ELP Turntable - Ask Me Anything!

[Jerfo]

Argh. It looks like I can't edit my initial post any more, so here's an update:

The ELP can actually play Quiex II pressings. These are high-quality vinyl pressings that are all slightly translucent - they're usually a translucent brown or even purple when viewed through a strong light source. To my surprise, the ELP handles them just fine!

I'm listening to the Quiex II pressing of Peter Gabriel's "Security" album and it sounds great!

 

[dghan]

Since the CD-4 information can approach 50 kHz, you'll need to sample higher than 96 kHz. I have a demo copy of Stereo Lab and I plan to digitize at 24-bits/192 kHz to try it out. For 96 kHz, the maximum resolved frequency is 48 kHz and the signal-to-noise ratio near 48 kHz is getting higher than the lower frequencies, maybe too high to decode well. Also at 192 kHz, the ADC anti-aliasing filter will have less phase distortion at 50 kHz. For QS and SQ LPs, I plan to sample at 96 kHz.

 

[dghan]

Maybe off-topic, but regarding Quiex LPs: I have copies of LPs that have normal and Quiex pressings and I can't tell the difference. I think the advantage of Quiex is the pressing is more likely to be quiet, whereas a normal pressing is more likely to have noise because of the recycled vinyl used at the time Quiex was introduced, but normal pressings can be as quiet as Quiex. Quiex uses virgin vinyl. Has anyone else noticed this?

 

[Jerfo]

Since the CD-4 information can approach 50 kHz, you'll need to sample higher than 96 kHz. I have a demo copy of Stereo Lab and I plan to digitize at 24-bits/192 kHz to try it out. For 96 kHz, the maximum resolved frequency is 48 kHz and the signal-to-noise ratio near 48 kHz is getting higher than the lower frequencies, maybe too high to decode well. Also at 192 kHz, the ADC anti-aliasing filter will have less phase distortion at 50 kHz. For QS and SQ LPs, I plan to sample at 96 kHz.

When I first tested the Stereo Lab software, I actually tried 24/192 at first but the software itself insisted on 24/96. This was a year ago, so it's possible that they've changed things since then.

 

[Plan9]

Maybe off-topic, but regarding Quiex LPs: I have copies of LPs that have normal and Quiex pressings and I can't tell the difference. I think the advantage of Quiex is the pressing is more likely to be quiet, whereas a normal pressing is more likely to have noise because of the recycled vinyl used at the time Quiex was introduced, but normal pressings can be as quiet as Quiex. Quiex uses virgin vinyl. Has anyone else noticed this?

My experience as well: I have had Quiex LPs noisier than normal LPs.

 

[lukpac]

Since the CD-4 information can approach 50 kHz, you'll need to sample higher than 96 kHz. I have a demo copy of Stereo Lab and I plan to digitize at 24-bits/192 kHz to try it out. For 96 kHz, the maximum resolved frequency is 48 kHz and the signal-to-noise ratio near 48 kHz is getting higher than the lower frequencies, maybe too high to decode well. Also at 192 kHz, the ADC anti-aliasing filter will have less phase distortion at 50 kHz. For QS and SQ LPs, I plan to sample at 96 kHz.

Have you tried it at 96 kHz? I haven't, but looking now, the carrier frequency is centered at 30 kHz, and it appears the crossover between the the audible spectrum and the FM modulation is about 18.5 kHz or so. With that in mind, I'm wondering if anything close to 48 kHz is actually necessary for decoding.

 

[dghan]

Have you tried it at 96 kHz? I haven't, but looking now, the carrier frequency is centered at 30 kHz, and it appears the crossover between the the audible spectrum and the FM modulation is about 18.5 kHz or so. With that in mind, I'm wondering if anything close to 48 kHz is actually necessary for decoding.

If the low end is 18.5 kHz, then the high end has to be 11.5 kHz above the carrier, so 41.5 kHz. Wikipedia states that the range is 18 to 45 kHz with a 30 kHz carrier, which might not be correct. Sampling at 96 kHz might work well as long as the SNR doesn't get too high just because there are so few bits to represent frequencies in the 24 to 48 kHz range, the sweet spot for the CD-4 information. That's why I would use 192 kHz, but maybe Stereo Lab doesn't support that for their CD-4 decoders. I haven't found anything regarding this and I haven't tried the demo yet.

 

[DuncanS]

Have you tried it at 96 kHz? I haven't, but looking now, the carrier frequency is centered at 30 kHz, and it appears the crossover between the the audible spectrum and the FM modulation is about 18.5 kHz or so. With that in mind, I'm wondering if anything close to 48 kHz is actually necessary for decoding.

There will need to be frequency 'space' between the carrier and the frequency modulated channels as the analogue filters used in the encoding (& old decoding) would need roll-off. So if I remember correctly CD4 only has 15kHz bandwidth, and you'd need a few kHz for the filter roll-off so you would need the full 48kHz.

 

[lukpac]

There will need to be frequency 'space' between the carrier and the frequency modulated channels as the analogue filters used in the encoding (& old decoding) would need roll-off. So if I remember correctly CD4 only has 15kHz bandwidth, and you'd need a few kHz for the filter roll-off so you would need the full 48kHz.

I'm not sure what you mean. Isn't the FM signal centered around the 30kHz carrier? Or is there more signal above 30kHz than there is below?

 

[DuncanS]

The audio was modulated onto the carriers using FM-PM-SSBFM, Frequency Modulation/Phase Modulation Single Side Band FM, and if I remember correctly its not quite symmetric around the 30kHz carrier as the FM signal range was 18kHz to 45kHz +/-?dB. The more available bandwidth the 'easier' it is to demodulate, so better to go above 45kHz, the encoding/modulation filters wouldn't have had hard cut-offs so the signal should be there.

 

[Jerfo]

BIG Update:

Short Version:

The ELP can playback CD-4 LPs with all of the high frequency information in tact.

Long Version:

Boy, did it take me a while to get to this point. Back when I ordered my ELP in late April 2020, I ordered a passive inverse RIAA board from a company called Motronix out of Israel. (For those of you who haven't read this entire thread, my turntable only supplies a line out, but CD-4 dedmodulators need a phono level signal to work their magic.) Due to COVID-19 restrictions, the board was taking an inordinate amount of time to get to me, so I tried a number of workarounds that are documented elsewhere in this thread. None of the workarounds was particularly satisfying to me, so I decided to take matters into my own hands and build my own passive inverse RIAA adapter.

Fortunately, I'm not the only person crazy enough to have attempted this, so someone else had already gone through the trouble of creating the parts list and the gerber files for the PCB: Side Project: Stereo Accurate Inverse RIAA from HIFISonix - Build - Muffsy Phono Kits . I ordered PCBs from PCBWay (who manufactured them extremely quickly and inexpensively) and ordered the caps, resistors, and miscellaneous parts from Mouser. Once the first wave of parts came in, I started assembling my board....no trivial task, since it involves surface mount soldering of many teeny tiny parts with my very non-steady hands. You all can probably guess what happened next:

The original fully-assembled board that I ordered in April finally arrived from Israel, more than two months from my original order date.

So, I excitedly hooked it up to my Marantz demodulator and shook with joy when the ol' RADAR light lit up like a Christmas tree. Alas, my enthusiasm was short-lived as things were not quite right. Separation on the left channels was good, but non-existent on the right channels. I was eventually able to determine (after much hair-pulling and swearing) that the Motronix anti-RIAA board was allowing the 30k carrier to pass through its left channel, but not its right. Big bummer!

After licking my wounds, I got back to my workbench and completed the assembly process of my homemade anti-RIAA adapter. Once complete, I hooked it all up to my Marantz demodulator and...hmmmm...something still isn't right. There's separation, but not to the degree I'd expect. No amount of adjustment is making a tangible improvement. In a last ditch effort, I connected everything up to my JVC demodulator that I thought was broken...and it worked! Channel separation is good, and overall sound quality is good. I've listened to a handful of CD-4 LPs so far and they all seem to be working and sounding great.

So, can an ELP laser turntable be used to playback CD-4 records with the high-frequency carrier information in tact? YES!

 

[furui_suterioo]

BIG Update:

Short Version:

The ELP can playback CD-4 LPs with all of the high frequency information in tact.

Long Version:

Boy, did it take me a while to get to this point. Back when I ordered my ELP in late April 2020, I ordered a passive inverse RIAA board from a company called Motronix out of Israel. (For those of you who haven't read this entire thread, my turntable only supplies a line out, but CD-4 dedmodulators need a phono level signal to work their magic.) Due to COVID-19 restrictions, the board was taking an inordinate amount of time to get to me, so I tried a number of workarounds that are documented elsewhere in this thread. None of the workarounds was particularly satisfying to me, so I decided to take matters into my own hands and build my own passive inverse RIAA adapter.

Fortunately, I'm not the only person crazy enough to have attempted this, so someone else had already gone through the trouble of creating the parts list and the gerber files for the PCB: Side Project: Stereo Accurate Inverse RIAA from HIFISonix - Build - Muffsy Phono Kits . I ordered PCBs from PCBWay (who manufactured them extremely quickly and inexpensively) and ordered the caps, resistors, and miscellaneous parts from Mouser. Once the first wave of parts came in, I started assembling my board....no trivial task, since it involves surface mount soldering of many teeny tiny parts with my very non-steady hands. You all can probably guess what happened next:

The original fully-assembled board that I ordered in April finally arrived from Israel, more than two months from my original order date.

So, I excitedly hooked it up to my Marantz demodulator and shook with joy when the ol' RADAR light lit up like a Christmas tree. Alas, my enthusiasm was short-lived as things were not quite right. Separation on the left channels was good, but non-existent on the right channels. I was eventually able to determine (after much hair-pulling and swearing) that the Motronix anti-RIAA board was allowing the 30k carrier to pass through its left channel, but not its right. Big bummer!

After licking my wounds, I got back to my workbench and completed the assembly process of my homemade anti-RIAA adapter. Once complete, I hooked it all up to my Marantz demodulator and...hmmmm...something still isn't right. There's separation, but not to the degree I'd expect. No amount of adjustment is making a tangible improvement. In a last ditch effort, I connected everything up to my JVC demodulator that I thought was broken...and it worked! Channel separation is good, and overall sound quality is good. I've listened to a handful of CD-4 LPs so far and they all seem to be working and sounding great.

So, can an ELP laser turntable be used to playback CD-4 records with the high-frequency carrier information in tact? YES!

Do you change the groove depth adjustment for CD-4, or does it make a difference?

 

[Jerfo]

Do you change the groove depth adjustment for CD-4, or does it make a difference?

It depends on the record. Sometimes it makes a significant different, and other times it makes no difference whatsoever.