US3164676A

US3164676A - Stereophonic system employing audio matrixing

Info

Publication number: US3164676A
Application number: US95394A
Authority: US
Inventors: David H Brunner
Original assignee: Philco Ford Corp
Current assignee: Space Systems Loral LLC
Priority date: 1961-03-13
Filing date: 1961-03-13
Publication date: 1965-01-05
Anticipated expiration: 1982-01-05

Description

Jan. 5, 1965 H. BRUNNER STEREOPHONIC SYSTEM EMPLOYING AUDIO MATRIXING Filed MarOh l5, 1961 2 Sheets-Sheet i Jan. 5, 1965 D. H. BRUNNER 3,164,676

sTEREoPx-:ONIC SYSTEM EMPLoYNG AUDIO MATRIXING Filed March 13, 1961 2 Sheets-Sheet 2 /0 /2 AMFL /f/IR fa werf PMMA,

ff/G. 5A.

United StatesPatent O 3,164,676 SYSTEM EMPLOYING AUDIO MATRIXING David H. Brunner, bAbington Township, Montgomery County, Pa., assigner, by mesne assignments, to Philco Corporation, Phiiadelphia, Pa., a corporation of Delaware Filed Mar. 13, 1961, Ser. No. 95,394

12 Claims. (Cl.l 179-1) STEREOPHONIC It is customary tol refer to the two stereophonic program signals as'the A signal and the B signal .and this vterminology will be employed herein. The

`tgwo channelsystem mentioned above produces excellent It has the additional advantage stereophonic effects. v that an existing high; fidelity monauralsystem may b e converted to a high fidelity stereophonic system by ex-V actly duplicating all of the parts of the monaural system `except the source of program signals. However Vsuch a system has several drawbacks. The necessity of having v two'high-powered,V high-fidelity amplifiers and two wide 3,154,676 Patented Jan. 5, 1965 lCe'Y wave in a wide band speaker unit. The difference signal is amplified 'in a circuit which may have a more restricted bandwidth than the one employed in the sum channel. The -acoustic wave representing the difference signal is radiated by an electro-acoustic transducer or speaker unit having a different orientation from the one employed in the sum channel. Again this second speaker unit may be a limited range speaker un-it. The present invention relies on acoustic matrixing of the sum and difference acoustic wave signals to produce the desired Y stereophonic effect.V

For a better understanding of the present inventio together with other and further objects thereof, reference should now be made to the following detailed description which is to be read in conjunction with the accompanying d-rawings in which: FIG. 1 `is a block diagram of onepreferred embodi- ,ment of the invention;

FIG. 1A is a schematic diagram of the matrix circuit of FIG. l; Y

j FIG. 2 is a vector diagram which explains the operation of the system of FIG. l;

. FIG. `3 is a block diagram of another preferred embodiment of the present invention;

FIG. 3A is an alternative loud speaker arrangement which may beemployed in the system of FIG. 3; and

i FIG. 4 is a viewpartially intsection ofV an enclosed range speaker units makes the system expensive. Also the two wide'rangespeaker units 4arel not vreadily accommodated in small rooms.

Several i modified stereophonic `reproducing "systems have been proposedin` an attempt to overcome one or `more of the disadvantages of .the stereophonic reproducspeaker' vassembly which may be employed in certain embodiments ofthe present invention,

Turning nowr to FIG. ,1, the source of two stereo- `phonically related program signals is represented by ,block 10.' This source may be a stereophonic phonograph' pickup, a'stereophonic tape reproducer, a stereo- 'phonic radio receiver or the like. VFor convenience, the

program signal a-t the output connection 12 of source 10 will bereferred to as the A program signal While the signal at connection 14 will be called the B program signal. Both the A and B `program signals ing system employing two identical channels. However 4 none of the systems proposed heretofore have fully met the need for a compact, relatively-low .cost stereophonlc reproduction system which iseither `self` contained or` constitutes an addition to existing monaural equipment.

. i Therefore it is an object of the present invention' to provide` a stereophonic reproduction system which p roi vides excellent stereophonic reproduction with .a mmi- 4 mum yof expensive components.

Another object object of the present invention-is tof provide a stereophonic reproduction system which does not require physically spaced speaker systems.

An additional objectof` the present invention is to provide a stereophonic reproduction system in: which the if' apparent spacing between speaker systems may be made greater than actual spacing. i

A further object of the present invention is to provide a stereophonic reproduction system in which the spacing between the apparent -or-effective sources of vacoustic waves maybe varied electronically..

Another object of the invention lstocprovide `animproved stereophonic reproduction system which provides excellent stereoph-onic reproduction withv only one full range iampliiier and speaker )unit and one or "more amplitiers and speaker units of limited frequency range.

Still another` object of the present invention is toprovide means for converting existing `monaural Vequipment to a stereophonic reproduction system without modiication of 4existing equipment. l

vention are achieved by matrixin'g thetwostereophonic program signals-at a low level to produce electrical sum and difference signals. The', suln'rv signalisv amplified in a `wide band amplier andV converted toA an acoustic l i In general these and otherobjects ofthe vpresent in- A 'difference signal (f1-B). -this operation are Well known in the art, but for 4the sake Itions-of audio frequency transformers. Output lead 22 of matrix circuit 16 is coupled to the .input `of an amplifier 26. The bandwidth and powerhandling capabilities of amplifier 26 must meetthe standards established forthe reproducing systemsincegit handles all are'wide range audio frequency signals. For example, in a high-fidelity stereophonic system thej A and B program signals may `from time to time include componentsihaving frequencies in the range from 20 cycles to 20,000 cycles. It is to bef'understood that, in general, the A and B program signals will have different amplitudes and different amplitude vversus time waveforms. y l A The A and Bfprogram signals are supplied to a matrix circuit 16 which combines the Aand B program. signals to producea sum signal (A+B) and a Matrix circuitsfor performing `of completeness one simple transformer matrix circuit is shown in FIG. 1A. This matrixV circuit comprises two

laudio frequencytransformers

18 and 20, each having a primary winding and two independent secondary windings. The A program signal is supplied to the primary winding'o'f transformer 18 and the B program signal is 4supplied to vthe primary winding of transformer 20. "One secondary winding" oftransformer 18 and one secondary Y :winding of transformer 20 are connected in series in the v'proper polarity toprovide on output lead 22 the (A +B) signal. Thexremaining secondaryjwindings are connected in series to provide on outputv lead r24the (A -B signal. V` It should be understood that in high'delity systems it is `usually preferable to substitute phase splitter ampliiersand resistor-capacitor coupling for the transformers `18 and Z0 because of the` well-known bandwidth limitan d ofthe A program signal components and all of the B program signal components. The output of amplifier 26 is connected to a wide range electroacoustic transducer or` loud speaker system 28 hy way of a pad 29. YPad 29 may be a conventional L-'pad or Tpad.

Gutputconnection 24 of matrix circuit 16 is coupled to the input of a second 1amplifier 30. Amplitier 30 may have a more restricted Abandwidth and alower' power handling capacity than amplier 26 without reducing the over-all quality of the system performance. As a typical example, ampliiier 26y may have a passband of from 300 to 10,000 cycles'f fAmpliiier 30 provides oppositely phased tsignals at outputs 32 and 34,' respectively. These signals.

may be provided by usual push-pull'circuitry in the output o'f amplifier 30. Gutput connection 32 is coupled to a lspeakerunit 36 by way of a pad 37 and output connection 34 is coupled to a speaker unit 33 by'way of pad 39.

. Pads 29, 37 'and 39 represent one conventional way of changing the relative amplitudes ofthe signals supplied to the three

speaker yunits

28, 36 and. Other-means of signal amplitude control may be substituted for padsV 29,

and 39. For example; suitable gain `control circuits may be included in ampliersld and 30. The speaker units .36 and 3S are so oriented that all three

speaker units

28, 36 and 38 projectlsound into a common volume of 4space centered at O fronrdifterent directions'. It is desirable but not strictly necessary that the directive axes of all three speakers intersect at 'the same point in the space..

For reasons which will Vappear presently, it isV desirable rthat speaker 28 occupy a positionbetween speakers 35 `and 138. AIt V'-is" also desirable but not essential thatrthe respectively, Bc, ABL and BR are corresponding B com be reliected only in the constants K1 and K2. In the'vecspacing PL between

speakersl

28 and 36 be equal to the.. i

the directiveaxis of'- speaker 2S. It is also "desirableV but 4distancePR between the speakers 23 and'SS and fthat the 'angle 6I', between the linerpar'allel to the .directive 'axis 'olf` -spealerrZS and thedirectiveaxis of speaker bevequal. -to the e193, Vthe! anglehetweengthe second line 'parallel tov not essential that the diiference in elevation'of the three fromthe speakers to the listener V'location O.V

speakers 2R, and 33 be `small compared to the distance.

, `It lwill now Lbe sho'wn `that thelacoustic waves from the three

speaker units

28, 36 and l38 combine atpointOA to i provide a wave which'appears to have'an A component "whichoriginat'es' from an apparent source TLaat a point to the leftiof'speaker unit 2d and a B `component which Aoriginates from apparent sourc'eTRV at a pointl to the .rightofrspeaker unit 2S. y

It will also be shown that-by properly selectingthe orientation of Yspeaker units 3d and 38 and the relative gains of ampliers 26 andi?, theresultantFA and ,B component signals at point O are the A program signal component and the B program signal (component, respectively. Since the ,subjective feffect experienced'by a listener at point O depends only on the Vnetacorustic waves present at point O` and not in the manner in which they are generated, a listener at point()y f receives the impression that he is listening to a two `sterophonic system employing two vappropriately spaced wide vrange speaker units. Portions of the following analytical explanationof the operation of the system of FIG.y l are Yjpliedby source 10;

Theacoustic'wav'e signals nradiated the three

speaker units

28, 36 Vand may; beufexpressed as follows:

' 'This condition is niet if `fect, iit isfnow-only necessarytoshow that the vectors AN tor diagram of FIG. 2 (but not'in the following analytical explanation) it is assumed that K1V=K2=l-v Treating the A and B components separatelyefor the moment, the component AF whichA is the resultant A component for the three speaker units at point O taken parallel to the directive axis of speaker uni-t 28, may be expressed. as'follows:

` AFA +I `1A cos @4x54 cos 0R 4) ywhich may be rewritten asi Y l i i .i AF=A(1+K1 cos @L-fKz cosaR) (5) The component AX which is/the vresultant lA com ponent taken at right angles to the directive axisjof speaker unit Z3 may be expressed as follows;`

AX=K2A sin (2L-.tina sin 9g y. (6) Awhich may be rewritten as: :fr a* i i .AX=A(K1 sin @La-Kzlsinag) (7) ik The yectorlsum oi components AF and Axifsthe vector AN, VThe magnitude of the componentsAN for the three 4speaker units may be expressed as:

,Y iANl=\/AF2+A`X2? Y (S) substituting Equations `5` and 7 lin Equation? 18, it will be Yseen that Y' If the system is totp'roduceatrue stereophonicfelf'ect, Ythe amplitude' ratio IAM/{EN} must equalthe .amplitude "A/ B. -Si'nc'e the second-term under the radical'is the same for |ANI- as for {EN} 7 1' same as wouldk be produced by a stereophonic'reproducer- "f havingtwo spacedspeaker units which radiate solely the f Y l Y K1 CS COS 0R i .y Since vectors AN and BN be made to'have the necessary aplitudes to produce the desired stereophoni'crefandyrBN liavel the properfdirect'ion to produce thejstere- Y fi Speaker unit 28Sum Signal: A(al-(.130 Y (1.)

f speaker unirse, Left DifriKgAg-BL), (2)

Y ophonic effects. n FIG, 2 that Wanna-:fili

where 6g isfthe angle between the apparent line of direc- VVtion of,theracousticrvector AN ,andthe directive. axis of speaker' .unit as. Equation, 16: e. f-

Substituting. EquationsY *andA v7.. in

' occurs yfor a considerable area around point O.

where B is the angle between the apparent line of direction of the acoustic vector BN and the directive axis of speaker unit 28. Substituting Equations and 11 in Equation 19,

- (K1 sin @L+ K2 sin 0R) Thus the apparent directions of the vectors AN and BN are the same as that requiredfor normal sterophonic effect. t

From Equation 18 or 21 it can be shown that for coincide with the actual sources 36 and 3S. Alternatively,

speaker units

36 and 38 may be spaced any convenient `distance apart and the constants K1 and K2 varied byV `changing the relative gainsof ampliers 26 and 30 or the attenuation provided by pads 37 and 39 until the desired apparent separation is achieved between thev apparent sources TL and TR.

` yIt has been demonstrated experimentally that the acoustic matrixing of the signals from speaker units -corresponding to

speaker units

28, 36 and `38 ofvl-iIG.- l yIt can also be shown that the acoustic matrixing of the signals from the three speaker units-producesrthe desired stereo` phonic effect for asynunet-rical as well as symmetrical arrangements of the speaker units. However asymmetrical placement of the speaker units may require that the constants K1 and K2 have different values.V

It is wellknown that the low yfrequency acoustic wave componentsl of typical original source material are rela- 6 Y to have a lower power rating than that of amplifier 26.

Speaker units

36 and 38 are not required to reproduce low frequency signals. Therefore these speaker units may be made quite small.

It will be seen that if amplifier 26 and speaker unit 28 comprise the ampliiier and speaker unit of a monaural system, no modification of these units is required in order to convert this monaural system to a stereophonic system of the ltype illustrated in FIG. 1. 'Furthermore the cost of adding amplifier 30,

speaker units

36 and 38 and matrix f unit 16 will, in general, be far less than the cost of duplieating the wide band amplifier 26 and the wide range speaker unit 28.

The system shown in FIG. 3 is similar to the system shown in FIG. 1 except that a single push-pull electrostatic speaker unit 52 has been substituted for the -two

speaker units

36 and 38 of FIG. 1. Components in FIG. 3 corresponding to like components in =FIG. 1 are identified by the same reference numerals. In the embodiments of FIG. 3, only one output 32 from amplier 30 is required. The push-pull electrostatic speaker 52 generates the push-pull acoustic waves (A -B) and -(A-B). VIn the embodiment of FIG. 3, the directive axis of electrostatic speaker unit 52 is oriented at right angles to the directive axis of speaker unit 28. The difference acoustic wave (A -B) yfrom the left side of speaker uni-t 52 isreected from a suitable reiiecting surface 62 to the listener location O. Similarly the acoustic wave -(A-B) from the right hand side of -speaker unit 52 is reflected from a second reflecting surface 64 to the listener locationA O. The reilecting surfaces 612 and 64 of FIG. 3 may 'be the walls of fthe room in which the system shown in FIGQS is located. This reection from the

surfaces

62 and 64` t produces an apparent source of the (A-B) signal at tively nondirectional and that the low'frequency compo- Y nents of the A and iB` program signals typically have i nearly identical amplitudes. As a result the low frequency components .of the difference: signal (A -B will 'have verysmall amplitudes.` Since the low Vfrequency components contribute little to theover-all stereophonic effect and since there is a very little low frequency energy in the difference signal (A -B), the ampliiier 30 may have a lower cutoff frequency of the order of 300 cycles without p materially affecting fthe fidelity over-allsystem. f

Since at frequencies above about 10,000 cycles the Wavelength of acoustic energy is comparable to or less than the inter-ear spacing of'the average listener lthe location of a source emitting only high frequency signals becomes ambiguous. `For this reason it has been found that in many instances components of the difference signal above approximately 10,000 4cycles may be eliminated without adversely affecting the over-all performance of the system or lessening of vthe subjective stereophonic'eect. For these reasons, ampliiier 30 may have a restricted bandwidth, for example a lbandwidth of 300 to 10,000 cycles. Since the peak power in most -acoustic Waves occurs in the of reproduction of the very low frequency signals, the elimination of the loW` frequency signals from amplifier 30 permits amplifier point 52 and an apparent source of the (A-B) signal at point 52". lt will Lbe --seen that speaker unit 28 and apparent sources SZand 52" correspond to

speaker units

28, 36 and 38, respectively, of FIG. 1. y

`It has been demonstrated that the acoustic waves above 300 cycles are sufiicientlydirective 'so that appreciable acoustic matrixing of the sum and diierence signals radi ated by

speaker units

28 and 52 of lFIG. 3 occurs at listener location O. The system shown in FIG. 3 has the advantage .over the embodiment of FIG. 1 that all speaker units may 'be at a single location. For example, the

electrostatic speaker unit 52 may be incorporated in or placed upon the cabinet or enclosure for 'the wide range speaker unit normally employed in a monaural system. One possible arrangement of the speaker unit 52 within the same enclosure as speaker unit 28 is shown in` FIG. 4.

Enclosure is provided with ports 82 and`84 for the acoustic waveV difference signals (A-B) and -(A `-B).` Bales 86.and 88 may .be provided adjacent -

ports

82 and 84, respectively, for concentrating the acoustic waves in the desired direction. If enclosure 80 is placed in the Corner of a room,

bailes

86 and 88 may be augmented by or replaced by the adjacent Walls of the room. In setting upV the system of FIG 3, due consideration must be given to the reliective properties of the acoustic

wave reilecting surfacesy

62 and 64.

FIG. 3A shows an alternative speaker arrangement which may be employed in the embodimentofPlG. 3. In FIG. 3A, two electrodynamic' speakers 70 and 72 are substituted yfor the electrostatic speaker 52 of FIG. 3.

Speakers 70 and 72 are pointed in opposite directions so that they produce approximately the same distributienetV that various modiiicationsand other embodiments thereof i l ,c will occur to those skilled in the art within the scope of the invention. Accordingly I desire the `scope of my invention to be limited only by the appended clair-ms.

l A stereophonicsignal reproducing system comprising system comprising input means for providing at first and second outputs electrical signals representative of the instantaneous sum and the rinstantaneous difference, respectively, of two' stereophonically related program sigynals, a first 'amplifier `coupled to said first output, a first electroacousticwave transducer coupled to the output of said first' amplifier, said first transducer directing acoustic wave energy representative of said instantaneous sum of said two stereophonically -related'program signals in a selectedV path in response to the signal supplied by said rst amplifier, additional amplifier means coupled to said secon-d output, additional electroacoustic wave transducer means coupled tothe output of lsaid additional amplifier means, said additional ktransducer means producing two differently directed, oppositely phased acoustic waves representative of said instantaneousdifierence of said two s tereophonically related program signals in' response to the Vsignals supplied by said additional amplifier means, said last-mentioned two acoustic waves being directed differently than the acoustic energy lwave produced by'said first transducer.

2. A stereophonic signal reproducing system in accordance with'claim 1f whereinA said first transducer and said aiddirtinal transducer means are so oriented that the paths i of the acoustic wave energy from said first transducer ant said oppositely phased waves from said additional transducer means are initially directed toward a common location.

3. A fstereophonic signalreproducing system in accord'- ance with claiml wherein said first transducer and said `additional transducer means vare so oriented that the acoustic wave energy from said first transducer and'v said oppositely phased waves from said additional transducer means'are initially directed along non-intersecting paths. 4.' A stereophonic signal reproducing system in accord'-y ance with claim l wherein said rst transducer and said additional transducer means are located at substantially i VVthe same region in space and wherein said first-transducer-V and said additional transducer means are so oriented that acoustic wavee'n'erg'y from said first transducer and said :oppositely phased waves from said additional transducer means are initially 'directed along non-intersecting paths. 5. A stereophonic reproducing system comprising a transducer means each adapted to project acouslic wave energy in ak path centeredaboutfa,separate directive axis,

saidtransducer means directing acoustic wave energy so asito arrive yat a common location fromrtdiferent directions in response Vtofelectrical signals supplied thereto,

a first amplifierfmeans coupling said first output of saidV matrix ymeans to saidfirsttransducer, said first transducer producing an acoustic wave'representativefof said instan- 'p taneous sum of said two program signals ink response to the signal supplied `by said Vfirst ampliner means, addiional amplifier vmeans coupling said second output of said matrix means to said second vand third transducers,` said second and saidlthird transducers producing oppositely phased Vacoustic waves representative of said instantaneous Vditference of said two program-signals in response to the `signals Supplied by said additional amplifier means.

6. A stereo honic sicnalre roducinf s stem in accord- P e P s Y l ance with claim wherein-said first transducer is located intermediate rsaid second and third transducers.

V7."A'stereophonic signal reproducing system in accordc ance with claim 5 wherein said three transducers are sooriented that the directive axis of said first transducer makes equal angleswith the directive axes of said second and said third transducers.

8. A stereophonic signal reproducing system in accordance with claim 5 wherein the directive axes of the three transducers lie substantiaily in a common plane and wherein the directive axis of said firsttransducer bisects the angie formed by the directive axes of said second and third transducers; l l Y v l c 9, A stereophonic signal reproducing system comprising input means for providing at iirst and second outputs eicctrical signals representative of the instantaneous'sum and the instantaneous difference, respectively,V of two stereophonically related program signais, a first electroacoustic wave transducer arranged to direct acoustic wave energy in a selected path, means coupling said first output to said first electro-acoustic wave transducer, said first transducer producing `an acoustic wave `representative of said instantaneous sum of said two stereophonically related program signals in response to signals supplied at said i'irst output, additional electro-acoustic wave transducer means, additional means coupling said second output tosaid additional electro-acoustic wave transducer means, said additional transducer means producing in response to the signals supplied at said ysecond output two difierentiy directed, opposiely phased acoustic waves representative of-'said instantaneous diiierence'of said two program signals, said last-mentioned two Yacoustic waves being directed differently than the acoustic energy wave produced by said first transducer.

10. A stereophonic signal reproducing system comprising input means for providing at first and second outputs electrical signals representative of the instantaneous sum and the instantaneous difference, respectively, of two stereophonically related program signals', a first amplifier 'coupled'tosaid first output, a lfirst full range electroacoustic wave transducer coupled to the output of said first amplifier, said Vfirst transducer `directing in a `selected path,

stantaneous difference of said -two stereophonically related program signalsin response ',to the ksignal supplied by said additional amplifier means, said difieren'tly directed acoustic waves having afrequency' range substantially less than the full frequency range oi said two program signalswhich is to be reproducedbyV said system,rsaid last-mentioned two acoustic waves `being directed differently than the acoustic-energy wave produced by said first transducer.

1l. A stereophonic reproducing system comprising a matrix means adapted to receive first and second stereo- `phonically related program signals, said matrix means including means for providing at first and-second Voutputs electrical signals representative of; the instantaneous sum and instantaneous difference,v respectivel, of said two program signals,` first full range electroacoustic wave transducer means, and second Vand third limited range electroacoustic 'wave transducer" means each 4adapted to project acoustic wave energy ina path centeredabout va separate directive axis, saidtransducer means directing acoustic wave energy so as to' arrive at a common location from different directions in response to electrical signals supplied thereto, a first amplifier means coupling said first output of said matrix means to said first transducer, said first transducerproducing in response to the signal supplied by said first amplifier means an' acoustic wave 'representative of said instantaneous sum of said two program signals throughout substantially the full frequency range of said two programsignals which is to be reproduced by said system, additional amplier means coupling said second output of said matrix means to said second and third transducers, said second and said third transducers producing in response to the signals supplied by said additional amplier means oppositely phased acoustic waves representative of the instantaneous difference of said two program signals in a frequency range substantially less than the full frequency range of the two program signals which is to be reproduced by said system.

12. A stereophonic signal reproducing system comprising input means for providing at first and second outputs electrical signals representative of the instantaneous sum and the instantaneous diierence, respectively, of two stereophonically related program signals, a rst full range electroacoustic wave transducer arranged to direct acoustic wave energy in a selected path, means coupling said rst output to said rst electroacoustic wave transducer, said first transducer producing in response to signals supplied by said rst output an acoustic wave representative of the instantaneous sum of said two stereophonically related program signals throughout substantially the full range of resentative of said instantaneous difference of said two program signals in a frequency range substantially less than the full frequency range of the two program signals which is to be reproduced by said system, said last-mentioned two acoustic waves being directed differently than the acoustic energy Wave produced by said first transducer.

Y References Cited in the le of this patent UNITED STATES PATENTS Owen etal Feb. 5, 1963 Olson Sept. 24, 1963 OTHER REFERENCES Crowhurst: Acoustic Matrixing, Audio, November 1960, pp. 19-21, 78-81.

Claims

9. A STEREOPHONIC SIGNAL REPRODUCING SYSTEM COMPRISING INPUT MEANS FOR PROVIDING AT FIRST AND SECOND OUTPUTS ELECTRICAL SIGNALS REPRESENTATIVE OF THE INSTANTANEOUS SUM AND THE INSTANTANEOUS DIFFERENCE, RESPECTIVELY, OF TWO STEREOPHONICALLY RELATED PROGRAM SIGNALS, A FIRST ELECTROACOUSTIC WAVE TRANSDUCER ARRANGED TO DIRECT ACOUSTIC WAVE ENERGY IN A SELECTED PATH, MEANS COUPLING SAID FIRST OUTPUT TO SAID FIRST ELECTRO-ACOUSTIC WAVE TRANSDUCER, SAID FIRST TRANSDUCER PRODUCING AN ACOUSTIC WAVE REPRESENTATIVE OF SAID INSTANTANEOUS SUM OF SAID TWO STEREOPHONICALLY RELATED PROGRAM SIGNALS IN RESPONSE TO SIGNALS SUPPLIED AT SAID FIRST OUTPUT, ADDITIONAL ELECTRO-ACOUSTIC WAVE TRANSDUCER MEANS, ADDITIONAL MEANS COUPLING SAID SECOND OUTPUT TO SAID ADDITIONAL ELECTRO-ACOUSTIC WAVE TRANSDUCER MEANS, SAID ADDITIONAL TRANSDUCER MEANS PRODUCING IN