April l2, 1966 Filed March 2l, 1962 ZFIGJ.
EDWARD s 3,246,081
EXTENDED STEREOPHONI C SYSTEMS 2 Sheets-Sheet 1 Wl LLIAM C. ECM/mans BY anw( /di/ ATTORNEYS April l2, 1966 w. c. EDWARDS EXTENDED STEREOPHONIC SYSTEMS 2 Sheets-Sheet 2 Filed March 21, 1962 zozaOmE-L INVENTOR Wl Lum/l GEEN/A205 iwan" BYV ATTORNEYS United States Patent C) 3,246,081 EXTENDED STEREPHONIC SYSTEMS William C. Edwards, 220 Circie Drive, Plandome, NY. Filed Mar. 21, 1962, Ser. No. 181,268 16 Claims. (Cl. 179-1) The present invention relates generally to stereophonic systems, and more particularly to systems for expanding the apparent physical separation of sources of signal, whereby to provide the audible effect of sound derived from Widely separated transducers when in fact the transducers are closely spaced.
Stereophonic systems may take many forms, i.e., record/reproduce, radio broadcast, public address. It is a common feature of all such systems to provide two (or more) physically separated microphones. The microphones see any given sound source from different directions, which introduces a phase difference into the output bands derivable from the microphone, yand it is this phase difference which is in at least considerable degree responsible for the stereophonic effect. Additional audible effects derive from amplitude differences between corresponding frequencies in the separate channels.
The system output, whether derivable from a broadcast receiver, a recorder-reproducer, or loud speakers and amplifiers alone, ultimately includes at least two loud speakers desirably separated in space by adequate amounts, say six feet. If both the microphones and the speakers are adequately separated satisfactory stereo effects will ensue. If the microphones or the speakers are not separated adequately, the stereo effects are not noticeable. Nevertheless, the situation may exist that microphones or speakers cannot be adequate physically separated, so that inadequate phase separations occur as between the outputs of the stereo channels, as heard by the listener.
The system intermediate the microphones and the speakers may be of wide variety. For example, the microphone output may be recorded, and reproduced at sorne later time. In such case it may be desir-able to accentuate phase or amplitude differences during reproduction only. The present invention is operative to accentuate phase or amplitude differences, so long as two stereo or binaural channels are available.
It is a feature of the present invention to provide a phase and amplitude enhancement `system for stereo and binaural systems, which employs the phase differences of the sound sources.
It is a further feature of the present invention to provide stereo effect enhancement as between parallel channels carrying audio bands having some initial relative phase separation, by means of cross-feed among the channels, 'the' feed-back circuits being shift circuits.
lThe above and still further objects, features and advantages of the present invention will become apparent upon consideration of the following detailed description of one specific embodiment thereof, especially when taken in conjunction with the accompanying drawings, wherein:
FIGURE 1 is a block diagram of a system according to the invention;
i FIGURE 2 is a block diagram, including a specific cross feed-back circuit of a modification of the system of FIG- n URE 1;
FIGURE 3 is a schematic circuit diagram correspondrated, in general. The separate microphones 10, 11 provide signal forseparate signal channels 12, 13, which in turn drive separate loads, 14, 15. In a recorder-reproducerloads 14, 15 may be magnetic record heads. In a public address or reproducersystem loads 14, 15 may be speakers. In abroadcast system loads 14, 15 may be radio transmitters. In a reproducer system microphones 1f), 11 may be reproduce heads. The system of FIGURE 1 is, therefore, intended as a generalized representation of applicants novel concepts.
According to the present invention, the phase separation between the bands handled bychannels 12, 13 is inadequate to provide desired stereo effects. Cross negative feed-back circuits are accordingly provided, the channel 12 `supplying signal tochannel 13, and vice versa, via feed-back paths 16 and 17, each of which introduces a phase shift. The effect of the feed-back networks is to enhance the phase separation of the corresponding frequencies in the two channels. The source phase differences will cause a corresponding amplitude difference between the two channels, due to one side becoming less de-` generative and the other more degenerative. For closely spaced microphones, however, it may be assumed that identical bands are received by the microphones except for phase, and that the stereo enhancement is accomplished in terms of phase enhancement, so thatloads 14, 15 provide output as if the microphones 10, 11 were greatly separated instead of slightly separated. A further phase separation occurs, in any stereo system, in proceeding from loud speakers to the ears of the listener, which is a function of speaker separation. The effect of phase enhancement is then to increase the apparent separation between the speakers, which is particularly valuable in stereo consoles.
The system of FIGURE 2 parallels the system of FIG- URE 1, except for the cross-feed back system. In the. system of FIGURE 2, twoseries resistances 19, 20 eX- tend betweenchannels 12, 13, the junction ofresistances 19, 20 being connected to ground by capacitor 2-1, which is selected to provide phase shift. The use, of phase shift is optimum, but smaller values can be used.
FIGURE 3 is a schematic circuit diagram corresponding with FIGURE 1, as applied to a tape recorder. Themicrophones 30, 31 are coupled, respectively, to the bases of transistors 32, 33, which in turn are coupled in cascade withtransistors 34, 35.Transistors 32, 34 arer thus cascaded amplifying elements of a first stereo channel, andtransistors 33, 35 of a second stereo channel.
The emitters oftransistors 32, 33, 34, 35 are connected to acommon lead 36, which is connected to the positive side ofbattery 37, and to thecommon point 38 of twostero recording heads 39, 40.
The negative side ofbattery 37 is connected via lead 40 to the collectors oftransistors 32, 33, 34, 35 viaload resistances 42, 43, 44, 45.
The bases oftransistors 32, 33, 34, 35 respectivelyare connected back to lead 40, for signal coupling purposes, viaresistances 46, 47, 48, 49. In addition, a common capacitor shunted bias circuit 50 for all the bases is connected between lead 40 andlead 36.
A feed-back circuit exists between the collector ofy transistor 32, i.e. at point. 52, back through capacitor 53- and resistance 48 topoint 54. Thereby, a feed-back voltage appears acrossresistance 55, and is communicated to the base of transistor 32 via resistance 46. In an analogous manner, the potential at the collector of transistor 33, i.e. acrossresistance 43, is fed back throughcapacitor 56 andresistance 49 topoint 54, and is thence communicated to the base of transistor 33. However, any voltage developed acrossresistance 55 appears at the bases of both transistors 32 and 33, identically.
All feed-back voltage, from both channels, is thus developed acrossresistance 55, and from that resistance is transferred to the bases of b oth transistors 32 and 33.
Each phase shift circuit introduces its shift individually into its own channel, and also into the alternate channel. Insofar as feed-back intrachannel is concerned, phase shift occurs, but these shifts are identical in each channel so that no differential phase shift occurs. Insofar as interchannel feed-back occurs, on the other hand, the signals in the channels have an initial relative phase displacement, so that the effect of feed-back is to accentuate and increase the relative phase shift which exists in any event as between the channels, i.e. which would have existed were the channels independent.
It may be noted that the phase shift circuits employed are differentiating circuits. The circuits are designed to provide approximately 90 phase shift, i.e. the maximum available from the type of circuit employed.
Thetransistors 34, 35 operate solely as drivers for therecord heads 39 40. The voltage at the collectors oftransistors 34, 35, appearing acrossresistances 44, 45, is transferred via D.C. isolating capacitors 60, 61, to recordheads 39, 40, respectively. Resistors 62, 63 provide impedance matching between transistors and record heads.
In operation, then, two amplifyingchannels 12, 13, exist, each of which has internal feed-back involving a large (90) phase shift. Where feed-back is internal,
the gain equations of the system show that the total phase shift, between amplifier input an-d amplifier output, is less than the phase shift introduced by the feed-back network. The precise shift is a fun-ction of the percentage of signal which is fed back, i.e. the so-called factor of the network.
In addition to internal feed-back, interchannel feedback occurs. This does not occur identically, considering the entire two channel system as a closed loop servo, to intrachannel feed-back, and has been found to introduce a spreading or increase of the relative phases and amplitudes of signals in the two channels.
Input signals from microphones 3f), 31 are applied to the separate bases of transistors 32, 33, relative tocommon lead 36, which is connected to the external conductors 66, 67 of the cables leading frommicrophones 30, 31.
'The phase separation between signals of the same frequency as they appear atheads 39, 40 is greater than the phase separation for those signals, as seen by the microphones. Clearly, the same effect would occur were theheads 39, 40 speakers, and the audible stereo effect would be enhanced for inadequate space separation of the speakers, in such case, or, if the speakers were well space separated, adequate audible stereo effect would be provided even if the microphones were not sufficiently space separated.
In the system of FIGURE 4, the microphones 70, 71 (connected 180 out of phase) are coupled capacitively, bycapacitors 72, 73, to the bases oftransistors 74, 75. Shunt condensers 76, 77 provide bias or high frequency filtering. The emitters oftransistors 74, 75 are connected directly together, and to a common lead 7S, which is also directly connected to the junction of microphones 70, 71 and to the positive terminal ofbattery 80.
The negative side ofbattery 80 is connected to a common lead S1, from which extends loads 86, 87, 88, 89 :for the collectors oftransistors 74, 75 and of two further transistors 90, 91 connected in cascade with transistors ".74, 75, respectively.
The emitters of transistors 90 and 91 are connected back to positivecommon lead 78, via -variable bias resistances 92, 93. Base biases fortransistors 74, 75 are provided by means of resistances94, 9S which connect back to the collectors oftransistors 74, 75, whereas in the case of transistors 90 and 91bias resistances 96, 97 extend between the bases and negative common .lead 81.
lConnected across the collector load-s 88, 89. are coupl-ing and hi-gh frequency lay-pass capacitors 100, 101,
102, 103, which supply A.C. output .signal to record heads 104, 105. The common point of the latter is connected to positivecommon lead 78, which may be taken as apoint of reference potential.
To this point in the description, the system involves a pair of channels for conveying microphone signal to recording heads, the separate channels being per se conventional, and' involving common elements for purpose of convenience only.
In accordance with the present invention, a pair of equal resistances 110, 111 extend in `series between the collectors of thetransistors 74, 75. From thejunction point 113 ofthe resistances 110, 111 to lead 78, extends a capacitor 112. The valves of resi-stances 110, 111, and capacitor 112 are such as to introduce a 90 phase shift, approximately, with respect to collector, at thejunction point 113.
The circuit composed of resistances 110, 111 and capacitor 112 provides interchannel coupling, with phase shift. Assuming an initial phase difference, as between the channels, for any specific frequency, the cross feed or interchannel coupling, will accentuate the difference.
The concept of cross feeding between channels to accentuate an initial random .phase separation ena-bles the accentuation to be proportional to lthe initial separation. This'cannot be accomplished by designing the amplifiers to have different phase shifts.
The microphones vare connected 180 out of/phase in the systems of FIGURE 2 and FIGURE 4.
This is so that there will be more degeneration or less generation in each channel, depending on the phase difference (from 180) of the two signals.
This is not the case in the circuit of FIGUREl 3, Where the degeneration is not obtained by having each amplifier signal 180 out of phase with the other.
The circuit of FlGURE 3 is degenerative because the signals between channels are fed by the following stage,l
which is 180 out of phase due to the'action of the grounded emitter transistor.
The microphones in FIGURES 2 and 4 are connected to be 180 out of phase. The cross fed signals are somewhat degenerative event if the sound source is directly in front of each michophone. vThe phase shift networkbetween the two amplifiers, 12, 13, then causes one amplifier to be less degenerative and the other more degenerative if the sound source is shifted so that the space time relation of the two microphones are different inY respect to the sound sources.
While I have described and illustrated one specific embodiment of my invention, it will be clear that variations of the details of construction which'are specifically illustrated and described may be resorted to without departing from the true spirit and scope of the invention as defined Y plitude difference between phase separated signals of corresponding frequency in discrete similar stereophonic amplifier channels, comprising circuitry for cross-feeding said signals between channels, said circuitry including means for deriving respective approximately ninety degree phase shifted signals from the first-mentioned signals in each of said channels and for respectively applying the phase shifted signals so derived each to a channel other than that fromwhich derived, to enhance the initial phase separation of signals conveyed by said chann-els.
- 2. The system according toclaim 1 wherein said circuitry is negative feed-back circuitry common to said channels. l
3. The system according toclaim 1 wherein said circuitry is interstage circuitry common to said channels.l
4. In a stereophonic system, wherein an initial phase separation exists as between signals of identical frequency in separate stereophonic channels, said separation contributing to stereophonic effect, comprising a first ampliier for amplifying a first ste-reophonic signal, a second amplifier for amplifying a second stereophonic signal, said amplifiers each having input terminals and output terminals, a separate load connected to each of said output circuits, a separate transducer connected to each of said input circuits, phase shift circuits interconnecting said channels and having a common element, said phase shift circuits being symmetrical with respect to said common element, and means for connecting said common element commonly to signal input points of said channels, said phase shift circuitry being arranged and adapted to enhance the existing phase separation between identical frequencies existing at said separate transducers, respectively.
5. The combination according to claim 4 wherein said phase shift circuits are constituted of a pair of series resistances connected between corresponding signal bearing points of said channels, and a phase shift condenser connected between the junction of said resistances and a point of reference potential.
6. The combination according to claim 4 wherein said phase shift circuits are separate series circuits, each of said series circuits including a resistance and a condenser, said circuits extending from corresponding signal bearing points of said channels to a common resistance, said common resistance being in the input circuits of both said amplifiers.
7. In a stereophonic or binaural system a pair of substantially identical stereophonic signal amplifying channels, circuitry interconnecting similar signal bearing points of said channels to a point of reference potential, said circuitry including two paths each leading from one of said signal bearing points, each of said paths including an approximately 90 phase divergence enhancement circuits for signals of the same frequency existing in said channels, respectively, and for feedback circuitry extending from a point common to said paths tothe inputs 0f said channels, respectively.
8. In a stereophonic amplifier, first and second amplifiers for amplifying separate stereophonic signals, negative feedback phase shaft circuitry interconnecting said amplifiers, said phase shift circuitry being arranged and adapted to enhace existing phase differences between signals of the same frequency in said amplifiers.
9. The combination according to claim 8 wherein said negative feedback phase shift circuitry includes two substantially equal resistances cross-connecting said amplifiers, and a capacitor connected between the junction of said resistances and a point of reference potential.
10. In an audio system, sources of first and second audio bands having components of the same frequency differing in phase and amplitude, and means for enhancing the phase and amplitude differences in proportion to the existing differences comprising means for shifting the phases of said first and second audio bands by approximately ninety degrees to provide first and second phase shifted audio bands, and means for combining the second phase shifted audio band with the first audio band, and the first phase shifted audio band with the second audio band, the relative phases being selected to provide phase difference enhancement,
11. The comb-ination according toclaim 1 wherein said means for shifting includes negative feedback circuits.
12. The combination according to claim 10 wherein said means for shifting includes means for shifting phases by ninety degrees.
13. In a system for enhancing directional discrimination among signals having different directional factors, means for deriving one version of said signals, means for deriving a second version of said signals, said versions having phase and amplitude differences representing said directional factors, and cross channel feedback means for increasing said phase and amplitude differences proportionally to the first mentioned phase and amplitude differences.
14. In combination, a first microphone, a second microphone, said microphones receiving identical audio signals of different directional characteristics, a first amplifier channel connected to said first microphone, a second substantially identical amplifier channel connected to said second microphone, said microphones being connected to said channels, respectively, in opposite phases, means for feeding signal from a point of said second channel to a point of said first channel, means for feeding signal from a point of said first channel to a point of said'second channel, the points corresponding in the separate channels, and a load device for each of said channels, said means for feeding signal including identical phase shift circuits.
15. The combination according to claim 14 wherein said phase shift circuits provide at least approximately ninety degrees of phase shift.
16. In combination, a first transducer, a second transducer receiving identical signals of different direction characteristics, a first amplifier channel connected to said first transducer, a second substantially identical amplifier channel connected to said second transducer, said transducers being connected to said channels, respectively, in opposite phases, means for feeding signal from a point of said lsecond channel to a point of said first channel, means for feeding signal from a point of said first channel to a point of said second channel, the points corresponding in the separate channels, and a load device for each of said channels, said means for feeding signal including identical phase shift circuits.
References Cited by the Examiner UNITED STATES PATENTS 2,481,911 9/1949 De Boer et al 179-13 2,762,870 9/1956 Sziklai et al 330-29 X 2,819,342 1/1958 Becker 179-13 3,050,583 8/1962l Berlant 179-13 3,170,991 2/1965 Glasgal 179-13 OTHER REFERENCES Stark: A Continuously Variable Stereo Dimension Control, Audio, July 1959, p. 22.
DAVID G. REDINBAUGH, Primary Examiner,
WILLIAM C. COOPER, Examiner,