparent- 80 volts. 'lhcenergy of the primarycircuitv is reduced from sponding reduction in the energy oi-the secondary cir cuitor main line. of Awhich alone energy is transferred from the primary to the. secondary of a transformer, is dependent uponV the value of the primary current and its power factor,

and the secondary voltage Vis 'directly proportional to the energy of the. magnetic field, it follows then that any reduction in the secondary or line cumenb, as a result of an.increase'd resistance or length ot'fige, causes a corresponding increase in the impedance of the primary, and therefore a reduction in the primary current,

which in turn operates to further reduce the secondary i or line current by reducing the secondary potential which is equal to 27IfnMI1. -Thus, itis seen that-,to the primary impedance and secondary resistance is due mostI oi the inefiiciency of telephonie induction coils or tiansiormers. and hysteresis losses(v it 'is impracticable to avail ourselves of a magnetic field ot suicient intensity to warrantfan eiicient transformation. On account oi second-ary resistance,and long distance telephonie induction coils may be 'said to operate on resistances 100 equivalent to open circuit-we are not allowed to take advantage ota back electromotive force which further 'reduces th'e transformation ratio. This is true to such. an extent that the transformer-referred to in Fig. l,

shows, by computation, a loss of about 93 percent be- 105 tweenthe primary andthe 'secondary current, even with the secondary on short circuit.

Having the probleinstated, considernow the proposition of Ycontrolling the primary self-induction, and

rendering it independent of the secondary or line current. I have found that this can be doncl bythe use of capacity in lthe primary' circuit, or preferablylnultiples thereof, Vwhereby the primary self-induction may be wholly or partially .neutralized iu the given blanches oi the' primary for any desired ireq'ucrnfy', 115

thus greatly increasing the primary current and thereby the secondary current in the main line. Asis well known, the reactions due to self-induction and electrostatiecapacity are diametrically opposed. lt thereforefollows that for any circuit containingboth seli- 20 induction andcapacity, it is possible to soadjust the values of each as to make the one neutralize the other; When this -is done, a current will flow through the circuit the same as if it'we're free from any .impedan whatsoeve'i. This isthe conditonoi resonance, ,and 12:3 is possible only when or anyperiodicjty Since the magnetic field, by means 80 0n account ol' primary self-induction 95 abovdi or below this the tivo ilalucs will not cancel, and 'Y perfect resonance will not oh in'. Ihave found, -however, ithat when a circuit is malleresoqiant'for any particular frequency, say 250.p.1p. s., the leiects are experienced hyall frequencies above 250, but in a 'gradually decreasing'value, as will be explained later,

in -this specification. 2

Itis proposed to'neutralize the impedance oi the primary circuit, by dividing the primary into two or more branched,l or multiple circuits and'inserting a sufficient capacity and' iriductance into each branch tir-multiple thereof to satisfy the' conditions of reso= Vnance for the frequencies to be. affected, whereby larger primaryV currents 'and' more power may be' obtained 'Neutralizing the self-induction of the prima-rydoes not. affectthe magnetic field of the 'coil or the mutual induction upomvrhih depends the secondary e. m. f.

v and current.- On the other hand, it enables us to obtain` larger currents through the primaries, to utilize more iron in the magnetic circuit and to transform more energy into the secondary. This will be hett'er understood by reference to Fig.. 2 which is an enlarged por- `tion of Fig. 1. From this diagram it .willbe seen that with the secondary on 'short circuitl the primary self ity reactance h included i the primary circuit. to

oppose the reactan'ce of'V self-induction' equal to vector J. Z., then which gives the same impedance for the primary as When'the secondarywas on short circuit; O. K.b`e comes (l. K.; the 'primary takes practically twice the,- nmount. of current as lefore and transforms an equivalcnt. amount-of 'energy into the secondary circuit, or

nmi'n' line, due to an'increased Vratio of transformation, und this as a result of onlypartial resonance.

l shall consider now, some numerical examples for the purpose oi shtnving theopemtion of the general rule. tonsidcr a primary circuit including an induction. or transmitter coil, ofthe iolliiwi'ng constantsz.

{1:.5 olnn. R1' cxt .ernal=9.5 ohrn, L,=.022 henry. i".,=l0 volts. From the curves oi Fig. 43, it, will he vscen that. this circuit. and coil, listed 'therein as circuit No. 4'. oi'ers an impedance of about 34 ohms to a currcutoi 2.5i) p. p. s.', and that for a frequency o 75h p. p. s., the impedance is over 103 ohms.

Fig'. 4 shows n series of curves, of current resulting from thef ixnpcdanccs in the circuits` given in Fig. 3 ef'irn-spondiizg to the circuits and .fonstants'given, mzdcr the impressed 1; iin-.flot l() volts.. rFlins in Fig. t.. circuit-l.. It is secu that, for un impedance oi CH ullltlS and :i-t'rt-qin'trt' ol'. n. :2.31). we get only .29 ampere, vwhile for Ttlmf p'. s.. we get. .(1117 unipr-rc. This is .l l upon ih: ussunilnun: ol secondary on liniof Stltlthihnls,

assunte' Inl the. case under consideration sum of the curnts -in the different coils. Fig. 4,- represcnts the resultant of lcoilsl and 2 in mul- 100 Vrents und their rela-tive phase angles. Refer or approaching open circuit: us introduce into the4 65 primary a capacity, C. The primary' impedance is then determinedhy the equation,

Assuming, n,=25llp. p. s., and that. 70

I l 2mm-Lio,

27ml: Y p then the impedance reduces to zero andthe current;- rises to the value ziyen by, I-.fs ,I l 75 this neutralization is approximately obtained by a capacity of 18 micrl farads and the primary current rises from .29 ,ampere 'to 1 ampere as a result of this capacity in the circuit. Referenceto Fig. 3 will also show 'that for a frequency 80f of 7-50 p. p. s., the curi-ent resulting from' the partial resonance Yamountlsto .108 ampere, against .097 'ampere for the same circuit without the capacity: thus indicating clearly the benefits o primary resonance for telephonie currents, even' where the circuit is attuncd, 85 orV syntonized -lo butone frequency. Now,we may divide the primary vcircuit into a plurality of circuits, and include` in each branch the primary ofaninrdilc-V l tion coil, together with the capacity necessary to tune it forA any given frequency, 'as illustrated in Fig.' 4,1 90 whercimthefcurrent l'values' for identical circuits-- I with R. and L. constants tlie same in each-are shown for differcntperiodicities, without capacity, :ind'with the' circuits tuned for the frequencies` oi 250 p.; p. s., 375p. p.'s. 'and.500 p.p. s. respectively. Figi/4 also 95 shows in dotted lines, curves representing the values of current in the secondary circuit, wherein the secondaries are connected in multiple, givingr the vectorv Curve 11",

tiple; curve b, the resultant of 1, 2 and'3 connected in multipleinto the secondary circuit which is considered under short circuihat a transformation ratio ot 1:1 neglecting losses. These curves are significant.

Figs. and tilluslratc the graphical method of determining the sum oi'thcsc currents, each vector of which is draven to u 'given scale jrepresentingihe curi lo y Fig. 5,' wherein the values nrc hase-d upon a frequency 11i] of 250 plp. s., it. will hc noted that as a. result ofcircuit l heing resonant for this frequency the current l1 has a.

value of l ampere, whereas circuits 2 und 3 represented hy currents l2 and la. owing to their very highitnpedancc'ior this frequency give very small valucsol' cur` 115 rent und of such relative phase relation as to hut. slightly increase the resultant current l". The fact is elicited,however. that fora frequency oi 250, only circuit No. l will live any material current iut-othe sccorNlary.aI1d-thut. though the coils :ire in multiple, tfhc 120 itnpcdanecs are so great. for this ncquency, und tht,I

phase relations are such that coil 1 is reinforced instead 'ot' heine: short circuited hy coils 2 und 2l. Now, :rs thc frequencies :u lvnncc the currents from the different coils come neuw-r und nearer intophuse und as n con- 125' sequence the resultant 'approaches the simple sunt ot'.

v the several currents as is illustrated by Fig. 6.- 'Fl-cse values may also be obtained analytically. For two coils in multiple the formula for thc resultant current is o where a, and a', represent. tlrcyalue of -tlie two currents and the phase anglebctwcerr them. For thc resultant. of threcor' more currents. find the resultant of thc first two t-lr'cu combirn` this resultant with thc third and so on for tlrc scri'cs.

Having illustrated the general thcory and umh-ity- Y ing principles upon which Vthc resonance iciiturcs'ot' my invention are based, I shall vnow gire. so tur as may bc necessary; the detail of method and apparatus l'or carrying this theory intocffect; In the art vof telephony,r the nretkjliairicul energy of the Voice is translated into electrical energy through the agency of a diaphragm in operative relation toa r'miable resistance medium infclosed circuit with ir. constant 20' The changes in the variable resistance ft'rllowing the yibrations offthe diaphragm creates a unidirectional ra- I riable current through' the closed circuit including the yprimary vof the induction coil. This primary. current being unidirectional.- itffollows that a condenser cannot be placed inseries'withthe primary, coil.. a necessary requisiterun'der the given conditions` and the circuit f heinga constant `potential `onenotlring mai-"he gained Llry placing a condenser in multiple with the primary ui the coil; vv To'transform the primary unidirectional cur- Y 30 rent into a'n alternating secondary current for the purtpose of utilizing the principle of resonance lin the. seo

ondary, then-transferringthis' current 4tio-line means of another transformer, involves t-helosses of both tra-nsformera together with the dielectric losses. of the con- 3 5 densers which may more than equalize the gains, consequent upon a proper tuning for the essential voice frequencies. These facts and the very complex nature of -the frequencies and phase relations of a voice-current i have apparently ledlto the belief, by telephonie au# oriti'es, that the principles of resonance cannot be suc; l lly applied to telephonie circuits except for the ofcounteracting thecapacity of the main linev.nrrtlthe'rehy reducing the attenuation factor of transmission. -Iv have disco'yered a method, however,

' 45 .whereby a capacity may be successfully placed in se- 7" ries ewith the of a transmitter induction coil', andtlie cireuittuned for any desired frequency, or fres duendes, .which will be better understood by reference toFigL?` Referring to this figure, it will be seen that 5'() we'have a variable resistance l., in serieswith a con-l 60 'energy, 'supplying airnidirectional current, can only cause a current to flow' through the brunch containing 'the variable resistance 1., the current through tlre other brunch being stopped by the condenser 6. This is the normal condition'of thecircuit,

potentialsource of electrical energy und a transformer.'

Consider, now, the result. of yarying the resistance 1. G5 Under normal conditions about 98 percent o the'total resistance of the circuit formed by 1, 2,'3 and 4, lies in Y the variable resistance l'itself. Therefore 98 percent ot' the drop ot' potential in the circuit vwill be between -tlre points 3 :md 4; or in other words, there will he a difference ot' potential betweenthese points 'equiva-A lcnt to 9S percent of the potential of the battery, or source of ener-gy 2'. While no current flows, normally, through thcbranchof `the circuitcontaining tlre'fprimary coil 5.- and condenser l. the condenser is charged to the same potenti-.rl that exists bet-wecu and 4. lt follows that. it' tht` resistance l, bersuddenly lowered, as is thccase with a tclcphonit` transmitter, when a sound i `wave strikes the diaphragm thereof, the potential bet'u'ccntlre points :i :md 4. isproport ioriat-ely lowered, and being7 now ot' a lower potential than the condenser 6, the corrdcuser'dsclrarges causingr a current toliow, as shown by thc arrows, through the point 4, the transmitter l the point 2' and the inductionl c oil 5, to the other side of the condenser'. an'dvtlris current notl only flows under a p0- tcntial equal to the maximum variation oi the resist- :urcerot' the system` liutsflows through a circuit whose irrrpcdancc has been neutralized or reduced by resoi nance to practically itsolrrnic equ'iyalent and at a time when the ltransmitterj resistanceis at`its minima, thus 0 enormouslyincreasing the available current inthe primary coil dwlrich in turn proportionately increases the voltitgc, or potential of the secondary circuit. When the transmitter returns to its normal position the potential of thecontlenser tl is again raised to equal the' poten- Atia'lbetrivecn 3 and 4, and as tirediaphragm passes hrouglrits normal position and to maxirnain the opposite direction, thereby raising the resistance of the circuit, there is a proportionally increased difference S potential across 3 and 4 and also-across the condenser 10.0 terminals causing it to take a maximum charg'epand to discharge again, upon a decrease in the'potential across i 3 andgll, due tp another reduction in the resistancel;

rrd. thus with each-excursion oi the diaphragm, to and the resistance of the transmitter lis alternately in- 105 creased Vand decreased, raising` oiloweiringthe difference of potential across the terminals of tle condenser .which being alternately charged and discharged, pro- .duces a current through theprimary coil 5 ltwill be noted that bytlris methodian absolutely alternating; llt) current is utilizedin the primary coil, wliiclrvaries from-a positive maxima thrpughzero to a negative maxima, thus insuring better transformer ecency and better .definition -of wave form, as a result of permitting the use oi a greater amount of iron in the core of the 1 15 primary coil, and tlrej/ ariation of v the magnetlcflux'in i the c ore of the coil niaximathmn'gh zero to; inaxiina again, instead of nierelytreducing'tbe valueof a magnetic flux'of constant polarity frdmmaxima to' a certain art From the' circuits traced in Fig.- 7, itis evident-that.'

Apoints 34 and 4 may be divided into a'plirrality of cir-'KHA cuits, each containing th'e primary of an ,induction .coil

with the necessary inductanceand capacity, requisite tomake each branch resonant iorthe desiredrequency; f By Winding the primaries of the induction coils with an increased number of turns,'the transformer efliciency may not only be increased but the in'cluctance of the percentage of itself.' This isa' distinct step'in the vis primary coil'itselfmaylargely'serve totune the circuit. i gihle as compared with 27T'nL. and the magnetic' It is pr.thabli'advisahle. however, to insert separate inductames in the circuitsso as to use theA largest .amount inductance in the primary circuit that is practican.

pensive.

As hasheeu stated in the foregoing. a circuitjs tuned sary for resonance, when ih e othertizo are lgiven. For

tz-'trete he ohtained'; in turing orrresonance. only hy the' thereforishere igiven. which is:

As inductatlce" is afactor whichfhtts` toghe determined, ordinarily, by computation, the` expression Where t f3 numlierof tums in the coil,

magnetic circuit, /1 :permeability of ironcore. In this connection. ifshould h'e stated that satisfactory results use ofV mica or air condensers'. as I have toundhy disappointing and" pt ainstaal-:ing` eiiperirnents. that;l paraffin condensersar Ahysteresis andlovv insulation. which varies within wide limits withchangesoftemperature. l`

A'Having' n ow. covered tlte'se\'eral fundarne tal ele' ments of my's \'stem.'entering intotheappiication of resonance principles tothe primary circuit for any desired frequency orffrequeucies. aud the method of pro# ducingthe same.4 l shall nonadvance to the next. step fin the' direction of ini-rcasingthe effective main line '.potentialso asttvgixe amaximum linefcurrent; As'is -well known. 'in 'order to effect the -long distance transmission'of a telephonit` transmitter current. it is neces.

sat-2v' to step up the low potential current. ofthe primary. But. there is a limit. in thus ohtainiugthe necessary main line'potential beyond which we'may not gro. and

vwherein tnuch is lost. to telephony. To throw light 'upon' this point. let. us hrief'ly consider the secondaryY elw-tromotive force in a transformer. As stated in the foregoing.- this electromotive force is equal tothe pritn'arjf e. m. f. multiplied hy the ratio of transformation.

leakage vet-'y lou. it therefore follows'that the coefiicient of self induction hein;r proportional to the square of the numltet' ot' turns. we mays v that.

. yehapproximatel This` leads us to consider the fictitious natunl of the secondariA e.m. t'. soohtaincd as affecting secondar current. By derivation, we ma) 4detera, Marti-Rr The denominatorin this fequation shows us..the limitatit ms of tmnsformaton foretfective secondary'potential,

vnetfijc corein Square c. m.lero`s-1 section. I mean length A v l teutlal of tujtce the original value. hutsmce We have tot to be relied upon. owing to dielectric i Furthermore. it not onlyshoujs usthat the-apparent secondary potential must .he dividedvhy the secondary impedance. hut that the-'primaryresistance isatactor transferred to the secondary.. and that the tnuist'or'mag tion ratio is also a factor in they divisor vwhich increases 'as the square.Y While the secondari' resistance increases directly.' To illustrate, consider-a transformer with `or made 4800. what. Wonidhe the result'?- t is adminv ted that. ywe should ohtain an apparent-secondary po'- From the equation. we would get. aA certain secondary Suppose the secondary turns,'l2. be doubled.

it `is. readily seen'by substitution, that in so far as sec-Y I ondary current 1s concerned,`we have lost. h v the operation. This is one reasonv vvhy telephonie t.rans.

secondary impedance in relation to the receiver inA the circuit. And, it. is weliknown-.hy telephone'e'ntoo les'

gineers that. by shunting out. thesecondary. `of `high potential coils, theetticiency of the receiver is raised by 50 percent.. or more. This again refers us to. and acca-.mts in part. fot-",.the low efficiency transmitter induction-coils, in general use on the long distance I lines of the American Telephone and Telegraph Coin- Non'. in my invcntion,.the csscntial effective transformation ratio is obtained h vtuning th'` primary cir- I jat the distant endof the circuit, similarly arranged,

' in serios.

"Moreover, this is done without interfering with the efficiency of the receiver. In addition to this, l have' diseoveredthat by placing the secondaries, of the plu-4 rality of primary cii'cuits'already referred to, in multiple, and relati igthem to compensating transformers, the terminal impedances of the'line may be reducedto practicallygzero andthe receiver efficiency irnproved by about 100'pereent.' This will he better understood by reference, to Figs; 8 and 9. -i j Referring 4now to Fig. '8, let ns consider the alternator 10, as the equivalent of the secondarysourcevof e. m. f. 10, in Fig.9,.and that4 10 is another generator the rest of Fig... 8 being identical, with Fig. Sand with similar indices. Alternator 10, is supposed toghe generating an e. in. f., whichgis considered at the moment when the `current is in a positive directionfollowing the ,arrows; 'llhetransformers Tmvv and-Tra; are taken( as identical, with a transformation ratio of 1 :1,

100 percent. efficiency, primary and secondary turns wound together on thefsame4 core, and connected as shown in the diagram;` Consider now the current flowing 'in the direction-fof the` arrows. Upon reaching the terminallZ of transformer Trl it passes through the. line coil, or primary, in adirectionaround the core, as shown, :reating a magneticfux through the ,core and an induced e'. inthe other, or 'secondary winding which Hows in the oppositedirection around Trl, which on., account of. the two secondaries bef' ing connected, places these two secondary e. m, fs, Now, it is evident that the current from generator 10, upon reachingv 14 has two paths; the one through the main circuit and the -generator-10 and the other through the two secondaries 15 and 16. lt is also obvious that the sec-ondary path Athrough windings 1.5 and 16, is impassable because this lpath contains an e. m. f.` of itsoij/n, due to mutual induction, opposed to the direction of the current. Y In fact, this second path has an e. m. f. twice as giat as the e; m. f. in coil 13, and itself produces a current, which is placed in multiplo with 'the current in the line primary. In addition to this, it is evident that secondary'coils 15 and 16 being in series and of -a potential twice as great as that in the main, a current is also forced back through the primary coil 13 and the generator 10, and the potential of the generatorreduccd in the main line byhalf its former value. Reversing the diagram and considering 10 as the generator and 10 as the other terminal, we shall also reverse the statusv of the system, and the potential through the circuit between 1l. and 12 and through generator 10 will be twice that given by generator 'lO/ These results are of course predicated upon ideal relations and efficiencies, and in actual practice would be modified by the losses in the transformers and the relative values of transfornrer and generator constants. Thus it is seen that bythe useof compensating transformers, we. muy raise or lower the lino voltage,

or by properly adjusting the transformers, or their constants of transformation, we may maintain thel original potential of the'source while reducing its internal impedance to incoming currents, as a result of the shunt to incoming currents through coils 15 and' 1G. lt 70 shouldlbe observed however, that incoming currents are not only shunted in multiple with the source l0, but are raised in potential to twice that. of genctutor 10, thus reducing the impedance of the terminal from two causes, and thereby iner'l'asing the current'throtfgh 75 the circuit.

` It will be `noted that l have shown and described two` compensating'tmnsformers in 'relation to the source 0f e. m. f., one oil each side thereof. Now, this' is not ds: sential, andin practice ono transformer' properly ad 80 justed as to constants and transformation ratio, so as to equal the line e. m. f., is found preferable, as shown at. '15., in Figs. 9 and 10. Both methods, however, are effective, and have advantages varying with eircumstances. This brings me -to .Fig. 9 which is a diagram 85 matic representation of alocal battery type of my system, in conventional symbols, which will be readily nn- 1 p (lerstood. From. this figure where similar references indicate similar parts at each terminal -of the line, it e will be seen that the principles disclosed and explained with reference to Figs. 7 and 8, are applied tothe opera:

tion of Fig. 9. Referring new to Fig. 9. specically, A

and B. are the two distant terminals of the line; 1. is a ytelephonie transmitter ofl any approved type; 2 isarf source of constant potential in circuit with` transmitter 95 1,. derived from which,` branching from 'points 3 and :j are'tw'o branch circuits, containing inductances L, and Ll, primary coils Sand 7, and capacities and 8., This is identical with Fig. 7 already explained, whereby pri- 'l "i to :.rig. s, which. ispiflentirijhiwith. the secondary or# minal system of Fig.",`9.;lh3,rfore, we' have a system' whereby the transform Y fficiency' is increasedby resonance percent. or m' .while at the same time, neutralizing the effective :impedance of the equivalent secondaries and thereby increasing the sensitiveness of the receiver 19, by 100 percent., making@ not gain over old methods o at least 200 percent.,as.verifed by my experiments. j 4

Figs. l0 and 11 are diagrammatic -representatidnspf 115 contra-l energy orcommon 'battery'types of my system, -with distant terminals A. and B., and corresponding central offices C; O., whic h .will be readily followedf from what has already been disclosed, and a knowledge of common battery ,systems in general. Like indices 120 1 refer to like or equivalent parts in Fig. 9; -Fig. 11, is a.` mod ifcation of Fig. 10,*showing the method of applying the principles of my invention to subscribersstations in a common batterysystem instead of at the central office, as under. some circumstances may liedesirable.

I do not in this application claim the art or method of telephony described, as that is claimed in my other application, companion hereto, entitled Art of telephony, filed August 2, 1905, Serial Number 272375, which application is a division hereof, pursuant to the requirement of the Patent Office, .so as to-'separate thel apparatus from the method claims.

'It willfbebbvfous that many changes and variations mi be f nsde in 'parting from thespirlt of my invention-.fA

Therefore, without lmitfing'myseli to the detailsV 'the detail of Amy'system Without deshopvn,-whs:t Iciaim and desire to secure by Letters Patent of t-lie UnitedStates, is

, transmitteri 1. In the art of t i1'e electrical'transmission` and reproduction of* und, means foi: -produclng resonance in the cuit, substantially as set forth.

of the electricaltransmisslon and repro- 2. In the art Y duction of sound; means for producing anddetermlning electrical resonance vin the` transmitter` circuit, substantially as set forth.

- 3..1n the artffof the electrical transmission and reproduction of sound, meansgfor producing electrical resonance whereby the current isA in 'the primary 4circuit or circuits. -increased in 4the said circuit or` circuits, substantially as set torth.

4. In the art ofthe' electricai'trsnsmission and repro- 'duction of sound,- means' tori producing electrical conso-- .nance in the primary and secondary lclrcults, substantially us' set f oith.

-' 5. In.the art' ot the 'electrical transmission and reproduction of sound, means for ,producing and determining eiectricalvconsonance i'n 'the transmitter circuits, whereby the current is increased, substantially as set' forth.

6. 'Inftbe artof the electrical transmission and repro! ductionof sound, means for attuning the transmitter cir- 7. In the art ot' the electrical transmission and reproduction of sound, means for electrically attl'ining the transmitter circuit for the essential ,harmonidfrequenciesvof the sound to'be transmitted, whereby the cur'rent is in creased in tiiefsaid transmitter cli-cuit, substantially as 8. In the art of the electrical transmission and repro-' duction of sound, means for electrically.,attuning" tl1e transmitter and line circuits for'tbe essential frequencies to be transmitted, substantially as set forth. 9.V in the art of tiie electrlcal transmission und repro- `duction of sound. means for electricallyV attuning the transmitte'and line circuits for the essential frequencies to be transmitted, whereby the current is increased in the said circuits, substantially as set roi-th. 10. In the art of the electrical transmision and reproduction of sound, means for reducing the terminal impedanceito incoming line currents; substantially as set-forth.- 11.' In the artof the electrical transmission u nd reproduction of sound, means for reducing the terminal impedance ot the line to incoming voice or sound currents,

wherebyfthellne current is increased, substantially as set 12. In the art ofthe electrical-transmission and reprov'duction of sound,`mesns,for generating alternatingcur-l l rents'in a' branch of the transmitter'clrcuit containing the primary coil, substantially as set iioiftli.A

13 In the art o! the-electrical transmission Vand reproduction of sound. means for generating alternating currents in circuits derived from t'he transmitter circuit and substantially as set forth.

14. In'the art of the electrical transmission and 'reprocontaining the primary colis of a plurality ot transformers,

duction of sound, the combination o f a transmitter, s 'g source of energy and. s transformgr in a circuit .sttuned for'theessentiai sound or voice frequencies andin operative relation to the main line and receivermsubstantiaily duction ot sound, the 'combination ot s transmitter, As

relation to'the main line and Arecelversand compensating transformers in the main line,l substantially as set forth,

In testimony specification in' the presence of two subscribing witnesses.

' D, MACLAUCHLIN Tiramisu.. Witnesses: M

-source of energy and vtransformers in circuits sttuned for the essential sound .or voice 'frequencies and in operativeA whereof 1 have signed my naine to this'

Images