CD-4 - The thin line between success and failure, but still in the fight!

QuadraphonicQuad

Help Support QuadraphonicQuad:

J. PUPSTER

πŸ’ΏπŸ• Senior Disc Chaser 🎸
QQ Supporter
Joined
May 30, 2017
Messages
11,050
Location
CALIFORNIA (CENTRAL)
Had a little thought this morning about that 30K carrier signal. Would like some discussion here about this as it pertains to lower Hz ranged carts and how they manage to get a good Demod. out. Obviously a cart like the AT440Ml series, topping out at a claimed 24-27K(?) is reaching out over 30K for the Demod. to happen right?

One thing I wondered about, is what does that 30K signal sound like :) Well since most human hearing is limited to around 20K IINM (my own now at my age at around 16K); I thought I'd try something to capture it. So I took someone's clean AT440Mlb Demod. of a song & did some Low Pass and High Pass filtering to Isolate it out from the other audio, then changed the Pitch to around 3.6-4K so it could be heard and boosted it 40dbs.

I'm now wondering if anyone here or elsewhere has done any extensive testing on this as to CD-4 Cart. effectiveness. I'm thinking if the Cart./Stylus is marginal, maybe there'd be the carrier signal un-locking and causing some either disruption in the overall demodulation/separation, or sandpaper, etc? Maybe digitally capturing the same song with various Carts, and analyzing the sound via Spectrogram would reveal any pertinent data on Cart. effectiveness.

Here's what the original file looks like unaltered in a Spectrogram view in Audacity:

CD-4 30K SPECTRO..jpg



Here's what it looks like after the filters, pitch change, and db increase:

CD-4 30K Isolated SPECTRO..jpg


and here's a 10 second sound clip of what it sounds like (sounds like Christmas Jingle Bells and Crickets to me):
 

Attachments

  • CD-4 - 30K ISO (W-LPF&HPF&CP2 4K+40dbs).wav
    1.7 MB

Doug G.

2K Club - QQ Super Nova
Joined
Jul 8, 2007
Messages
2,596
Location
Austin, Minnesota
We have discussed, before, how manufacturers most likely publish only specifications they can guarantee and that's why some cartridges can deliver sufficient output of a 30kHz signal to light the radar. They obviously HAVE enough output there or the radar wouldn't light. Some cartridges obviously have output at even higher frequencies than those specified (AT440MLa/b, etc) or they would not work for CD-4.

And, of course, as has also been discussed, lighting the radar is only the beginning. Satisfactory output and separation up to at least 40kHz, ideally 50kHz, is necessary.

There is really no definitive way to predict whether any given two channel cartridge will work for CD-4 without analysis like you did there.

Unreliable output at those frequencies would, indeed, cause the problems you mention and do.

Doug
 
Last edited:

Sonik Wiz

πŸ‘‚ 500 MPH EARS πŸ‘‚
QQ Supporter
Joined
May 30, 2005
Messages
3,599
Location
Kansas City
Had a little thought this morning about that 30K carrier signal. Would like some discussion here about this as it pertains to lower Hz ranged carts and how they manage to get a good Demod. out. Obviously a cart like the AT440Ml series, topping out at a claimed 24-27K(?) is reaching out over 30K for the Demod. to happen right?

One thing I wondered about, is what does that 30K signal sound like :) Well since most human hearing is limited to around 20K IINM (my own now at my age at around 16K); I thought I'd try something to capture it. So I took someone's clean AT440Mlb Demod. of a song & did some Low Pass and High Pass filtering to Isolate it out from the other audio, then changed the Pitch to around 3.6-4K so it could be heard and boosted it 40dbs.

I'm now wondering if anyone here or elsewhere has done any extensive testing on this as to CD-4 Cart. effectiveness. I'm thinking if the Cart./Stylus is marginal, maybe there'd be the carrier signal un-locking and causing some either disruption in the overall demodulation/separation, or sandpaper, etc? Maybe digitally capturing the same song with various Carts, and analyzing the sound via Spectrogram would reveal any pertinent data on Cart. effectiveness.

Here's what the original file looks like unaltered in a Spectrogram view in Audacity:

View attachment 83209


Here's what it looks like after the filters, pitch change, and db increase:

View attachment 83211

and here's a 10 second sound clip of what it sounds like (sounds like Christmas Jingle Bells and Crickets to me):
Yup, sounds like jingle bells & crickets to me as well!
Now when you say you took this from a "clean demod", I'll point out if it has been demodulated already there is no carrier signal to capture or analyze.

And to reiterate what @Doug G. said, just going to 30kHz isn't good enough. As I said elsewhere that 30kHz carrier signal is frequency modulated & bounces up & down a freq range of 18kHz to 45 kHz & that's why a bit extra to 50kHz is beneficial. There is no way to verify this on modern cartridges so you look at the specs & give it a try.

I'm pretty sure no one but you has tried this approach before. Not sure if it has any value.

And another note on CD-4 bandwidth... base band this time. It's commonly mentioned that the basic music bandwidth of CD-4 is 15kHz. Seems not great but pretty good since most will accept FM radio with the same high end cut off as ok. But when that nominal 15kHz spec is mentioned it is already -3dB down. That's just the way the spec is measured. And it is rolled off at -18dB/octave from there.So probably CD-4 is only flat to 12>13kHz. Just sayin'.
 

J. PUPSTER

πŸ’ΏπŸ• Senior Disc Chaser 🎸
QQ Supporter
Joined
May 30, 2017
Messages
11,050
Location
CALIFORNIA (CENTRAL)
Yup, sounds like jingle bells & crickets to me as well!
Now when you say you took this from a "clean demod", I'll point out if it has been demodulated already there is no carrier signal to capture or analyze.
OK, so what does that horizontal line right at 30K in the Spectrogram represent if not the carrier signal; some kind of distortion from it?
 

Sonik Wiz

πŸ‘‚ 500 MPH EARS πŸ‘‚
QQ Supporter
Joined
May 30, 2005
Messages
3,599
Location
Kansas City
OK, so what does that horizontal line right at 30K in the Spectrogram represent if not the carrier signal; some kind of distortion from it?
Dunno. Thinking about it. Maybe someone else has some insight? Perhaps query thePSpatial folks who have some good experience with CD-4 on the computer.

I think it would have greater validity if you were to use an UN-demodulted signal to analyze. From the phono cart to a MOTU and bump the gain up a bunch & bandpass filter ( 18kHz to 50kHz?), then pitch shift to make it audible.
 

fredblue

2K Club - QQ Super Nova
QQ Supporter
Joined
Mar 30, 2009
Messages
26,960
Location
London, England
Had a little thought this morning about that 30K carrier signal. Would like some discussion here about this as it pertains to lower Hz ranged carts and how they manage to get a good Demod. out. Obviously a cart like the AT440Ml series, topping out at a claimed 24-27K(?) is reaching out over 30K for the Demod. to happen right?

One thing I wondered about, is what does that 30K signal sound like :) Well since most human hearing is limited to around 20K IINM (my own now at my age at around 16K); I thought I'd try something to capture it. So I took someone's clean AT440Mlb Demod. of a song & did some Low Pass and High Pass filtering to Isolate it out from the other audio, then changed the Pitch to around 3.6-4K so it could be heard and boosted it 40dbs.

I'm now wondering if anyone here or elsewhere has done any extensive testing on this as to CD-4 Cart. effectiveness. I'm thinking if the Cart./Stylus is marginal, maybe there'd be the carrier signal un-locking and causing some either disruption in the overall demodulation/separation, or sandpaper, etc? Maybe digitally capturing the same song with various Carts, and analyzing the sound via Spectrogram would reveal any pertinent data on Cart. effectiveness.

Here's what the original file looks like unaltered in a Spectrogram view in Audacity:

View attachment 83209


Here's what it looks like after the filters, pitch change, and db increase:

View attachment 83211

and here's a 10 second sound clip of what it sounds like (sounds like Christmas Jingle Bells and Crickets to me):

very interesting findings Mister Pupster! πŸ₯°

fwiw, in trying out the AT440MLb on my newly-acquired SL-10, its been rather fascinating so far, so i'll share my experiences with you!

the following things have come to light;

1.) the (supposedly fabled?) EPS-310MC that came with the turntable, despite having a rated upper frequency response of 60kHz and seemingly separating stuff out rather well was noisy throughout, in a kind of wooshy/white noise way the whole time way i've never encountered with CD-4 before, so i'd say maybe its not suitable for CD-4 and as its a nude Elliptical rather than a Shibata or MicroLine maybe that shouldn't come as a big surprise?,

2.) i tried out 2 x AT440MLb's (2 x different bodies and styli) and this is where things started to get weird.. the older and i guess more worn 440 sounded much cleaner with CD-4 than the one i bought more recently - and even weirder, this is the exact opposite situation when i run these carts on my SL-1210! on the SL-1210 the newer AT440 sounds a lot cleaner (!),

3.) Capacitance of the RCA leads between the turntable and the Demodulator is a much bigger deal than i imagined! my SL-1210's phono cables are hardwired (and said to be the originals which were lower capacitance than on later production runs) so its not something i ever questioned. however, now swapping out RCA cables on the SL-10 has a real impact!

so far, the thickest and shortest cable i've tried (which funnily came with the Ortofon MCA-76 pre-pre-amplifier, its approximately 1/2 metre long and thick) is the best and is giving the cleanest demodulated CD-4 i have ever heard, with barely any noise and interference despite not even having the ground cable connected (!!) now by stark contrast if i disconnected the ground on the SL-1210 there'd be noise galore!

well, the rest, as they say, is "to be continued..." πŸ‘€πŸ§πŸ˜‹
 

J. PUPSTER

πŸ’ΏπŸ• Senior Disc Chaser 🎸
QQ Supporter
Joined
May 30, 2017
Messages
11,050
Location
CALIFORNIA (CENTRAL)
very interesting findings Mister Pupster! πŸ₯°

fwiw, in trying out the AT440MLb on my newly-acquired SL-10, its been rather fascinating so far, so i'll share my experiences with you!

the following things have come to light;

1.) the (supposedly fabled?) EPS-310MC that came with the turntable, despite having a rated upper frequency response of 60kHz and seemingly separating stuff out rather well was noisy throughout, in a kind of wooshy/white noise way the whole time way i've never encountered with CD-4 before, so i'd say maybe its not suitable for CD-4 and as its a nude Elliptical rather than a Shibata or MicroLine maybe that shouldn't come as a big surprise?,

2.) i tried out 2 x AT440MLb's (2 x different bodies and styli) and this is where things started to get weird.. the older and i guess more worn 440 sounded much cleaner with CD-4 than the one i bought more recently - and even weirder, this is the exact opposite situation when i run these carts on my SL-1210! on the SL-1210 the newer AT440 sounds a lot cleaner (!),

3.) Capacitance of the RCA leads between the turntable and the Demodulator is a much bigger deal than i imagined! my SL-1210's phono cables are hardwired (and said to be the originals which were lower capacitance than on later production runs) so its not something i ever questioned. however, now swapping out RCA cables on the SL-10 has a real impact!

so far, the thickest and shortest cable i've tried (which funnily came with the Ortofon MCA-76 pre-pre-amplifier, its approximately 1/2 metre long and thick) is the best and is giving the cleanest demodulated CD-4 i have ever heard, with barely any noise and interference despite not even having the ground cable connected (!!) now by stark contrast if i disconnected the ground on the SL-1210 there'd be noise galore!

well, the rest, as they say, is "to be continued..." πŸ‘€πŸ§πŸ˜‹
Love it!

You gonna try something like an AT12-20 with it?
 

Doug G.

2K Club - QQ Super Nova
Joined
Jul 8, 2007
Messages
2,596
Location
Austin, Minnesota
The only way I can think of that a 30kHz signal would show at the output of a demodulator is if the demodulator, itself, is not suppressing the carriers or those frequencies are somehow otherwise leaking into the output, something not impossible once you get into those frequencies. Normally, you would never know.

But yes, to do any comparisons of different cartridges and their outputs at 30kHz, you need to scope the signals before demodulation/suppression.

Doug
 

Quad D

Active Member
Joined
Aug 10, 2022
Messages
61
Location
NYC
I think this cartridge behavior is very dependent on the demodulator, so personal experiences might vary a lot depending on the hardware used. Also, the RCA and tonearm cable capacitance has a large impact, of course that also in combination with the demodulator. So, one CD-4 user can find a lot of non CD-4 spec cartridges that work mostly fine, while another user using different equipment can struggle even with CD-4 approved cartridges. But, I can't say I tested this theory myself, because I used only two demods, and both made by the same brand.
 

jupp369

Member
Joined
May 9, 2021
Messages
36
Location
Berlin
I used different carts for recording: VM35F (eliptical/50kHz), AT 20 sla, panasonic epc-450, Nagaoka JT-322 and the results look similar, depending on their age.
(hope i can add some pics and wav here)
I'm recording after the Riaa demodulator with the Panas.450 or JVC 5436 direct to an XP-PC line-in or now via a Scarlett 2i2 to USB at Win8 (win8 limits input frequency to 2x24kHz).
Sounds good, no problems
 

Attachments

  • CCR1.gif
    CCR1.gif
    89.1 KB · Views: 0
  • CCR2.gif
    CCR2.gif
    56.8 KB · Views: 0
  • DDavis Caribe A1 Marianne 15sec.wav
    16.5 MB

Quad D

Active Member
Joined
Aug 10, 2022
Messages
61
Location
NYC
Top