Elso Clock in CDT or DAC

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Dear Elso,

The X-DAC has a quartz oscillator 16.9344Hz driving Yamaha YM3623B chip. I think the Denon CD player is also equipped with 16.9344Hz quartz oscillator.

I have changed the dual op amps in the X-DAC to AD826, analog caps to 10uf Elna Cerafine, 2200pf Silver Micas and digital caps to 10uf SC Os-con. I have also change the power caps to Blackgate 1000uf and schoktty diodes too. I want to upgrade the clock in X-DAC so as to realise all the potentials with the tweaked items. Is it the right way?

Or you advise me to change the clock in the Denon player instead?

I do not have the Denon CD player's schematic. Can u advise on which caps to take out. Thanks

Patrick
 
KWAK-CLOCK

patricklyw said:
Dear Elso,

The X-DAC has a quartz oscillator 16.9344Hz driving Yamaha YM3623B chip. I think the Denon CD player is also equipped with 16.9344Hz quartz oscillator.

I have changed the dual op amps in the X-DAC to AD826, analog caps to 10uf Elna Cerafine, 2200pf Silver Micas and digital caps to 10uf SC Os-con. I have also change the power caps to Blackgate 1000uf and schoktty diodes too. I want to upgrade the clock in X-DAC so as to realise all the potentials with the tweaked items. Is it the right way?

Or you advise me to change the clock in the Denon player instead?

I do not have the Denon CD player's schematic. Can u advise on which caps to take out. Thanks

Patrick
Hi Patrick,
I think it is best to change the clock in the Denon (transport).
The clock in the DAC is only used to recognise the frequency and is best switched off thereafter. If I remember right Elector had a scheme to do that.
Personally I would use 2200pF polypropylenes instead of silver mica. I don't have any experience with Cerafines and Oscons so I don't know about that upgrade. Schottky diodes seem OK. You could also try soft or ultrasoft recovery diodes.
To install the clock you have to remove the crystal and the two small caps electrically connected to it. Save the crystal for use in the KWAK-CLOCK.
etc.
:cool:
 
Hi. Elso,

Thanks for your advice. I think you refer to Elektor Electronic magazine. Right? Do you have the schematic or link to show how to switch off the clock in the DAC?

What if I change only the DAC clock? Would it not worth the effort? Just wonder the difference between changing the clock in CD transport and the one in the DAC. Could u tell me more?

Thanks

Patrick
 
Ryssen: It is better to install a clock in the DAC (where it is needed) and then make the nessescary syncronization of the transport.

Installing a clock in the transport only is an easy way out, but will NOT give you the full benefit of your clock investment. In fact it will not give you much improvement, and therefore is not worth while.
 
It's best to put the good clock in DAC and feed that into the CD transport for use there, also. That way you get a low-jitter S/PDIF signal coming in, and synchronicity between clock domains. In addition you have an excellent word clock source in the DAC where it really matters.
 
Just one small problem......

The 16.whatever MHz oscillator does not "drive" the '3623. It only serves as source to feed something downstream when there is no signal to lock on.

You do need that clock frequency, but putting a quality clock to the X_in pin of the '3623 will do nothing. You need to feed it directly to the filter chip AND DAC chips.

Jocko
 
I've read a few times now, about (commercial) modders performing a 16MHz clock upgrade to the YM3623B.:D

Once the YM3623B is locked, it makes sense to disconnect the 16MHz oscilator from the YM3623B with a fet switch for example using the ERR output of the YM3623B.
 
This is my implementation of Jocko's suggestion:

C1 is used for making a small delay of the xtal shutdown. Otherwise you may find the YM3623 sometimes has problems locking on to the SP/DIF.

Operation is simple, when YM3623 locks, pin 21 goes low, T2 stops drawing current, and a voltage builds up on C1. After some mS, the voltage is 0,7V and T1 starts to turn on, tilting the internal oscillator, so it stops.
This circuit works fine, some considerate people would spare the xtal of DC by connecting a small capacitor in series with it (>1nF) but it is not strictly nessescary.
 

Attachments

  • ym3623sd.gif
    ym3623sd.gif
    10.1 KB · Views: 1,043
Ryssen: If you have a 6.144 MHz crystal in your DAC, i would guess it is a CS-8412 or CS-8414 receiver.

In this case your DAC runs 11.2896 MHz, so if your transport runs on 16.9344 MHz, you can (most probably) not sync the two units.

(That is one of the drawbacks of using separate DAC / Transport solutions in the first place - there are benefits too ... :) )

What is the DAC chip in your DAC appliance?
 
If its the 18bit XDAC with HDCD then its a YM3623,PMD100 and PCM67/69 combination. The PCM67/69 will run off a 384Fs SYSCLK but I am not sure about the PMD100. Assuming the PMD100 will run off 384Fs, you have the same oscillator frequency in MF and the Denon. Sounds like a job for Jocko's mystery multiple output circuit.

ray.
 
If the XDAC does not deviate significantly from BB's suggested schematic for the PCM69, then the whole thing probably runs at 16.9MHz with the PMD100 and the PCM69 driven off Pin6 of the YM3623. In that case driving the dac and the transport with the same clock should not be a problem.

ray.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.