Home Digital Drive

Upsamplers, DACs, jitter, shakes and analogue withdrawals, this is it.

Re: Asynchronous Reclocking in the DAC

Hi protege of herbe,
I have a hard time trying to understand what you mean. Do you mean by clocking replacing the crystal oscillator by a better one or do you mean bringing the clock from the CDP to the DAC or vice versa?? In any case I absolutely don't see why this will reduce the life of the player. Why would this create extra work for the servo's?
Basically there are three servo's in a CDP; one for focus; keeping the laser in focus. One for tracking; keeping the laserbeam on the track and one for the carriage; shifting slowly the whole laser from the center of the disc to the outher edge. The latter servo is a lowpassed signal from the tracking. If you put on a very scratchy disc that is not flat too, the servo's have to work hard but how reclocking comes in the play.....?

Anyway I would like to draw attention to a very simple scheme called asynchronous reclocking.
You don't have to make a extra connection from the CDP to the DAC for the clock signal.
It is simply done with 74AC74 D-type flip-flops and I did not invent it. All the outputs from the CS8412 are reclocked in the D-flip-flop. with a 45.158 MHz low jitter clock. I found the higher the reclockingfrequency the better, more analog the sound is. Ryohei Kusunoki http://www.sakurasystems.com appears to use a 50 MHz clock. I will go to 100 MHz or even 125 MHz but first have to build a clock with such a high frequency.
I just finished the 45 MHz clock {schematic available on request by sending me a email} as this will also come in handy for Sony players. In fact I used the crystal from my Sony.
While others are struggling with extra PLL's 74HC4046, FIFO's etc. this scheme only needs a clock and a D-flipflop. If you can get the 47AC175 you only need one of this.
There appears to be not much concensus which signals to reclock, Ryohei appears only to reclock the latch while in a article in the Japanese MJ Magazine all three signals are reclocked except the DATA.
Credits go to Bob Fitz. and Barend who already in June 2000 saw the benefits of reclocking and Bob noticed that the clock does not need to be synchronous.
Aha Bob, in rereading your post I see why they don't reclock the DATA; it makes no sense. Below is the link to the DIYaudio.com forum where a schematic is found and how to connect the 45 MHz or higher clock.(It is quite unlogical posted in the Pass Labs forum). In the future I will post more details there.
Please note that my results are obtained with a low jitter KWAK-CLOCK in the transport, CS8412, Wildmonkeysects loopfilter and non oversampling AD1865 in the KWAK-DAC.

This post is made possible by the generous support of people like you and our sponsors:
  Michael Percy Audio  

Follow Ups Full Thread
Follow Ups

You can not post to an archived thread.