Nakamichi CDP2 cdplayer

I have a ~1990 Nakamichi CDP2 cdplayer, it is one with the tda1541 dac. I was looking around to find some information about getting it into shape. I decided I needed to replace all electrolytics earlier, the player sounded overly smooth, not very detailed too, especially bass was funny, just air is best to describe it, uncrontrolled loose air without real tone where bass used to be (on my other cdplayer).
To get leadspacing for the electrolytics, to be sure, I looked at bottomside pcb. While doing that I noticed all voltage regulators had very bad soldercontacts, cracked all around the leads, so I resoldered them...now there is tone with bass and I would not describe sound as smooth Better detail both treble and bass, but treble a little flat, some might call it harsh, bass not strong enough and really could be better I think. It is so different I need to listen to other cdplayer again to get firm grip on what I hear.
But looking at electrolytics I also noticed the glue around them getting dark edges, and some oxidation on parts touched by the glue, so might be better to get that glue off anyway.

 

Ok needed to look if posting pics would work
My question is, does any of you have suggestions how to improve bass from a tda1541 player? And maybe same time loose a little flatness/harshness in the treble?

Anyone seen the glue discolouring and oxidation of parts in glue befor?

 

It looks like you've got a fun project there. Have you managed to find a circuit diagram yet? Try the HiFi engine, they've got a huge database there.

Starting with replacing the large electrolytics wouldn't be a bad idea (the five big green ones and the single big black one). Panasonic FC, FM and Rubycon ZLH are all excellent. Go with the same voltage rating or higher. You are doing the right thing by checking the lead pitch and physical dimensions first to check that the replacements will fit neatly.

It'll be a massive help if we can see a circuit diagram.

 

I have the hifiengine servicemanual, but it has a bad sort of picture of pcb, invisable traces, some diagrams for parts of the cdplayer, but labelled as on pcb, this diagram may be best overall

 

We need a schematic showing the electrical connections and the components used. I see if I can download the schematic myself from the HiFi engine.

 

Well my Hifi engine download has the parts tabel, but schematic is too poor quality I feel. Maybe I got a bad download The above diagram does show same chipsets as some Sony's I think. The two 5532 opamps have 100nF decoupling closeby I saw, not even the 1uF the Sony has, further away near the 4 2200uF's and the +15V regulator there is a 470uF capacitor for the opamps. Maybe I put the 100nF as SMD between the current capacitor pinholes and on topside pcb I put a larger value? So 4 filtercaps for 15V are 2200uF/25v, four 10D1 rectifiers where leads are having black flakes coming off them, for "main" low voltage secundaries there is 2200uF/16V and 3300uF/16V(dark blue one), positive or negative side has bigger filtercap there, and a small bridge rectifier. Then there also is a "'single" output secondary, 2 leads instead of three, that has 1000uF/16V and 1 diode 1SR35-100 covered in glue (and in other places oxidation shows were glue touches metal). I read Schottsky's could be an improvement, I think they should be replaced anyway. The small bridge looks ok but maybe there are better ones, it is really tiny, 15mm lenght.

 

On a better quality CD player like this one they'll usually be a separate transformer winding for the DAC and analogue stage which looks to be the case here. The block diagram shows +15/-15v/+5v/-5v for the analogue circuit which makes sense. The TDA1541 needs +5v, -15v and -5v. The op-amps will probably be getting +/-15v and the filter chip will likely be +5v.

Most of the sound quality improvements are going to be made by improving the analogue circuit and associated power supply. I suspect the four discrete rectifier diodes and the four green 2200uF caps are the power supply for the analogue circuit. I'd replace the four diodes with Schottky 11DQ10. The four filter caps can be replaced with good quality long life low ESR types, Panasonic FC, FM and Rubycon ZLH are some of my favourites.

I wouldn't bother changing the rectifier for the digital circuit/servo circuit but fitting new electrolytics might be a wise precaution to make sure the player is in tip top order.

Near the output there will probably be a pair of electrolytic caps, sometimes polar and sometimes non-polar which will be directly in the signal path and DC blocking coupling caps. These can be upgraded to a better 'audio' quality electrolytics or better still film caps. The rough order of preference is:

polypropylene/foil (best)
metallised polypropylene film
Polyester film

BUT large axial polyproylene film caps are going to be very large and difficult to install neatly. The Wima XL MKS2 range if you can find them are very compact and will usually just drop straight into the 5mm pitch PCB holes. Try these:

http://www.ebay.co.uk/itm/3PCS-10uF...366?pt=LH_DefaultDomain_3&hash=item51c6cd31c6

or these

http://www.ebay.co.uk/itm/10PCS-4-7...747?pt=LH_DefaultDomain_3&hash=item3cf16f6b2b

 

If I would select 105 degrees lowest ESR/highest ripple/long life/highest capacitance that would fit in place of the 4 green Toshin Kogyo 2200uF/25v, Nippon Chemicon KZN 6800uF/25v would be best (Mouser), but wouldn't there be a chance inrush current would be too high ? 13600uF per side instead of 4400uF seems big difference, but saw you went from 3300uF to 5600uF too, so maybe nothing to worry about, I have no experience with this.

The standard rectifiers/diodes are 10D1, 100V, 1.5A average, 50A peak. At Mouser closest schottky would be Vishay SB2H100, 11DQ10 is maybe previous version, non rohs eg?

Output side opamps indeed has a electrolytic, 47uF. On main pcb those are the only non Toshin Kogyo branded electrolytics, Nakamichi used Chemicon AWD series there, the Wima's 10uF indeed are same size, amazing for film cap, I thought they would be huge. I will order some.

 

Extra capacitance and low ESR in the smoothing caps should bring benefits (up to a point) but yes going from 2200uF to 6800uF does seem a big jump. Having said that, I've seen plenty of modified CD players using huge smoothing caps (like 22,000uF) without problems. Why not compromise and go with something like 3300uF which will give you a modest increase and will also give you more series of capacitors to choose from. If in doubt stick with the standard values. I've no experience with Nippon Chemicon as I always stick to Panasonic and Rubycon myself so I can't comment on the KZN. The specific smoothing capacitors used do seem to affect the sound and I've found that some manufacturers/series sound better than others.

You don't want to use capacitors with super low ESR immediately after voltage regulators as it can cause problems (ringing). My personal favourites to use after the regs on the supply to the op-amps would be Elna Silmic.

11DQ10 are 1A and 100V. I'm not sure what the equivalents are but you can buy 11DQ10 easily from most suppliers (e.g. Farnell) and also on eBay etc.

 

11DQ10 are no longer produced/obsolete at digikey/mouser/farnell, some kind of toxic substance no longer allowed Probably harmfull if you chew on them, silly rules most of times. But the one I mentioned is probably succesor.

I have read advice to go for biggest value low esr that could fit for this position, probably shouldn't worry if some use 22.000uF. Do you put a small SMD film as bypass under the PSU caps ?
Post regulators SilmicII I will order, same question, do you use SMD bypass there? Currently the opamps have a 100nF film real close to chippin, I think a larger value SilmicII there with bypass same as existing value could improve supply. 5532 probably not best opamp, I might try another but need to read a little about application first, wider bandwidth could oscillate I guess, so could bring problems too. You like the opa2604 I think?
Output from dac goes to a de-emphasis transistor here and then to opamp, most cdplayers I have seen skip the de-emphasis, don't really know what it was, but hasn't been used much I believe, only maybe first produced cd's ?? After opamps there is a transistor again in each output, this time for muting. I also wonder about removing that one, could I try that without transistor and put them back in if I don't like the "noise?" without them?

 

I increased the value of the smoothing caps in both my Arcam and Sony by about 50% without any issues. I would advise against adding a film bypass under the smoothing caps as it could create extra noise on the supplies.

Silmics are not low ESR which makes them safe to use after a reg without fear that they might cause problems. I have found that the capacitors used to supply op-amps can have a really noticeable effect on the sound and I liked Silmics the best when I tried different caps in my Arcam. In my Sony the op-amps power supply pins were only decoupled to ground with electrolytics and since I wanted to upgrade the op-amps I chose to add the PPS film bypasses. Usually I'd expect to see op-amp power supply pins decoupled to ground with something like an electrolytic of 10-100uF plus a small film or ceramic bypass of 22-100nF. You need an oscilloscope to test the stability of an op-amp in circuit properly but I don't have one yet (I'm thinking about getting one). I've bought some OPA2604 but haven't tried them yet. At the moment I'm using LM4562 in the Sony which I'm really liking, they're very clean sounding. OPA604 has a reputation for being very musical but slightly coloured. OPA2132 also seems popular with modders.

 

You can usually remove the muting transistors without any problems but you might hear some pops and clicks when changing tracks on the CD etc.

When I asked forum member Felix (who is very knowledgeable) about removing the de-emph transistors on my Arcam he advised me not to bother.

I suggest you de-solder just the muting transistors and try it. You can put them back in if you get excessive unwanted noises.

 

Thanks for your answers, I completed an order for parts

 

Glad I could help. Let us know how it goes.

 

In first picture it seems back two of the grouped four electrolytics are bigger and not mounted neat, both at back side were leaking under glue and in fact a little bigger than equal front parts, so Panasonics FC adviced I decided to order will probably be a nice improvement Parts arrive probably near end of week...lets see if other caps leaked too...

 

That sounds nasty! If some of the caps have started leaking then there's a strong argument for replacing all of the electrolytics on the board, just to be on the safe side.

It looks to me like there are two caps 220uF? near the TDA1541, possibly after the two regs IC304 and the one next to it. My best guess is that these will be for the +5v and -5v supplies to the DAC. I suspect the -15v is shared with the +/-15v supply for the op-amps.

The +5v for the DAC is on pins 27+28 and the -5v is on pin 26. Something like 220uF Silmic or Cerafine will do nicely here.

Is C211 connected to pins 16 and 17 of the DAC? If it is then it's the DEM oscillator cap.

 

I did order all electrolytics in cdplayer. Both 2200uF on +15V (going to LM7815) side of supply had dumped smelly stuff on pcb under glue. one tried to force the rubber out of can and pushed itself a little to one side. For DEM it has a 680p I recall, will try this:

 

If you have an external DEM cap it means that you can upgrade from TDA1541 to TDA1541A

I did some measurements a while ago and decided that 560pF was the optimum value (see my Arcam Alpha 5 thread) but to be honest the audible difference is quite insignificant so I wouldn't worry too much, anything from 470-680pF ought to be fine.

I use a very simple DEM circuit which just ties each leg of the 560pF to -15v (at pin 15 of the TDA1541A) via a 6K8 resistor. It was a circuit which was recommended on the gigantic TDA1541A DAC thread over on diyaudio.

 

I read somewhere that Phillips datasheet for some time indicated NC for pin16/17, but that internally in chip this was not true (blue c211 in my pic). This Phillips solution in Grundig FineArts is also from diyaudio, but not very long thread Phillips wrote advice for companies buying the chip from them, but was (near) EOL as it turned out, which is typically Phillips btw. Long thread probably is about N(o)OS too, this Phillips solution doesn't work with NOS, or better it will work but DEM will be set too low for audio

 

This from lampizator is a little confusing to me:


He calls these treble muting (and noisekilling) caps. Somehow that makes me think they are in the outputsignal, but I looked and they are not. They are coupling outside gnd (rca shield) to an inside cdplayer gnd, the opamp gnd. Question now is did lampizator think they were filtering output signal, to me it looks like that...