Thanks Hachiman for link. They talk about Guido's nonos dac, PCM1704 running on 16 bit too.
Rfbrw, have you done something to reduce jitter in clock signal out of PMD? Only curious. But problably not, you don't like reclocking if i am not mistaken.
The pdf of universal logic doesn't invert clock into shift chips.
Listened to the experimental cdp i have with 1541 nonos for a moment this afternoon. It can not reach the clarity of the pcm56's. There seems to be more noise in the music too, like a slight rumble, but then around tones.
Rfbrw, have you done something to reduce jitter in clock signal out of PMD? Only curious. But problably not, you don't like reclocking if i am not mistaken.
The pdf of universal logic doesn't invert clock into shift chips.
Listened to the experimental cdp i have with 1541 nonos for a moment this afternoon. It can not reach the clarity of the pcm56's. There seems to be more noise in the music too, like a slight rumble, but then around tones.
PMD100 is not easy to find:
http://cgi.ebay.com/Pacific-Microso...70669875QQihZ009QQcategoryZ4663QQcmdZViewItem
http://cgi.ebay.com/Pacific-Microso...70669875QQihZ009QQcategoryZ4663QQcmdZViewItem
Have some questions:
What is the maximum voltages that the PCM56 is to be run at?
What's the voltage swing when using passive I/V using say 560R?
What happens if I feed the DAC with 24bit data?
AD1851 and AD1861 (18bit) are pin compatible. Anyone tried them? Good idea to use sockets
Anyone settled for a 'definitive' glue logic I2s 'converter'?
What is the maximum voltages that the PCM56 is to be run at?
What's the voltage swing when using passive I/V using say 560R?
What happens if I feed the DAC with 24bit data?
AD1851 and AD1861 (18bit) are pin compatible. Anyone tried them? Good idea to use sockets
Anyone settled for a 'definitive' glue logic I2s 'converter'?
Right now i am listening to it, it has still some very small artefacts with beginning of tracks, at the track itself it dissapears. I used the universal shifter, recommended, for universality.
http://www.geocities.com/nonospcm1704/universal_shifter.html
For other dac tests i only have swap the shift wires.
I havent shielded the logic, nor decoupled the logic chips yet!
But sounds good.
Play a cd with a 'crowded' track in which many voices are recorded. CD Doe maar, Klaar, track 12
For the first time i can understand what they say:
"Het is wel druk hier. Laten we ergens anders gaan zitten."
Btw tried an other cdp as base, maybe this could declare the previous problem. Will swap clocks, now use Tents XO.
For the first time i can understand what they say:
"Het is wel druk hier. Laten we ergens anders gaan zitten."
Btw tried an other cdp as base, maybe this could declare the previous problem. Will swap clocks, now use Tents XO.
I can correct now easily MSB on ear with low level signal music by means of headphone. The digital noise dissapears, and there is the best adjust point. But the Right channel i can not get completely silent, the pot is at his end. Another PCM56 dac could be tried for R, and see if this one is better adjustable for MSB.
Indeed the MSB adjust is very very important, no adjust gives low level distorsion.
The culted TDA1541 doesn't have MSB adjust, is stereo, so it must be hard to select a good sounding stereo pair, lots of dac chips has to be selected and small percentage of them will pass.
Have a handfull of 1541's and should sell them in future.
The XO gives an analytic sound, but can be slightly sharp in treble. Experienced that in previous cdp's. Kwak 7 was more analog sounding, will try Kwak on the PCM56's
Indeed the MSB adjust is very very important, no adjust gives low level distorsion.
The culted TDA1541 doesn't have MSB adjust, is stereo, so it must be hard to select a good sounding stereo pair, lots of dac chips has to be selected and small percentage of them will pass.
Have a handfull of 1541's and should sell them in future.
The XO gives an analytic sound, but can be slightly sharp in treble. Experienced that in previous cdp's. Kwak 7 was more analog sounding, will try Kwak on the PCM56's
tubee said:
Only use the chips marked 63. The other two are not good and one is outside the adjustment range, those were only to hear the difference. Best throw them away.
The adjustment range can be made more wide or narrow or shifted by changing resistor and pot values.
I think it is best to make the pot as small as possible to avoid temperature drift.
Also a temperature stable voltage reference should be used for the supply.
Parallel chips and using MSB on one chip only will further reduce drift.
Hi Bernard
I have only two PCM56 dacs, not four, and are not stamped.
I swapped the dacs and the very small distorsion stays at R, so the regulators and/or caps could cause it. IV i swapped too, no difference. But as i said this distorsion is very small, L is very clean and i want both channels clean. Have to decouple each logic chip also. Was surprised it worked allmost immediately, after correcting 2 fault this morning, yesterday i was soldering with a fine glass of red wine
Will add the dacs beside the logic on the perfboard pcb and keep room for some extra PCM's
The universal logic has also a provision to feed dacs differential (Q-not output of flipflop) This could give another 6dB or so noise cancelling imo. But have to find out about I/V then, i want to use a passive I/V resistor and tube stage, have parts allready.
Bernard, did you made the MSB adjust schematic for only 1 of the four dacs? Would it be possible to adjust each dac, and listen/adjust each dac by a switch or jumper? Then after all 8 dacs each are adjusted, they can be paralled again. Idea?
I have only two PCM56 dacs, not four, and are not stamped.
I swapped the dacs and the very small distorsion stays at R, so the regulators and/or caps could cause it. IV i swapped too, no difference. But as i said this distorsion is very small, L is very clean and i want both channels clean. Have to decouple each logic chip also. Was surprised it worked allmost immediately, after correcting 2 fault this morning, yesterday i was soldering with a fine glass of red wine
Will add the dacs beside the logic on the perfboard pcb and keep room for some extra PCM's
The universal logic has also a provision to feed dacs differential (Q-not output of flipflop) This could give another 6dB or so noise cancelling imo. But have to find out about I/V then, i want to use a passive I/V resistor and tube stage, have parts allready.
Bernard, did you made the MSB adjust schematic for only 1 of the four dacs? Would it be possible to adjust each dac, and listen/adjust each dac by a switch or jumper? Then after all 8 dacs each are adjusted, they can be paralled again. Idea?
Guido, have you bought this lot?http://cgi.ebay.com/Pacific-Microso...70669875QQihZ009QQcategoryZ4663QQcmdZViewItem (i ask because of your smile)
tubee said:
Ups, I recently sold a pair of selected PCMs + two bad ones.
I thought it was you...
Very possible one of your PCMs is outside the range.
Have seen that even on K's.
tubee said:
I made MSB adjust only for one DAC, all 4 together behave like one, MSB currents are added and adjustment range is only 1/4 of normal.
Adjust every single DAC is useless.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.