Finally, an affordable CD Transport: the Shigaclone story

While the transport I'm presenting in this thread is not really a clone, it all started with me picking inside Shigaraki and posting some pics on audiostereo.pl forum.

As those guys are quite resourceful, they quickly found that the same mechanism and similar chips are used in JVC RC-EZ31 boombox and the rest is history.

Now, after all those months, I'm quite puzzled why nobody so far came up with similar application? After all, Gainclone and Phonoclone gained so much popularity that easily surpasses the originals.

It all started when last September somebody asked me to build a PS for Shigaraki transport. I did that and had a chance to play with the transport in my own system. And while I was extensively using CEC TL0, ML31.5 and ML37 at that time, I was quite impressed by Shigaraki. In some ways it offered performance virtues none of my other transports could dream of, specifically the freshness and vitality of the sound, with very well defined PRAT (BTW, the PS based on 70VA Hammond transformer was in no way inferior to original Dumpty).

I was almost ready to buy a used Shigaraki or Flatfish when the alternative option emerged on Polish Forum. And although I was quite involved during all those months in development process, the discovery of the transport source and some fine points of modifications should be credited to other forum members there.

Now, after 6 months, my transport is almost completed and it took the spot of CEC TL0, becoming my main digital reference (I still keep TL0, both Levinsons and loaned Flatfish for comparisons).

The whole story can be read here: http://www.audiostereo.pl/forum_wpisy.html?temat=36475&p=1#k

.
An externally hosted image should be here but it was not working when we last tested it.



While I was quite hesitant to post this info here, I think that discovery is too good to be kept secret, with time I will provide more details.
I used Shigaclone reference in the thread topic, but after some time will ask moderators to remove it.
 

Attachments

  • cdp.jpg
    cdp.jpg
    87.8 KB · Views: 37,746
Administrator
Joined 2004
Paid Member
Looks very interesting. I'm looking for a small, inexpensive transport to use with my Zhaolu which will soon be retired from my main system and put to use elsewhere. This could be the right project..
:D

I'm going to look for one of those JVC boom boxes, anything else you might recommend?

I will tweek against my modded Sony SCD-777ES..
 
with peter's help and the photos from the polish forum i have also put together this CD player.

i was previously using a NADC541i as a transport. I found the modified JVC to be a significant improvement over the NAD. i was quite surprised by this result as i did not expect a change in the transport to have an audible effect.

The cost of all the electronic components comes in under 200 USD making it a relative deal when compared to other DIY digital source options.

I made a PDF documenting the changes, with peters permission and some updating i will post here.
 
Sure, that PDF will be helpful

markushbm said:
I will be patiently waiting, until it`s ready.

It is ready and don't wait. All that's left to do in my transport is to nicely install the display board.

Who knows how long that JVC model will be available (the higher models, like EZ35 are no good). I am certainly stocked well;)
 

Attachments

  • boxes.jpg
    boxes.jpg
    97.1 KB · Views: 37,053
Here's the PS schematic. I tried some popular regulators (LM317, LM2937, LT1086, LT1129) as well dual regulation and separate regulation for 5V. Nothing worked better for me than a single LM7808 from National.

I also tried few diode types, including 31DQ09 Schottkies that are supposedly used in Zanden, but again, MSR860 worked best.

Using BG STD 1000/25 after regulator is quite critical, before regulator you see BG FK 2200/35. I later switched to not available any longer BG F 1000/35 which provided more high frequency extention. You could use BG STD 1000/25 with BG N 4.7/50 bypass to compensate slight brightness but you may loose some liquidity and naturalness.

Other caps are also OK, feel free to experiment. What I present here is exactly what worked in my system (for me).
 

Attachments

  • ps.jpg
    ps.jpg
    19.5 KB · Views: 33,073
The frame is made of bronze nad copper and weighs 5kg.

I came to this matrial choice by accident, when I decided to try the copper frame I had originally made for CD-Pro. Adding later bronze base and springs improved things further.

The frame you see in the pics is the second generation where I had to compromise few things in order to gain more functionality.

Originally I had the mechanism mounted on acrylic board and later on spruce board.

Bronze/copper and springs add more dimentionality and refinement. Brass is a bit ringy, but depending on application it may worked as well.

The mechanism works best if mounted on two standoffs only, it sounded dull when I tried to hang it underneath brass frame with more contact surface. Here the pics of one prototype that did not work as I expected:

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.
 

Attachments

  • crystal.jpg
    crystal.jpg
    11.2 KB · Views: 17,952
Beside all that, there are few components that I replaced on original board (replacing too many will kill the sonics).

I basicaly removed the main filter cap and replaced all other electrolytics tied directly to rails with BG N caps and also removed their SMD bypasses. One other cap was replaced to ERO 1830, 0.1uF and that's basically all. Okapi will post PDF when it's ready.

One other thing, the output resistors are quite important, I got the best results with Caddock MK132 75R to ground and VishayS102 60R (or similar value) in series.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.