1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  2. Things you need to know about the new ‘Conversations’ PM system:

    a) DO NOT REPLY TO THE NOTIFICATION EMAIL! I get them, not the intended recipient. I get a lot of them and I do not want them! It is just a notification, log into the site and reply from there.

    b) To delete old conversations use the ‘Leave conversation’ option. This is just delete by another name.
    Dismiss Notice

Flea and other mods to CDi...

Discussion in 'd.i.y.' started by Dowser, Apr 14, 2010.

  1. martin clark

    martin clark pinko bodger

    2 for 2! Full of win!


    (Cricklewood Electronics, that eminently-reliable supplier, have some late-date TDA1541A in stock for c£25. My bro has just fitted one as a replacement into his Avondale'd CD3 - works a treat)
  2. Dowser

    Dowser Learning to bodge again..

    Thanks Martin - they're a little more now, £30+vat, but I've emailed them to confirm Taiwan date code. Even at that price I will order a few :)

    I rushed re-assembly of #1 and have f*cked something up, a brief spin and almost immediate error. Intermittent initially, now seems constant. Need to strip the mech out and start again with more patience.

    I'm listening to Laverda's now, with some 10uF wet tants op caps to replace the solids, and very nice it is too. Have not re-assembled yet, all balanced on a table under my rack.

    Relax time - I put it back together, and investigate #1, tomorrow.
  3. Cirkus

    Cirkus pfm Member

    Delighted to hear your progress. As stated by Martin a well & encouraging documented journey - warts and all - it gives us all hope for a positive outcome should we have the guts to endure.

    Please keep us up to date with the trials of #1
  4. Mike P

    Mike P pfm Member

    Well done Richard!
  5. Dowser

    Dowser Learning to bodge again..

    Thanks again all - I've actually learned heaps, but it was pretty hard work to stay motivated. Main motivator - always at the back of mind was the question of all CDPs I've heard/modified/played with since, "yes, this is good, but was it as good as my CDI that broke?".

    I haven't progressed with #1 yet, but have left #2 on repeat for last 2 days. Just had the first serious listen, and can say with some certainty I have answered the question - Laverda's old CDI (4 ALWSRs powering final 2 opamps, which have been swapped for OPA627s and is a mod I am not familiar with) is the best sounding CDP I have in my place currently. It really is rather stunning - it has all the strengths of the various players I have but none of the weaknesses.

    I now need to get #1 running again to see how much of this is down to the OP stage mods, but the possibility is creeping in that I keep the 2 CDIs - one for home, one for office :)

  6. Dowser

    Dowser Learning to bodge again..

    Finally looked at #1 today - operator error again... I didn't have 16v caps when I swapped a couple of 47uFs on the servo board earlier, so used 35v items with a higher profile...causing this;


    Not a problem when I only had a lossy connection between mech and servo board (I initially used the 2 plastic spacers that slipped over the screws for temporarily mounting mech to servo while fault finding, and it worked fine - but testing again tonight with the 2 tight fit spacers was enough to have the same issue).

    I simply refitted the original 16v items and it is back to life :)

    Very interesting to compare the 2 CDIs;

    #1 bog standard except for a recap, with a dual crown 1541

    #2 with Flea and dedicated PSU, and 4 ALWSRs powering OPA627s instead of stock OPA42s at output. Against it, it is running an R version of the TDA1541A instead of the normal single/double crown version

    Both have had output caps swapped for 10uF wet tants.

    Objectively, #2 is better in every way. Subjectively I prefer listening to #1 :)

    I will leave #1 on repeat for 24 hours, those wet tants take a while to come on song. Logically, I should swap the TDA1541As between players to confirm how much of a role that is playing, but that's a pretty high risk based approach given my soldering skills :)

    If anyone has a single or double crown 1541A, or a Taiwan date code one, they'd consider selling me, please let me know. I'm in contact with Cricklewood, they are running low and have no Taiwan items, but I will buy one of their later items just to see whether the R version has much of a role to play here.

    I think I am about to mount 2 disassembled CDIs into my main system permanently to allow me to compare further :)

  7. martin clark

    martin clark pinko bodger

    While you can, stick your scope on the output of the ALWSRs and just check they aren't singing at HF (noise at idle with nothing playing)

    Interesting finding, I can quite understand such preferences ( I've had a few such moments myself with other kit) Just shows that listening for pleasure should be exactly that, and not confused with gear lust :)
  8. Dowser

    Dowser Learning to bodge again..

    Thanks - I'm fairly confident that the Flea when I originally fitted it in #1 produced better SQ & a more enjoyable listening experience. That leaves either the TDA1541A-R or the ALWSRs/OPS627s in #2 - hence my questions on the other thread about OPA627 decoupling...currently it just has stock Naim decoupling for the OPA42. I have a whole bunch of new caps to play with when the time is right.

    However, what I do hear with #2 is rock, rock solid imaging that #1 does not have. Plus it has all the subtle nice edges I'd normally associate with the Flea. But there is something that makes it less enjoyable than #1 overall (at a loss to explain why though!) - I'm going to assume the 1541 DAC version until I prove otherwise :) I will scope the ALWSRs output though, thanks.
  9. Dowser

    Dowser Learning to bodge again..

    Hmmm - I never thought I'd be able to do this, 2 functioning CDIs in my system :)




    After further listening, I could easily live with both - differences, for better and worse, are really small. I need to take my time with this :)

    What both do though is relegate these to the storage shelf;

    Who would have thunk it?

    Woodside WS3 and Cambridge Audio CD3 coming to a "For Sale" ad near you soon :)
  10. Dowser

    Dowser Learning to bodge again..

    Hehe - I think I have just found a supply of TDA S1 DAC chips - I've just ordered 4 :)

    More to follow once I receive & test one of them.

  11. Dowser

    Dowser Learning to bodge again..

    Ah well, 4 S1's was all they had - their stock exhausted, mine refreshed...which is good, but no golden stash :)

  12. Dowser

    Dowser Learning to bodge again..



    I'm on sick leave after an operation earlier this week - guess what I am going to do today :)

  13. Dowser

    Dowser Learning to bodge again..

    Before changing the DAC chip on #2, I just scoped the ALWSRs feeding the OPA627s - they look fine. Only slight issue is that one of the positive items is outputting 16v, whereas other 3 are all dead on +/- 14.9v. This is within spec from the datasheet (16.5v max), but does the imbalance between positive and negative feeds to one of the opamps cause an issue? It seems silly that I do not know :)

    I'm going to go ahead and change the DAC anyhow, and add some decoupling around the 627s.

    Thanks, Richard
  14. martin clark

    martin clark pinko bodger

    interesting parts score!

    (and no, the DC voltage difference doesn't matter.)
  15. Dowser

    Dowser Learning to bodge again..

    Thanks Martin - after swapping the DAC chip for an S1 (hard work - removing an already replaced chip from a Naim PCB is not so easy!), I just looked again - Graham has changed all 3 pairs of opamps for OPA627s, for some reason I only thought it was the last 2 pairs he had done. I assume he left the original Naim supplies to power the first pair...I didn't scope those... I'm going to decouple all now, and scope the first pair on re-assembly.

    On the S1s, I wont say exactly how much I paid, but they did not differentiate between S1s and standard 1541s (I bought 2 of them too)...and they were around 20% less than the standard one I got from Cricklewood last week :)

  16. Dowser

    Dowser Learning to bodge again..

    Checking the Naim CDi layout around the opamps - they already have 100nF from +/- pins to ground pane very close to pins, plus 10uF tants a little further away. I do not see the value of adding more decoupling?

    What they do not have is 10nF directly between + & - pins, so I will add that (at least on the 2 opamps fed by ALWSRs, not sure if I should also add to the first 2 opamps being driven by the stock 317/337 regs?).

  17. martin clark

    martin clark pinko bodger

    Agree with leaving the standard decoupling if that's how its done.
    be careful with tying 10nF between two ALWSRs, could well end up with stability issues; try a couple and scope before and after.

    In any case I'd try the trick first on the I/V opamp, the pair nearest the dac, because these 'see' the toughest loading (exposed to full RF output of the dac etc) and if it helps, it will do most good here.
  18. Dowser

    Dowser Learning to bodge again..

    Haha - thanks, OK, I remove from the second 2 pairs of opamps, and I test. I was just about finished!

    I was also wrong above - the same pair of +/- ALWSRs power all three opamps, dedicated ALWSRs per channel.
  19. martin clark

    martin clark pinko bodger

    Oh don't let me stop you - just try your handiwork and report back :)
  20. laverda

    laverda pfm Member

    I've got to say 'nice one' Richard and glad you now have both up and running. I just didn't/don't have the enthusiasm.... Doing other projects....

Share This Page