Non-LP sources of CD-4 music

[jupp369]

The block diagrams of CD-4 supplied in the 1970s show that the carrier is split off just after the first preamp. Then the demodulation is done, and then the RIAA de-emphasis is applied.



The problem is that the recording medium would have to have a frequency response out to 50KHz

RIAA is not needed for just certain kinds of magnetic cartridge. It is needed for all magnetic cartridges. Ceramic cartridge bars are specially "cut" so the ceramic itself has the RIAA curve.

The RIAA recording curve has several purposes:
1. Increase recording time by reducing the deep bass to reduce cutter swing at low frequencies.
2. Increase high frequency response by increasing the high frequencies sent to the cutter.
3. Make all records play the same way - a standard.

View attachment 89073
The RIAA playback curve is the opposite of the recording curve. It boost the bass and cuts the treble in exactly the same amounts the recording curve changed them. The result is an exact reproduction of the original sound.

Note that both curves steeply cut frequencies lower than 15Hz and higher than 22KHz.

Yes, but the carrier survives, without bypass.

 

[barfle]

The block diagrams of CD-4 supplied in the 1970s show that the carrier is split off just after the first preamp. Then the demodulation is done, and then the RIAA de-emphasis is applied.



The problem is that the recording medium would have to have a frequency response out to 50KHz

RIAA is not needed for just certain kinds of magnetic cartridge. It is needed for all magnetic cartridges. Ceramic cartridge bars are specially "cut" so the ceramic itself has the RIAA curve.

The RIAA recording curve has several purposes:
1. Increase recording time by reducing the deep bass to reduce cutter swing at low frequencies.
2. Increase high frequency response by increasing the high frequencies sent to the cutter.
3. Make all records play the same way - a standard.

View attachment 89073
The RIAA playback curve is the opposite of the recording curve. It boost the bass and cuts the treble in exactly the same amounts the recording curve changed them. The result is an exact reproduction of the original sound.

Note that both curves steeply cut frequencies lower than 15Hz and higher than 22KHz.

Excellent explanation of why trying to get CD4 carrier out of an RIAA preamp won’t be successful.

 

[par4ken]

I suppose that It would depend on the design of the actual demodulator.

This diagram from the Panasonic SE-405H manual (the demodulator that jupp369 is working with) illustrates just what he is saying. The RIAA comes first.

 

[par4ken]

Here is a JVC diagram that shows the RIAA done as a separate path.

 

[Sonik Wiz]

Here is a JVC diagram that shows the RIAA done as a separate path.

View attachment 89105

When I did some service work on a Marantz 400-B I was surprised to see the RIAA EQ pre-amp was also done before low/high pass filtering. That really seemed odd to me. But as you have demonstrated it seems to be done both ways depending on the unit.

 

[jupp369]

Here is a JVC diagram that shows the RIAA done as a separate path.

View attachment 89105

In the SE-405H scheme "Equalizer" is the RIAA.
The JVC scheme comes after the RIAA and what is signed as "EQ" is an internal pre amp to lift the signal.
(I have 3 amp's from JVC) In 5436 and GX500 the RIAA is on a separate circuit board, In 5446 the RIAA is on the demod board.

 

[jupp369]

...
The problem is that the recording medium would have to have a frequency response out to 50KHz

RIAA is not needed for just certain .....

as I said: from win7 on the max input freq is limited to 48kHz (/2). Win XP did it well. So I use a cheap external guitar to USB (Focusrite Scarlett 2i2 3rd Gen = 150$) which goes up to 192/24 and have no more problems with hum.
(From win 7 on there is also a limiter for the audio out. For playback it has to be lifted. Realtek can do, my HP's can not)

 

[par4ken]

In the SE-405H scheme "Equalizer" is the RIAA.
The JVC scheme comes after the RIAA and what is signed as "EQ" is an internal pre amp to lift the signal.
(I have 3 amp's from JVC) In 5436 and GX500 the RIAA is on a separate circuit board, In 5446 the RIAA is on the demod board.

The JVC diagram is from their publication "Discrete Four Channel Record System" that describes the operation of CD-4, I haven't investigated their demodulators actual circuitry. RIAA is specified to what, 20 KHz? You would expect it to roll off even more beyond that point but it doesn't have to. Obviously the carrier has to make it through! Seems an odd why to do things though, but no point in second guessing the designers choices!

 

[Sonik Wiz]

The JVC diagram is from their publication "Discrete Four Channel Record System" that describes the operation of CD-4, I haven't investigated their demodulators actual circuitry. RIAA is specified to what, 20 KHz? You would expect it to roll off even more beyond that point but it doesn't have to. Obviously the carrier has to make it through! Seems an odd why to do things though, but no point in second guessing the designers choices!

Thanks Ken that was a very interesting piece of data I haven't seen before.

II am drawn to the paragraph under Recording Process which says:

"Regarding the difference signal, the ranges lower than 80Hz and higher than 6KHz are frequency modulated (FM) and range between 800Hz and 6KHz is phase modulated (PM)."

Firstly I'll say I think there is a typo there as I think it should say the range between 80Hz and 6KHz is PM.

Now this is just plain crazy to me. Very low and quite high frequency's are FM but the mid-range is PM?

I have always read that the carrier signal was simple FM. What would be the benefit to using two different forms of modulation? Or would the demodulator simply see them as both the same?

Maybe I can't see the forest for the trees? I would appreciate someone knowing more than me explain this!

 

[par4ken]

Yes the carrier is actually a combination of FM-PM-FM as to why without further research I can only guess, maybe it is desirable not to vary the carrier frequency too much? FM and PM are related, but why the combination of both?

Perhaps there is an explanation in"Quadraphile", or in the patent document. So sad that Lou Dorran passed on, he could've given us the answer I'm sure!

 

[aludra]

It was a modulation combination for the lowest distortion possible in CD-4.......document from "A better understanding of CD-4"