Lou Dorren on his CD-4 45rpm record, and FM Quadraplex

QuadraphonicQuad

Help Support QuadraphonicQuad:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I've wondered why the CD-4 system used 2 FM carriers,
it seems to me putting both LF-LB and RF-RB in a single
FM carrier (in both L and R groove sides) would have made
the system more reliable (the phono cartridge wouldn't need
channel separation in the 20kHz to 45kHz range, but would
need a wider frequency range, maybe up to 70kHz or so).

Kirk Bayne
 
Last edited:
I'm no defender of HD radio, but I just wanted to say that having listened to a ton of HD radio at this point, I think that the FM system does sound better then the analog system. It is compressed, but the I-Boc codec is really amazing. I do hear a few artifacts occasionally but for the most part it's a very clean sound.

It's vast improvement over the mp3 codec.

I haven't had a chance to hear I-Boc surround my self. My friend that went to NAB this year had glowing reports about the system that was demoed down there.
 
Like half-speed playback to recover the subcarrier?
Just like half-speed mastering was used to get it on there in the first place.
The idea is so simple there has to be something wrong with it.

Been there done that...
and while half-speed playback for recovering the best audio works wonders with 8 track quad carts, for records it does face a problem that IMHO is very hard to solve: rumble.

Playing a cd4 record half speed can be done with new or old hardware; you may get a old turntable (Dual, Lenco, Garrard...) that features the 16 2/3 speed, or you can get a new DJ turntable (Stanton, Gemini) that features +-50% pitch control (-50% of 33 1/3 is 16 2/3!), so mechanically is not a big problem.

Cartridge: any decent cartridge at half-speed picks up correctly the needed bandwidth (0-22,5k) than you can then double to get the correct f/r. Very good cartridges sampled at hi-rez picks up also the carrier harmonics: since at h/s the carrier is at 15KHz, sampling at 96k (0-48KHz bandwidth) shows 2 and 3 harmonics (30 and 45KHz). A filtering after h/s sampling is mandatory.

Carrier: WOW! Even cd4 in so-so state or that suffer of "sandpaper" played normally do now have a very precise waveform and a strong carrier that shows NO problem on decoding when feeded back into a cd4 demod. A definitive improvement. I do suspect that even hard-to-play CD4, such as Cat Stevens GH, played back at h/s etc will works ok.

These are the easy steps... the BIG problem, as i said before, is not applying a RIAA-Eq or inverse-RIAA or decliking, this works pretty straightfoward... the noise level of the rotating record, aka rumble, is also doubled BOTH in frequency and in output level so it's very present in the mid-low band. I have tried some techniques to get rid of it and for now i haven't found a decent way to do it without killing the sound big time.

Maybe this can be solved only by playing a cd4 record h/s on a optical deck, such as the modern ELP or the old Toshiba: don't know how these deals with the surface noise but if someone has access in such a player it will be nice to do some test.


BTW, the mastering speed for cd4 was reduced by a factor of 2.7, not 2. The JVC pdf shows the equipment used and some other nice stuff.
 
Winopener,

I'm very interested in your experiments with half-speed CD-4 recovery.
Do you have any sound samples, or examples of discs that have shown marked improvement? The one disc I would love to try is Barry Manilow's This One's For You, which has a severe carrier problem on all pressings.

I've never heard of half-speed Q8 playback. What advantages does it have? Lower speed playback should improve bass response, but lower high frequency response and increase noise, wow and flutter.
 
I've done tests using the WEA germany CD4 sampler, a japan JVC of Perez Prado and a serriously mistreated Sinatra "my way" usa. Even the Sinatra decoded back well. Don't have the BM cd4 you asked, so i cant verify it.
Advantage for h/s for q8? Cleaner signal, better defined than n/s, thus easier to clean up. No problem of w/f unless the tape has been duplicated badly (and there are such tapes... believe me!).
 
There was also a Sacramento station that supposedly broadcast Quadraphonic 24 / 7, KWOD (Quad) FM 106.5. I always wondered back then how that station broadcast Quad and how to receive the Quadraphonic music.

Weren't all quad radio stations just running everything through a QS encoder? KZAP 98.5 was quad for a few years as well.

I always wondered how careful those stations were--if they played a QS source, did they bypass any additional encoding? For SQ or discrete, did they decode to four channels then re-encode to QS? Somehow I doubt it...
 
http://www.rockininquad.com


Check out the studio equipment at our own Quad station in the Portland,Oregon area in the 70's.
Here's a "secret" about KQIV... They were never really in quad!!!

I know a few of the former KQIV jocks, and they have great stories about that place. The coolest is how they had the studio monitoring set up. I'm told it was the nicest on air studio monitoring they ever heard!
 
Weren't all quad radio stations just running everything through a QS encoder? KZAP 98.5 was quad for a few years as well.

I always wondered how careful those stations were--if they played a QS source, did they bypass any additional encoding? For SQ or discrete, did they decode to four channels then re-encode to QS? Somehow I doubt it...

Aloha from Kauai today...

It was my understanding that some of the stations Broadcasted In Quad using some sort of encoder with SQ & QS material. They also must have run stereo through it too, as I remember hearing stereo stuff being said to be broadcasted as Quad. I always wondered how to receive such a broadcast - I thought one needed a Sansui at the time.
 
They also must have run stereo through it too, as I remember hearing stereo stuff being said to be broadcasted as Quad. I always wondered how to receive such a broadcast - I thought one needed a Sansui at the time.

Yes, Sansui made the QSE-5B, a broadcast encoder that synthesized stereo material into QS Quad.

Like all of the quad synthesizer systems, sometimes it works great, sometimes not so great.

I have one in my collection, and the manual is included in this link

http://tinyurl.com/2vxclk

from one of my earlier posts. Right click and save.
 
Lou,
I stumbled upon this thread while searching (yet again) for data on the QSI5022.

I'm also an electronics engineer, and worked for an Australian consumer electronics company (AWA, now defunct) in the early 1970s, developing a 'lossless' passive matrix quad decoder while working there (by 'passive', I'm talking about circuits that were hung off the speaker terminals of a conventional stereo amplifier, with no amplification of their own - Hafler developed one of the early [lossy] examples of this type), as well as patenting a front-rear balance control for matrix systems. I've been a great fan of quadraphonics/surround ever since.

One of my current projects is repairing a Technics SH-400 CD-4 demodulator, which uses 2 of your QSI5022 chips. Those chips are OK (I swapped them, and the problem stayed in the same channel), so I'm now working through the (extensive!) peripheral circuitry. The service manual (which contains a schematic) has a block diagram of the QSI5022, but it's a bit thin on detail (many pins emanate from internal blocks without explanation as to their [non-obvious] function). Lucanu's Popular Electronics schematic adds a little more information in the block diagram, but still insufficient for a full understanding.

So I can better understand how the chip works, and thus debug the whole circuit operation, I was hoping you might have something akin to a datasheet for your chip. If not, even just a list of every pin's function would be a great help. Any information would be appreciated.

I found this whole thread most illuminating, and thank you and the other contributors for sharing your deep knowledge of the early quad industry.

Regards,
Daniel
Gerroa, Australia
 
Yes, Sansui made the QSE-5B, a broadcast encoder that synthesized stereo material into QS Quad.

Like all of the quad synthesizer systems, sometimes it works great, sometimes not so great.

I have one in my collection, and the manual is included in this link

http://tinyurl.com/2vxclk

from one of my earlier posts. Right click and save.

Does anyone else have a terrible time trying to get these tiny urls to open or save or anything??
 
Hello Folks,

Sorry for the delay in this posting. I have been busy both in the lab and the recording studio.

Lucanu, A purely real time software approach is still not possible because of processor speeds. To accomplish this task would require a CPU to be running at about 24 GHz. In my new design however, I am using a combination of hardware (Digital Signal Processors, DSP's) and the software to program them. This system is more then fast enough to handle real time CD-4 demodulation.

eggplant, I am including some FFT (Fast Fourier Transform) noise removal in the sum and difference matrix to improve the playback signal to noise ratio.

kfbkfb, The subcarrier situation is a unique an interesting problem. First, it was very hard to get the 30 KHz subcarrier on the disc in the first place. The original master lathe was 1/3 speed cutting to achieve this feat. The simplistic description is that the carriers are FM. This is not completely accurate. These carriers are actually FM-PM-SSBFM carriers, which make the record, playback more difficult. Although it is possible to get a 90 KHz signal recorded on a vinyl record, complex signal integrity above 50 KHz is quite iffy. This and the Nyquist rate are why two 30 KHz carriers where chosen.

Dylan Berichon, The IBOC "CODEC" actually does not create the problem, it is the Adaptive Predictive Throw Away Data Processor that does all the damage. They use high bit rate A to D and D to A conversion which is good and low bit rate 99kbps (FM) and 66 kbps (AM) date compression to loose all the high fidelity information. On receiving the system "predicts" what is supposed to be there.

winopener, Interesting premise. IF however, you find harmonics of the subcarriers at 30 and 45 KHz something is wrong with the signal being sampled. The base 30KHz carrier unmodulated is a low distortion sine wave with 2nd and 3rd harmonic at least 60 dB down. Second harmonic would mean the carrier was becoming triangular and third harmonic would mean that the carrier was becoming square. If both occurred the carrier was becoming trapezoidal. The sampler, unless very nonlinear will not cause this problem.

Rumble tends to be a vertical movement which is Left-Right difference information and can be easily dealt with.

I have tested the ELP laser player and it is not as good as one would think. First the disc must be vacuumed before being played. Even then the 5 beam pickup is very susceptible to dirt impulse noise. The frequency response rolls of rapidly at 21 KHz.

Your Idea is interesting for archive purposes, but obviously not for real time playback.

DanielTheGreat, Send me your email and I will send you a PDF of the QSI5022 Data sheet.

All for now

Lou Dorren
 
Lucanu, A purely real time software approach is still not possible because of processor speeds. To accomplish this task would require a CPU to be running at about 24 GHz. In my new design however, I am using a combination of hardware (Digital Signal Processors, DSP's) and the software to program them. This system is more then fast enough to handle real time CD-4 demodulation.
Lou Dorren

Hi again Lou,

so you're saying we will see a brand new cd-4 demodulator in the future? :banana:

I'm still hoping we can find a way to do a full demodulator via software, even if it's not real time it would be great, really.(y)

Ciao.
 
Hello Lou, here another asking for a later kindly collective answer. I have heard from a JVC demodulator, which the quadraphonic scene would have never seen. This item was developed even in the latest or after quadraphonic time - by a a hope of nevertheless may be 4-channel radio - around 1978-1980 and it should work for such quadraphonic broadcasting. So far I know, this demodulator (with no name) could demodulate CD-4 source and was automatic knowing other discrete and even matrix sources and switching them for the 4-channel output. Of course also a quadraphonic wonder machine or demodulator. Do you know something about this machine -for my curiousity to know all about quadraphonic technics? But I think nevertheless , that your coming new demodulator will be a lot more
interesting for us waiting 4-channel discrete fans.

Dietich
 
Aloha from Kauai today...

It was my understanding that some of the stations Broadcasted In Quad using some sort of encoder with SQ & QS material. They also must have run stereo through it too, as I remember hearing stereo stuff being said to be broadcasted as Quad. I always wondered how to receive such a broadcast - I thought one needed a Sansui at the time.

It was my understanding that FM stations would broadcast matrix-encoded quad material as is, meaning SQ from an SQ-encoded record, QS from a QS record, and they would use the encoder to enhance stereo recordings, or encode discrete to one of the matrix formats. I do have a Sansui QSE-5B encoder that I have used to encode CD-4 to QS for recording onto CD's. It does a great job in doing so. All of the FM stations in South Florida that did quad did it as I mentioned above. Too bad Lou's Quadracast system came too late to have done us any good.
 
winopener, Interesting premise. IF however, you find harmonics of the subcarriers at 30 and 45 KHz something is wrong with the signal being sampled. The base 30KHz carrier unmodulated is a low distortion sine wave with 2nd and 3rd harmonic at least 60 dB down. Second harmonic would mean the carrier was becoming triangular and third harmonic would mean that the carrier was becoming square. If both occurred the carrier was becoming trapezoidal. The sampler, unless very nonlinear will not cause this problem.

Rumble tends to be a vertical movement which is Left-Right difference information and can be easily dealt with.

I have tested the ELP laser player and it is not as good as one would think. First the disc must be vacuumed before being played. Even then the 5 beam pickup is very susceptible to dirt impulse noise. The frequency response rolls of rapidly at 21 KHz.

Your Idea is interesting for archive purposes, but obviously not for real time playback.

Hi Lou,
- the sampling at half-speed is very non-linear; first of all, if done tapping the signal out a phono preamp, RIAA will be wrong big time, so
1. the signal from the phono cartridge is going thru a flat pre-amplifier
2. after the sampling the resulting file is doubled in frequency
3. then RIAA is applied in software.
The harmonics i have seen are between 2. and 3., applying a RIAA curve get rid of most of them.

Any idea to deal with rumble is very welcomed, and not only by me.

ELP may be susceptible to dirt anyway but i think it still holds the advantage of a better separation between channels, which is *the* critical element for any matrix decoder. I know Toshiba did a optical LP player in the seventies, do you had any experience on this?

I prefer to avoid the "magic" :) of real time playback of quad sources for various reasons:
- wear: if something happen to the very rare japanese cd4-only copy of a very sought-after LP, chances to find out another one are pretty slim;
- sound can be improved: NR, decliking and other digital approach can improve a lot the sound just leaving the sound alone and getting rid of the noise due to the media used; especially true for quad carts.
- ease of use: playing a cd4 records on a car is kinda hard... unless it is converted in another format; digital multichannel players, be DD, DTS or DVD-A are available and easier to find than Q8, plus they performs a bit better :D than carts.
- a digital archive can be stored in the little space of a hard disk drive, which make "portability" something easier than before (when you had to move every 3-4 year you start to appreciate that!) and is a near-zero cost sources of a new fresh digital multichannel disc if the first one goes bad.

So, archiving of the analog sources IS my goal for these quad sources, then enjoyable in digital format without any hassle of something going wrong.
 
Back
Top