(1974-12) SQ Stereo Enhancement [CBS Labs]

[Soundfield]

Here is a copy of the final version of my SQ Encoder...
To use to synthesise SQ from stereo, only the front inputs are used. With the pot turned up a left input is encoded as a left back and right as right back, while center front stays upfront. , 270° enhancement. The other extreme produces no enhancement 0° .

I have found this discussion particularly interesting, thanks.
So to reduce your circuit to the absolute minimum, all you need to create a 270° Synthesiser from two SQ decoders is a single inverter-

I must try this with two Surround Masters!

 

[par4ken]

I have found this discussion particularly interesting, thanks.
So to reduce your circuit to the absolute minimum, all you need to create a 270° Synthesiser from two SQ decoders is a single inverter-

View attachment 76490



I must try this with two Surround Masters!

The first decoder has to be the basic type (no logic), so two Surround Masters won't work.

Edit,: Also the first decoder (used as an encoder) can't use any blend across the back outputs.

 

[Soundfield]

The first decoder has to be the basic type (no logic), so two Surround Masters won't work.

Edit,: Also the first decoder (used as an encoder) can't use any blend across the back outputs.

Ah, yes good point - I'll have to find my MC1312 chips!

 

[par4ken]

The CBS designed SQ 4-channel Synthesizer was featured in the Oct 1975 issue of Radio Electronics. I don't remember if it was posted here before or not. I recommend following the original CBS document as the RE article contains many errors, the description and even the schematic. I'm posting it simply as an additional reference.

 

[Soundfield]

The CBS designed SQ 4-channel Synthesizer was featured in the Oct 1975 issue of Radio Electronics. I don't remember if it was posted here before or not. I recommend following the original CBS document as the RE article contains many errors, the description and even the schematic. I'm posting it simply as an additional reference.

Both sources are a little odd. I wonder why for example neither has a coupling capacitor from the MC1312 RB out (pin 14) to the transistor stage whilst the other three outputs are coupled via 2.2uF capacitors

 

[par4ken]

Both sources are a little odd. I wonder why for example neither has a coupling capacitor from the MC1312 RB out (pin 14) to the transistor stage whilst the other three outputs are coupled via 2.2uF capacitors

That is because the coupling capacitors are required so that the transistors bias is not disrupted. Q3 is biased directly by the voltage from pin 14 of the decoder IC (about half the supply voltage). Q2 gets it's bias from the collector of Q1. You might be able to get away without C12 and C13, but they could be there to block any differing DC levels between the various outputs.

 

[Soundfield]

That is because the coupling capacitors are required so that the transistors bias is not disrupted. Q3 is biased directly by the voltage from pin 14 of the decoder IC (about half the supply voltage). Q2 gets it's bias from the collector of Q1. You might be able to get away without C12 and C13, but they could be their to block any differing DC levels between the various outputs.

Oh, yes- I'd not noticed there was no other means of biasing Q3! I should have looked more closely - thanks!

 

[par4ken]

I found this link if anyone wants to construct a decoder using stripboard.

https://www.petervis.com/record_players_and_turntables/quad-sq-part-1/quad-decoder.html

 

[Soundfield]

I found this link if anyone wants to construct a decoder using stripboard.

https://www.petervis.com/record_players_and_turntables/quad-sq-part-1/quad-decoder.html

He also published a decoder using all discrete components - rather bizarrely using MOSFETS (!) -

MOSFET SQ Decoder

 

[par4ken]

Years ago I purchased a number of SQ boards from surplus dealer Addison Electronics. They were rather unique in that they used the MC1313P chip. It is the same as the MC1312P but designed for 12V operation. I guess that those boards had been intended for use in an automotive decoder. I depleted my small supply of boards building encoders for myself and others. For the most part I replaced the 1313 chip with the 1312 and so still have a stock of those now rare parts.

I was even going to make a car decoder, it was to include bass and treble controls as well. The boards are still sitting in one of my junk boxes. As an avid DIYer I often had a number of aborted or unfinished projects kicking around. About that time Fosgate was announcing their Tetra Sound car decoder. It was my intention of getting one (regardless of cost) but sadly it never materialised.

 

[par4ken]

I would suggest using Fosgate's 6 pole filter. It uses standard value capacitors and standard 1% resistors.

 

[Sonik Wiz]

I would suggest using Fosgate's 6 pole filter. It uses standard value capacitors and standard 1% resistors.

For some reason I just "get along" better with IC's doing DIY. If I were to muck about with SQ or 90 deg audio circuits I would probably use the design used in the Integrex Ambisonic decoder. The article sez:

The design shown is a high-quality unity -gain eight -pole design giving 90° relative phase shift over the frequency range 30Hz to 16kHz, ideally with an error of ± 1'/3°, but with an error of ± 3°approximately using 2% tolerance components.

It just seems simpler to me if I were to translate schematic to perfboard. It doesn't use stock components as you can see some resistors are in series to get the required value. From PDF I can copy text but not images so if you want to check it out I can only give you the Radio History Archive link:

https://worldradiohistory.com/UK/Wireless-World/70s/Wireless-World-1977-08.pdf
But then I bet a few of us already have that DL'ed somewhere...

 

[par4ken]

Yes an eight pole filter would be even better than the six pole one. I don't think that either circuit would be hard to breadboard. I would use 1% components for either though, all readily available.

 

[par4ken]

t just seems simpler to me if I were to translate schematic to perfboard. It doesn't use stock components as you can see some resistors are in series to get the required value. From PDF I can copy text but not images so if you want to check it out I can only give you the Radio History Archive link:

The Fosgate circuit parallels capacitors to make up different values but the capacitors used are all standard. The Ambisonic decoder connects resistors in series to make up a non standard values but still the parts all all standard. The Audionics S&IC had their own special (non standard ) capacitors made for the phase shift network. At one time I was going to try to duplicate the Audionics circuit by paralleling capacitors, not an easy task! I recall seeing the circuit used by CBS as well and each pole of that ten pole filter used it's own op-amp buffer, so ten op-amps in series!

Often those old circuits from the seventies used 2% tolerance components but more often 5%. Today 1% even 0.5% components are readily available (and inexpensive).

With the Fosgate circuit you could lay the components out the same way that they are drawn in the schematic. A large piece of perf-board would be easier than a copper clad stripboard, simply push the leads though the holes and solder the component leads together on the bottom of the board.

 

[jupp369]

MC1312P, MC1314P, MC1315P
is it worth building it with the normal CBS plan ?

 

[Sonik Wiz]

MC1312P, MC1314P, MC1315P
is it worth building it with the normal CBS plan ?

I think the answer depends on what your motivations or goals are for doing so.

If you just want the fun of building & the chance to hear vintage full logic SQ, sure. I built a decoder with these chips back in '74(?) & it sounded ok. It really was a nifty design solution from CBS/Motorola to otherwise very complex design.

On the other hand much has been improved since then. I am dubious of the quality of control voltage time constants, the quality of internal VCA's, the quality of the internal op amps. There are several other choices, vintage & new, that I would reccomend if you are aiming for best quality sound.

 

[jupp369]

Ok Sonik, what should I use instead of this?
Since '70s I was happy with the 4 Matrixes offfert by my Toshiba 404 and the CD-4. Normal TV and normal Stereo was always played by their 5 transistor SQ, which I also implementet in my JVC 5436's. But while taking a closer look at the CD-4's ripped to computer, I think, the pure SQ matrix is not so bad at all if decoded better than with 5 transistors.

 

[Sonik Wiz]

Ok Sonik, what should I use instead of this?
Since '70s I was happy with the 4 Matrixes offfert by my Toshiba 404 and the CD-4. Normal TV and normal Stereo was always played by their 5 transistor SQ, which I also implementet in my JVC 5436's. But while taking a closer look at the CD-4's ripped to computer, I think, the pure SQ matrix is not so bad at all if decoded better than with 5 transistors.

As we are drifting off topic from @kfbkfb original post I'll keep this short. Maybe start a new thread of your own on the topic?

Legacy equipment that has held it's respect and value over the years is the Audionics Space & Image Composer and Fosgate Tate 101A, both great for SQ decoding & stereo to surround. Also most any of the Sansui QS Vario-matrix decoders/receivers have excellent QS & stereo surround, and a function of pretty good SQ decoding. Any of these would probably represent a big step up from the simple decoders in the Toshiba. Of course with these units you have the challenge that all vintage gear brings.

For modern gear there is a huge amount of processing modes & functions in many audio video receivers. But the Involve Surround Master would easily integrate into your system. Excellent decoding for QS/SQ and almost as good stereo to surround as on the older gear. Many advantages to this and it's well documented on the forum. Poke around a bit. Really the only choice I would consider a wise investment.

 

[par4ken]

MC1312P, MC1314P, MC1315P
is it worth building it with the normal CBS plan ?

I agree with Sonic, If you want to experiment the Motorola chips are excellent. IMHO the Motorola based decoder is as good as or better than the Sony decoders which are much more complex and require a very finicky adjustment procedure and (not really suitable for a DIY project). Too bad PC Boards aren't available anymore, but the circuit shouldn't be too bad to breadboard.
For higher level decoding the modern alternative is Involve, but at a much higher cost.

 

[barfle]

MC1312P, MC1314P, MC1315P
is it worth building it with the normal CBS plan ?

Good luck finding the chips. I’ve had an ebay search for them active for a decade.