No, not more diy lunacy

I 'think' many of the built in output stages in the dac chip 'only' have tuner/line level output, say a volt or less and the output stage appears to be the bit that takes the levels to 2V or so... The Naim would have their single rail output stage to give it the Naim 'sound' I think, so if you can bypass this, maybe the machine would sound better for it as I don't think CD players need to be so doctored these days?

I'm still suspicious of smps to be honest, but current legislation (I think - 'citation needed') may well cut down on rf Sh#t going in or out. Just use loads of clip-on ferrites around it - my bodge and it 'works' for me


I may have just 'upgraded' my geriatric Micro Seiki player. This one sounds best from its transformer coupled balanced outputs BUT, I think these invert the phase, so XLR pin 2 is actually negative rather then conventional positive. For me, a five minute job to check & change and the downstairs cack sounds more 'immediate' than it has done over the years on CD. I'm too bloody sentimental and don't want to chuck the speakers or sell the Harbeths right now... I also use an old rs-sourced mic cable from this player, it was the flexible mic cable with good size conductors, 'string' spacers and a double lap copper stranded screen, brown coloured (the nearest one is now a grey colour). Not for everyoine, but it likes 'this' player in musical terms.

You may be right Dave. I might have to have a poke around with the digital multimeter set on ac and see if i can read some voltages at various places. The volume may be more similar to  phono if the voltage out is reduced ? The output relay may need to be kept though. Might be some nasty noises when skipping tracks etc. 

I am all for getting rid of unneeded stuff. Musicality is more important to me now than "hifi" .

Its fun messing about with stuff innit

Nick G (Lurcher) would be the one to ask (or even Stan B), but I'd possibly suggest the higher the better (Nothing wrong to me in Nichicon FG electrolytics, although the audiophiles here would prefer the largest and most expensive Poly type possible - 'cos they look posher too and inspire audiophool confidence to go with the possibly slightly better sound!). Stan indicated with his original Caiman that the cap value depends on the load as seen by the amp, but this is where my lack of knowledge comes crashing in! My crown preamp has 25uF bi-polar caps (25V from memory) on the output for safety and the ton-up QED DAC (with 5534? stereo op amp) has 100uF caps at 16V or so - they're tiny whatever...

I do honestly believe *a bit* of maths/electronics knowledge can be a help rather than a hindrance, as it can save a hell of a lot of messing around and tail-chasing. I often look at something and wonder why it's done that way, but if I knew summat about it, I'd have a fair idea. Douglas Self let 'me' down severely on something he wrote which has been proven to be absolute incorrect in actual real-world practise and contributed to my having a fall out with Harbeth's Alan Shaw, who took the said article hook line and sinker... I trust Nick to accurately point you in the right direction once he knows the basic characteristics of the dac chip used... Colin W may also respond to a PM although I gather he despises digital and all it stands for (I don't know why, as it's usually totally transparent to stuff going on further back, rather than disguising it or painting over the cracks as vinyl usually does, but I digress...)

Can you double up to 20uF or higher? Instinct suggests that even 100uF may be better as any filtering and resultant phase shifts below 20hz may well be audible - If only I knew what I was talking about in this as regards a suitable value..

Output stage now bypassed 

10uf coupling cap and 100k resistor to ground. Output relay bypassed too. No pops, clicks or unwanted noise when using any of the remote control functions which is a bonus.

Volume out is definitely lower but not by much.  Sound has changed somewhat, more detail especially midrange, less fatigue and an easier listen which is not what i expected. Some reservations about the highest frequencies being rolled off a little but i need much more listening before coming to any final conclusions. Maybe a buffer is needed or a valve or an output transformer 

It has bypassed 6 opamps and 10's of resistors and caps, not to mention an output relay and a big ass dual secondary psu with super regs.

I could do that (and ought to) as I have a stock 3.5 in the cupboard which makes an appearance from time to time  This modded one has phono sockets though to accept TIS which makes comparison with the same cable impossible. I have a din to phono cable somewhere but not TIS.

Currently listening to Led Zep1 and I am hearing subtle detail that I thought never existed. I am trying to go through lots of different music before I commit to this new strategy. 

The 3.5 as stock is bold and lively. Some find it too much but the naim cdp that was said by some to strip enamel off teeth was the CDX. I have never heard one but always drooled over the top of the range models like Cdi, CD2, CDS2&3.

More laters

Did Copland CDA266 have any valves in it ? I have been reading the lampizator site about plonking a tube on the end of the dac. Never played with valves in fact I don't understand the circuits much at all except the psu. Will have to read up before I attempt that mod as larger voltages are involved.

Well I must have liked the lively presentation to buy this 3.5 way back in 1997 when it was new (one careful owner from new ). Modifications have improved it beyond recognition to my ears, in a good way which may be bad for some one else. Different mods have changed different things. I would like to believe that this modified 3.5 really is much better than any of naim's offerings bar the 555 maybe, although I will never hear the other models to compare.


Anyway it all depends on your ears and preferences. When you put lots of effort in to building and modifying your own stuff it really is hard to accept that something you just did is not as good. That's why I do prototypes first that don't take as much time to make as the finished article. Then not as much time wasted. Also use parts to hand until you know you are on to a winner. People really need to come clean on forums to say "no its not as good, back one step". Us diy nutters owe that to people who may be following in our footsteps. When you start out with little knowledge it can all be a bit daunting and a bit monkey see monkey do. I still need guidance from others and I would hate to pass on wrong info just because my ego wont let me.


So everyone, bypass that cdp anologue stage now. Its the best mod ever    Seriously I am still going through disc after disc hearing new things. Because of my fixed potential divider I cant change volume and coupled with the fact that I have lost nearly  half a volt of output from the cdp then I am listening at a lower volume whilst I try and compare to what it was like before. Hardly ideal and sonic memory is short and fades quick. There is a lot of merit in removing as much un needed circuitry as you can. I only realised recently that I could get by without the output stage. Phase shift anybody ?


Stu

Hey Dave,

I remember seeing the Copeland (Vickers hifi York) but it was out of my price range if I remember right.

Tubes - I have no experience of them outside guitar amps. I read every where that they add distortion in audio and some that defend them to the death. What is one to do apart from trying them yourself ?

Opamps in my CD3.5 are a mixture of the stock opa604 and some opa627 that I put in. They had their own twin secondary 500va transformer, Avondale TPR4 tracking pre regulator followed by some ALWSR super regs. That supply was as clean as I could get it and low impedance to boot. But there are still details covered over by the anologue stage.

Single rail might have something to do with it. If I had the time I might try and convert the output stage to split rail supply but I am not sure I will ever find the time for that. Plus it sounds so good now what could be more simple and effective. All the chips in the digital section run on single rail only including the dac chip and mostly at 5V (a couple use 9V but not sure they have to).

Its all fun

I was stunned by that Copland. I had owned a CSA 14 just a few months earlier and it was about as pedestrian an amp as you will ever hear. Don’t get me wrong, it was a sophisticated thing, and great with light jazz, but it did not know the meaning of boogie. The CD player on the other hand was more dynamoc than the CDX and still had all the sophistication. I’ve read a few people preferring it to the more expensive 288. I really should have kept it but I’m very happy with what I have now and it all matches. Is it as good or better than the Copland? I really have no idea, but I also have no intention of changing because it never fails to satisfy me. I don’t even look at other CD players.

Do you not hear anything... "radical" in the sound without the output stage?

When I was messing about with the output stage of a Technics SL-P1200 for Martin, I did try alternative op-amps, metal film resistors and Panasonic F Series capacitors. At one point, after replacing some of the capacitors, and having replaced the carbon resistors with standard metal films, there was an obvious improvement in terms of "torque". There was a much greater sense of sounds starting and stopping; much greater control. I stayed up until about 3AM listening.

Sadly, very sadly, the effect disappeared after a burn-in period of about 3 days.

Would a valve offer loads more distortion? I don't know. Some seem to have found merit in using a cheap valve buffer in their system. Perhaps a simple distorted circuit is preferable to a more complex less distorted circuit.

This is good. The designer of PS Audio's preamp/power amp. Even he doesn't know what's going on with valves. (Second video. First PS Audio's Paul McGown. Love his vids.):

I cant argue with what you have said Dave (and wouldn't want to).

Just glad some one is interested. Most of my diy threads turn into a one poster blog.

Never stopped me - cough.......

My thing - and I think Stu's too? is that when you change parts, what is actually happening? is stability improving, is distortion genuinely reduced and also, and important for 'digital' I think, what the hell's going on at much higher frequencies where the dacs are working?

Dave, I do think about what is happening although without test equipment I only have a DMM, a hunch and my ears to go on. RD would say that is all you need 

I am so enjoying listening to the 3.5 without output stage. Well, it still has a 1st order filter (don't mention that word) on the dac output which is a 680pf cap between two of the pins. The data sheet suggests 1n cap but Naim must have done that for a reason. Maybe without the rest of the output stage I should try 1n as per the data sheet. It also says that this 1st order filter is all that is required.

There is no doubt the presentation has changed and all for the good as far as I can tell so far. Or maybe I am biased ? The sound has tightened up considerably. The stock player was a bit to fatiguing for long term listening. I managed to get rid of some of that but it still bugged me - vocal sibilance is something I am particularly sensitive to. This mod seems to have cured it once and for all.

Opamps and regulators don't always play nicely together and you can get all sorts of ringing even oscillation. Getting rid of the opamps, regulators and psu's really has made a big difference. I would recommend it to anyone with a 3.5. There is a view that getting rid of the output stage can increase the output impedance  - I suppose it depends on the output impedance of the dac. Without a buffer o lower the output impedance the view of many is that you will run in to problems with passive preamps. Not so here. RD does bang on about impedance matching being a load of bollocks and that having enough voltage or current to bash the door open is what counts. I am leaning towards that view myself although I don't have the technical understanding of it.

Chemical Brothers are banging at the moment. Loving it.