Lou Dorren: A new CD-4 Demodulator!!! [ARCHIVE]

QuadraphonicQuad

Help Support QuadraphonicQuad:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
A quick QQ CD-4 demodulator update. Time is getting short for the Jon Urban delivery and evaluation. Stay Tuned.
First, thanks Lou for the technical explanations. It's comforting to be aware that the demod is built on knowledge, caring and common sense.

Second, will we have to choose between the 1st issue of the Beatles remasters and the demod? :D
 
Hello QQ Quadies,

jrborg,

You are quite correct. For single channel and two channel play back this is an acceptable method. The problem with this configuration is that if precision gain matching is required (CD-4) the gain setting components must be extremely accurate. A .25dB difference in gain can limit Front to Back separation to less then 25 dB. For your application however, it should work just fine.

Lou Dorren
 
It seems to me that if you reversed the phase of the signal going into the demodulator, because only the baseband audio would be affected and not the subcarrier, that the positions of the front and back speakers would be switched and be out of phase for the side in question. So, if you did not have correction in the preamp, for quad, you would have to reverse not only the phase of the two speakers but you would have to switch their signals to the opposite position. Opun listening to another source, such as a reel to reel or quad eight, it would have to be switched back, and phase corrected. Plus a broken switch could burn up your amp by shorting the speaker line. A very poor way to handle this problem.
 
Lou,
What's the gain difference left to right of a strain gauge cartridge? I know the gain difference left to right isn't 0.25db for any magnetic cartridge that I know of, that's why many CD-4 preamps have adjustments to slightly alter gain levels.
All,
If you look at the mechanical arrangement of the straing gauge cartridge, the need for one channel phase inversion is apparent. When the stylus is moving one direction, one strain gauge is being compressed and the other stretched, leading to out of phase voltages, whereas coil cartridges can be wired internally so that lateral movement can have in phase voltages.
 
It is possible to use the same op amp to have +gain (non inverting, 0 degrees) and -gain (inverting, 180 degrees). I'm working on a restoration preamp that uses just that to switch between lateral and vertical records.

Here you have a nice solution:
 

Attachments

  • phaseswitch.JPG
    phaseswitch.JPG
    88.1 KB · Views: 533
Last edited:
Hello CD-4 Fans,

Quadfather,
You are correct.

jrborg,

I can not remember the published spec for the 450, 451, and 460. I do, however, have my measurements for 10 EPC451C pickups. The worst one was .63dB left to right and the best one was .22 dB left to right. One thing to remember is that in CD-4 playback the F-B signal is coming from a fixed known FM demodulated signal. The F+B signal comes from the pickup which has a relative output of as much as +- 5dB. This is why there is a separation control for the Left side and one for the Right.

rolvkarsten,

This circuit will work, however, to maintain a match in each phase position to yield 25 dB Front to Back CD-4 separation resistors R2, R3, and R4 must be matched to .05% or better. I also assume the you mean to have the inputs independent or switched. The way it is currently drawn when the switch is open, depending on the resistor match, very little to nothing will come out the output node of the amplifier.

Lou Dorren
 
BTW, what can we expect regarding stereo LP reproduction through the demod with a high-output MC cartridge?

Thanks in advance.
 
Hello Quadience,

kfbkfb,

Kirk, The problem with the automatic separation circuit is that it does not perform as well as manual separation adjustments. It also uses a great deal of circuitry to do what we can do with our fingers and ears, and without the need of a special test record (it has to be redone often because "automatic" separation adjust does not hold for very long). I argued this point quite a bit with the Technics engineers including several demonstrations, but alas you can't win them all.

proufo,

As long as the pickup has at least 5 dB of separation at 50KHz and amplitude variance of less than 5dB, Front to Back separation will be 25 dB or more. This is regardless of pickup type (MM, MC, SC).

Lou Dorren
 
Last edited:
Oh my goodness - I stray away from QQ for a year and this turns up!

Wow my head hurts... I've just read through all 30 pages of this thread and (perhaps I should just come back tomorrow) one question above all else remains - it's not too late to join the list? Or? (yes I did that on the other thread pointed to for that purpose). Golly I hope not.

More things...

I'm puzzled by Lou's decision to enforce the shipping of all demod's with a cartridge of his selection (with spare stylii at 39.00 USD??) I've always committed myself to using good cartridges whatever the format and have settled on the Denon DL-304 m/c which has a frequency response up to 50K and when I record a CD-4 (in stereo) to a 96Khz WAV I can see the carrier signal lit up like an airstrip. Surely I do not need to use the cartridge that will come with the demod? I'm not griping, just curious. It seems to me that it would be near impossible to produce a stylus that can perform as well as my Denon m/c for that sort of money. I also have the problem that I can no longer quickly change cartridges by using another headshell now that I've fitted a one-piece tonearm to my Technics!

More on the Technics: Someone mentioned the problems with internal wiring on the Technics SL1200 - I saw too that it wasn't a problem in itself, after all, but thought you'd like to know that you can upgrade this venerable machine to something approaching a reeeeally good deck by simply fitting a new arm. I fit the Rega RB250 with mods by Origin Live! (that's how they! spell! it) and frankly my old Michell is still awaiting a better power supply while this Technics rocks away.
Also some clumsy deejay somewhere will glady help you pay for this when you list your crap Technics arm on ebay!

I'm a committed 'digitizer' for what it's worth, largely because I'm too poor to justify any spending on vinyl to keep when I can record it and then sell it on... but also because I have been an audio restorer for record labels for some years now and I'm darn good at it. (Nobody wants me to restore any quad stuff - just stereo and mono).

I've got a couple of questions for anyone else restoring the audio from CD-4 discs:

Is it better to do click removal before or after the demod? Bearing in mind that I have software that can (sort of) recreate the missing audio where a click was, what is the maximum and/or minimum time for a drop-out to occur in the carrier signal?

What are the groove noise issues in a demod'ed signal? Will sophisticated algo's do anything to help (CEDAR etc)? Is there a thread for this?

I am absolutely thrilled by the idea of using a restored signal on, say a DVD-Audio disc (I can make those myself) and passing it through the demod to get pure quad! I'm also excited by the idea that I could just record the demod'ed 4 channels and do a DVD-A to listen to that way. Whatever works best. Any other 'digitizers' out there care to communicate with me?

All these delights will soon be mine to discover for myself I guess... I hope! I think that a couple of people on this thread maybe started to run out of patience with Lou but I reckon the more patient among us may be rewarded beyond our wildest quad dreams!

Lou? Do what you need to do to pay the bills - I've got 510 dollars here for you when you're ready! Not too poor for that.

:smokin

Colin Young AKA See Why Audio
 
Hello 4 Channelphiles,

See Why Audio,

Colin, Welcome back to QQ. My idea to ship a CD-4 pickup is because a lot of people are using pickups that do not fully yield a full CD-4 signal. The sub-channel signal is quite complex being an FM-PM-SSBFM. While some lower performance pickups with a good demodulator will yield some CD-4 output, it will not be optimum. I want to make sure that anybody shelling out $505.00 for the new Demodulator are going to get maximum performance. The $39.00 price is not for a pickup, it is for a spare stylus. Since the pickup with a Shibata stylus is a part of the system, it comes with the demodulator at the $505.00.

This pickup is being custom made and tipped with a diamond Shibata stylus. The eval units that I have tested have a frequency response above 60KHz. The real important factor is sub-channel Left to Right separation. At 60KHz it has 15dB or greater. No pickup will ship until it is tested. This will guarantee maximum CD-4 performance.

The Denon DL-304 is a very good pickup and should work well. It does have a flaw however. It uses a .7 X .1 elliptical stylus which at 1.5 grams tracking pressure is quite hard on LP records, both stereo and CD-4. Re-tipping with a Shibata would fix the flaw.

As far as CD-4 digitizing, the new demodulator is set up for it. You should use the QSI record cleaner and play the record wet. Carrier drops generally occur from dirty records and the new demodulator has a dropout compensator which prevents burst noise. The record cleaner will fix most of the impulse noises. Procedures for doing this are in the manual that comes with the cleaner.

General noise noise reduction should be done after demodulation and sum and difference matrixing. I recommend an FFT algorithm for this purpose.

Lou Dorren
 
Hi Lou and thanks for the detailed reply - I see you are a man of conviction and no mean knowledge! I salute you. By the way the one question I really wanted an answer to was 'is it too late to join the list'? :D

Re: the Denon DL-304 cartridge, good to know it will 'work well' but even more intrigued by the concept of re-tipping it with a Shibata stylus. I also have an old worn Lyra Helikon that could use a re-tip.
Does anyone out there know someone who could do this for me? I have a contact for someone called 'the cartridge man' here in the UK but he has some odd restrictions on price-points (the Denon would be too cheap although I have been meaning to send him the Lyra) I haven't seen any mention of a Shibata stylus option. Anyone know?

Lou, further to your comments on record cleaning:
It was depressing to note your comments on the effects of alcohol. After years of searching for the most effective way to reduce surface noise, and many different (and sometimes expensive) solutions (liquid) I have alighted on what seems to be the most effective one so far: Isopropyl alcohol at 30%, distilled water at 70% with a couple of drops of photographic wetting agent, and I run discs through my Vpi HW-17F vacuum wet solution cleaner. 'Depressing' because I must have cleaned 2 or 3 thousand LPs this way! I hope I haven't ruined all of them. Some of the records I have cleaned for clients are the only known copy FFS! Thing is I cannot see any discs that were damaged by it? How would I even know? As far as I can tell it has only ever been an improvement and for getting rid of so much background noise artifacts on dirty records it makes my audio restoration a LOT easier. In fact I use it on every vinyl I record as standard practice no matter how clean it appears. I have genuinely yet to come across a piece of vinyl that became more noisy after using the cleaner. For what it's worth I don't usually allow the solution to stay on the surface for more than about 15-20 seconds at a time which seems to be as long as is necessary to dissove any encrusted stuff. How long does it take for alcohol to do it's damage?

I just can't get my head around the idea of using the stylus to 'scoop dirt out' and as far as I know there are many kinds of 'matter' that wouldn't be emulsified even with neat hydrochloric acid! So I still feel that physically removing the dirt and dust is the best concept of record cleaning. Would your solution be of any use in a Vpi cleaner?
Unfortunately at my large rate of use I could never really afford to use it regularly, I find myself cleaning approx 30-40 discs a week! Do you or anyone else know of a less dangerous yet effective solution I could safely mix myself? I'm afraid it's a cost thing Lou...

Look forward to any replies.

Oh another PS: Doobie Brothers Stampede German pressing - on close inspection the matrix numbers appear to be similar to many otehr hand written numbers that came from the USA. The 'made in Germany' is something that is/was a legal requirement for German products and most certainly was applied at a seperate stage in production. Beautiful pressing I must say. Can't wait to hear it quadraphonic!

Colin
 
Doobie Brothers Stampede German pressing - on close inspection the matrix numbers appear to be similar to many otehr hand written numbers that came from the USA. The 'made in Germany' is something that is/was a legal requirement for German products and most certainly was applied at a seperate stage in production. Beautiful pressing I must say. Can't wait to hear it quadraphonic!
Colin

Like all of the early Doobie Brothers CD-4s, Stampede has a fine Surround Sound mix. You'll enjoy it !
 
Hello AudioStereoQuadraphonicCD-4philes,

See Why Audio,

Colin, First it is not to late. Sign up on the other thread (If you are a low distortion signal generator Sine up on the other thread). A little audio humor (very little).

I will have some re-tipping info for you shortly. In one quantities it is not cheap!

I did a lot of work on the cleaning problem. In previous posts I have some electron scan microscopy photos of the problem and the result. The QSI cleaner we developed in the early CD-4 days was one of the main reasons for improved record compound (Q540) development. The stylus scoop method works very well with all styli, however the Shibata stylus, because of its shape works the best. The QSI liquid is very good at masking surface noise when used in archival playback.

Alcohol removes the lubrication that is in the record compound and the environment and and playing of the record create the micro cracks and dirt build up. The damage is done on the first use by removing the lube. The time for the microcracks to form depends on tracking pressure, number of plays and storage environment.

There is a great deal of particles in the air that are emulsified by the QSI cleaner. Material that does not break down becomes suspended in the liquid waiting for pickup. We found this out by analyzing the "hair ball" on the stylus after the first playback of the cleaning cycle.

I tried several different cleaning devices including vacuum types, and even though it is a bit time consuming, the scoopy play method gives the best result.

Unless you are in a sand storm or are playing records in a green house full of pollen, one cleaning application of the QSI Cleaner should last 4 to 6 months or longer depending on the non use storage of the record.

That pressing you have was mastered in the USA and pressed in Germany.

Lou Dorren
 
Regarding the CD-4 FM based Difference Signal S/N:

Tandberg TPT-3001 FM Tuner:
FM mono S/N 88dB
http://www.audio-extasa.eu/tandberg-tpt-3001-fm-tuner-p-1024.html
(for example, my Pioneer Receivers:
VSX-D209 FM mono S/N 76dB
VSX-D514 FM mono S/N 73dB)

Can the technology Tandberg used be
adapted to maximize the demodulated
CD-4 FM based difference signal S/N?

Kirk Bayne
 
Hello QQers,

kfbkfb,
Kirk,

The problem here is not the FM processing system, it is the system being detected. Commercial FM broadcasting has a maximum FM deviation of plus and minus 75 KHz which is a modulation index at the highest audio modulation frequency (15KHz) of 5. In mono that can give you a theoretical dynamic range of over 90 dB. This means that at full quieting RF level, 80 to 90 dB Signal to noise ratios are possible.

With CD-4 the deviation is low with an modulation index at the highest audio modulation frequency (15KHz)of well below 1. This is a stand alone dynamic range of 48 dB. When you add the ANRS you get a dynamic range of about 60 dB thus a Signal to Noise ratio of about 60 dB. This was the reason for ANRS.

The best signal to noise ratio I have ever measured from the audio frequency range of an LP record( 30Hz to 15KHz) is 68 dB. That was first playback from audio cut on a master lacquer. On a pressing with Q540 compound, the best signal to noise ration in the audio range I measured upon the first play was 61 dB.

In the development of the CD-4 system, the conventional detectors (discriminator, ratio detector, and quadrature phase detector) would all yield no more than 35 dB signal to noise, while the limited driven Phase Lock Loop yielded greater than 40dB. It was also much less prone to burbs and noise spurts than the other detectors.

Good Labor Day to Everyone,

Lou Dorren
 
Regarding CD-4 "radar" false triggering:

Some early 1970s CD-4 related speculation in
various Hi-Fi/Electronics magazines was that
some mono or stereo albums/45s could have
enough distortion products in the 20kHz to 45kHz
frequency range to trigger the CD-4 "radar".

I tried the Peace Train song from my Sheffield
Lab Direct Cut LP (the cymbal at the end of the
song has a great deal of high frequency content
as shown on P.60 in the 1974-02 Stereo Review)
with the demodulator in my Pioneer QX-747 with
the carrier level control at maximum, the CD-4
"radar" wasn't triggered.

Have any mono or stereo albums/45s been found that
do have enough high frequency content to trigger
(even momentarily) the CD-4 "radar"?

Kirk Bayne
 
I have a copy of Elton John's Goodbye Yellow Brick Road that drives my 4DD5 nuts. I have to switch to stereo mode to play it. I have noticed places on other albums as well that will cause false trigger of the "radar" circuits, and insertion of discriminator noise. It is always best to use the stereo position on the switch on the demodulator to play a stereo record, or just use the amplifier's preamp. Most demods have a provision for this. This also applies to SQ, QS and other matrix records. UD-4? I don't know about that one.

The Quadfather
 
The Ortofon MCA-76 moving coil prepreamplifier had a button marked "CD-4" that is really a high frequency filter button to eliminate this effect. Some people wrongly believe that it is used FOR CD-4 playback, but the matching SL15Q moving coil cartridge was equipped with a Shibata stylus that could extract very high frequency harmonics from vinyl. Moving coils are great for CD-4 playback because they are immune to cable capacitance ... tougher to match up with cheaper tonearms though, but they have much better high frequency range than magnetic pickups.
 
Back
Top