Marantz CD63 & CD67 mods list

Ot

I put the Black Gate AC 6.8uF caps in my preamp's output for DC blocking duties.

They're definately coloured to my ears!! Very nice though, instruments are full of warmth, in particular guitars and vocals sound lovely.

Hopefully the slight hazy/fuzzy quality will reduce with "burn-in".

Already I can safely recommend them above cheap film caps (for example, at output of CD63/7 if needed).

Simon
 
Hi Everybody,

I have installed a low jitter clock from TentLabs (http://www.tentlabs.com/Components/XO/index.html) powered by a "Sjöström Super Regulator" (http://sjostromaudio.com/joomla/index.php?option=com_content&task=view&id=55&Itemid=27) into my new Marantz CD 67 mkII OSE. The player suddenly began to play real stereo, with incredible detailes to be heard. The improvement is obvious, even guests of mine, who know nothing about electronics, confirmed that it sounds better.

I have made some measurements of the player characteristics before and after installing the new clock, using RightMark Audio Analyzer (http://audio.rightmark.org/index_new.shtml).

The original player:

http://www.sinigersky.com/docs/Marantz CD 67 mkII OSE.htm

The player with a TentLabs XO:

http://www.sinigersky.com/docs/Marantz CD 67 mkII OSE reclocked.htm

For these measurements, a M-AUDIO Audiophile USB sound card has been used. For reference, here is the card only:

http://www.sinigersky.com/docs/Audiophile Out 1_2.htm

When looking at the results one can see a worse frequency response after installing the clock, and also slightly larger distorsion and somehow, more hum is now entering the system (100 Hz etc. spikes everywhere).

My theory about the frequency response is that now there is less jitter content in the high range. Would it be a good idea to modify the analog filter at the DAC output in order to compensate this?

I would be glad to read some comments on the results. Has anybody done similar tests?

I will post some pictures of the modification later.
 
SimontY said:
That's better.

Firstly, I wonder why your tests show the steady state measurements to be worse with the reclocking.

Secondly, a jitter measurement would be more meaningful. It won't be reflected in any of those graphs I don't believe, as it is a timing error, not a response/amplitude problem.

Simon


I am wondering too.
Maybe the following is happening: the jitter of the clock is a phase noise which causes signal-related "amplitude" noise. When jitter is improved, the noise content (or "power") disappears from the frequency response plot causing more roll off at the high frequencies.

The hum effect should be something else. I tried to move the cables, to reverse the mains power cord plug, but maybe I'm still forgetting something.

I have no explanation about the slight distortion effects.

I agree that the change should not be visible in such an experiment, I made the comparison because I was curious if the sonic improvement would be measurable in this way.
Unfortunately, I have no equipment to measure real jitter performance.