Quadraphonic audio on vinyl records was problematic because some systems were based on discrete sound channels (allowing for full separation of the four original recorded channels, albeit with restricted high-frequency response and reduced record life), while others were matrix encoded into two tracks that would also play back in standard, two-channel, stereo on normal audio equipment (so-called 'compatible' quadraphonic).
There were some experiments done with radio broadcasts (e.g. a Cliff Richard concert by the BBC), but they were short-lived. One of the longest-lived radio broadcasts was WQSR-FM "Quad 102 1/2" in Sarasota, Florida. Throughout most of the 1970s this station broadcast a signal which could be tuned as two separate stations with conventional stereo receivers. In addition, San Francisco classical music station KKHI broadcast the San Francisco Opera in 'compatible' (that is, matrix encoded) quadraphonic format during the 1970s.
Compatible Discrete 4 (CD-4) or Quadradisc (not to be confused with compact disc) was introduced in 1971 as a discrete quadraphonic system created by JVC. Record companies who adopted this format include Arista, Atlantic, Capricorn, Elektra, Fantasy, JVC, Nonesuch, RCA, Reprise and Warner.
This was the only fully discrete Quadraphonic Phonograph record system to gain major industry acceptance.
In the CD-4 system, the Quadraphonic audio was divided into Left and Right channels with the Left recorded on one groove wall and the Right on the other, which is the case with normal stereo. The audio frequencies (20Hz to 15KHz), often referred to as the sum channel, would contain the sum of the Left Front plus Left Back signals in the Left channel and the sum of the Right Front plus the Right Back signals in the Right channel. In other words, if you looked at the audio frequencies only, you had an ordinary stereo recording. Along with this audio, a separate 30-kHz sub-carrier was recorded on each groove wall. The sub-carrier on each side carried the difference signal for that side. This was the information that enabled a combined signal to be resolved into two separate signals. For the Left sub-carrier it would be Left Front minus Left Back, and for the Right sub-carrier it would be the Right Front minus the Right Back. These audio signals were modulated onto the carriers using a special FM-PM-SSBFM (Frequency Modulation-Phase Modulation-Single Sideband Frequency Modulation) technique. This created an extended sub-carrier frequency range from 18 kHz to 45 kHz for the left and right channels. The algebraic addition and subtraction of the sum and difference signals would then yield compatible and discrete Quadraphonic playback. CD-4 was responsible for major improvements in phonograph technology including better compliance, lower distortion levels, pick-up cartridges with a significantly higher frequency range, and new record compounds such as Q-540, which were highly anti-static. Also a direct by-product of CD-4 technology was the Shibata contact stylus. Invented by Dr. Norio Shibata, the Shibata stylus greatly improved the contact area on the groove walls, which in turn reduced wear over time, thus maintaining the original LP's fidelity and frequency range for much longer than normal. This, of course, prolonged the overall playable life of a record – a definite boon for any audiophile or serious record collector. But what was (arguably) even more impressive was that the Shibata stylus was also capable of recovering sizeable portions of program audio that would otherwise be compromised or lost on a damaged or scratched disc (audio that a standard, non-CD-4-compatible stylus would fail to pick up). A typical CD-4 system would have a turntable with a CD-4 cartridge, a CD-4 demodulator, a discrete four-channel amplifier, and (ideally) four full-range loudspeakers. Some manufactures built the CD-4 demodulator into complete four-channel receivers.
Simply put, CD-4 consists of four recorded signals (LF, LB, RB, RF) and the following coding matrix.
The CD-4 encoding/decoding matrix:
This coding is needed because the high frequency "difference" signals have limited bandwidth, and so are not suitable as normal audio channels. However, when a limited bandwidth is used as a difference signal, the resulting four channels of music are of full bandwidth, although there is a very slight limitation in how each side (or sum) is resolved into front and rear sounds.
Although CD-4 (and Quadraphonic audio in general) failed due to late FCC approval of FM Quadraphonic broadcasting, the improvements CD-4 engendered spilled over into, and substantially improved, the production of conventional stereo LP records.
UD-4/UMX - Developed by Nippon/Columbia (Denon). This is a hybrid discrete/matrix system. Only 35 to 40 items are encoded in this format, and it was marketed only in the UK, Europe and Japan.
A regular matrix decoder could be used to play back these recordings, but, by adding a special cartridge and a UD-4 demodulator, two supplementary channels could be extracted and used to enhance directional resolution. UD-4 systems first encoded the four original channels into four new channels. Two of these new channels contained the original four channels, matrix encoded. The other two contained only band-limited localization information, and were encoded with sub-carriers similar to the CD-4 system. UD-4 was less critical in its set-up than CD-4, because the sub-carriers did not have to carry frequencies as high as those found in the CD-4 system.
The last release in the quadraphonic 8-track format was in 1978.
Stereo Quadraphonic was a matrix quadraphonic system for vinyl. It was introduced by CBS in 1972, and record companies who adopted this format include: Angel, Capitol, CBS, CTI, Columbia, EMI, Epic, Eurodisc, Harvest, HMV, Seraphim, Supraphon and Vanguard.
The system was based on the work of Peter Scheiber, who created the basic mathematical formulas used to matrix four channels into two in 1970. SQ stands for "Stereo Quadraphonic." This system made good sense, as, in the absence of a quad decoder, SQ-encoded records would play as normal stereo records, and CBS stated their desire to maintain excellent compatibility between their SQ-encoded records and standard stereo systems. Additionally – and perhaps most importantly – these types of record, along with the competing QS format, allowed the full bandwidth from 20 Hz to 20 kHz to be reproduced, giving a much more "open" and detailed top end.
The early SQ decoders could not produce more than 3 dB of separation from front to back. Early "Logic" circuits were introduced to enhance separation to 20 dB, but provided poor performance, very noticeable gain-pumping and an unstable, 'swaying' soundfield. The SQ system also faced resistance from broadcasters, since while essentially a 2-channel system, and totally stereo-compatible, it could have substantial mono compatibility problems, which posed serious problems with all televisions and monophonic radios of the era.
By the time that the most advanced Logic System was introduced for SQ, the Tate Directional Enhancement System "quad" was already considered a failure. Initially, the Tate DES chips were in short supply, as the original run was committed for use at cinemas in the first generation of Dolby matrix surround-sound systems installed for the first release of Star Wars (1977). They were also used in decoders produced by Audionics of Oregon and Fosgate. These units are still sought by SQ collectors for their superior performance.
A Prologic II decoder will recover some of the surround sound information present in an SQ mix, as the matrices used are somewhat similar, but directional cues will not be properly located. Some of the SQ mixes are still present on CDs, especially on early, non-remastered editions, on which the original master is, in itself, SQ. (SQ is compatible with 2-channel stereo, so there was no need to create different versions of it.)
| Basic SQ Encoding Matrix | Left Front | Right Front | Left Back | Right Back |
|---|---|---|---|---|
| Left Total | 1.0 | 0.0 | -j0.7 | 0.7 |
| Right Total | 0.0 | 1.0 | -0.7 | j0.7 |
To provide mono-compatibility a variation on this matrix was proposed:
| Modified SQ Encoding Matrix | Left Front | Right Front | Left Back | Right Back |
|---|---|---|---|---|
| Left Total | 1.0 | 0.0 | 0.7 | -j0.7 |
| Right Total | 0.0 | 1.0 | -j0.7 | 0.7 |
The four channels were coded and decoded normally in this proposal, but the back-center channel was coded in phase and therefore decoded in front-center.
Quadraphonic Stereo (sometimes referred to as RM or "Regular Matrix") was a system that was conceptually very similar to SQ, but developed independently by engineer Isao Itoh of Sansui, adopted by ABC, Advent, Bluesway, Candide, Command, Decca, Impulse, Longines, MCA, Ovation, Pye, Turnabout and Vox record companies. It was freely licensed to record companies, but was rarely found on receivers other than those made by Sansui. The QS matrix has been found to offer the advantages of excellent diagonal separation and stereo compatibility, and although the adjacent speaker separation is only 3 dB, this symmetrical distribution produces more stable quadraphonic images. Sansui's decoders also had good stereo-to-quad capabilities, wrapping the L-R panorama to LB-LF-RF-RB in a horseshoe topology. Two outboard decoders, the QSD-1 and -2, as well as the QRX- series of larger receiver-amplifiers, incorporated this matrix and up-conversion.
The QS matrix system was employed to create the five-channel Quintaphonic Sound system used for première engagements of the 1975 film Tommy. The left and right 35 mm magnetic soundtracks were QS-encoded to create four channels around the cinema audience, while the centre mag track was assigned to the speaker behind the screen. The mag FX track was unused. This channel layout came close (5.0) to the later usage in video 5.1 surround sound.
| QS Encoding Matrix | Left Front | Right Front | Left Back | Right Back |
|---|---|---|---|---|
| Left Total | 0.92 | 0.38 | j0.92 | j0.38 |
| Right Total | 0.38 | 0.92 | -j0.38 | -j0.92 |
A much simpler form of the Dynaco patent keeps the four speakers in their normal left and right plus front and rear positions. The left and right rear speakers are connected to the two-channel stereo amplifier via a passive matrix circuit, while the front ones stay directly connected to the amplifier. A lot of stereo material, recorded with a central, non-directional microphone (so-called kidney sensitivity diagram) placed in front of the orchestra, possessed suitable difference signals across the stereo signals. When taken across this passive speaker matrix for the rear channels, these produced a quasi-quadraphonic effect at low cost (the patent specifies the use of one fixed 10-Ohm resistor and three variable 20-Ohms resistors in a star arrangement). Especially for classical music, a fine impression of concert-hall ambience is achieved with such a system.
Dynaco sold this matrix circuit with a large and triple high-wattage potentiometer inside, for a sum equivalent to 70 euros or so in present-day currency. Electronic amateurs could build this circuit much more cheaply – e.g., with a 4-position switch (four steps in level of the rear sound from min. to max. level) using fixed resistors of, for example, 20, 10, 5 and 0 (short-circuit) Ohms. Because, in practice, only the highest level was of any use, a more basic set-up with only the fixed 10-Ohm resistor at close-to-zero cost is possible.
Note that the system requires more or less flat impedance curves for the rear speakers to work properly, which was often the case in the tube-amplifier days. Tube amplifiers had a constant impedance over a wide range, and worked best with high-efficiency speakers. Later on, when transistor amplifiers were used, speakers tended to lose that design feature. (Lower impedance meant higher power output for these amplifiers, compensating for the lower efficiency of such designs.) The system worked best using a transistor-based stereo amplifier, low-efficiency front speakers, and high-efficiency, constant impedance rear speakers.
The rear sound level in a live performance recorded in stereo is reproduced about 7 dB below the front level, but clearly audible. The rear ambient sounds, applause, and coughs from the audience, are sometimes received out of phase by the stereo microphones, while sounds from the musicians mostly are in "synchronous phase". Thus, if rear speakers are fed with the difference between the stereo channels, audience noises and reverberation from the auditorium may be heard from behind the listener. This can be most easily achieved by wiring two similar additional rear speakers in series between the live feeds (positive terminals) from the stereo amplifier. Alternatively, one rear speaker can be used on its own. (See External links for a circuit diagram.)