Philips CD951 • A DAC7 (TDA1547) vs TDA1541A R1 vs SM5872 story

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello there,

I've been given a Philips CD951, CDM9/65 mech and TDA1547. (Thanks again Jan).
The mech is hold as a good one (it plays CDR-W wich is nice), the TDA1547, with its decoder and filter on separated ICs, is great on paper and specs, sounded nice in some good stuff.

My aim, wich will tell how much and how I'll go, is to compare it to my soon to be rebuilt Marantz CD-53 (check the huge thread about Marantz CD-63 mods) and a Xmas gift wich will be a familly unused Marantz CD-40 with a CDM4 and TDA1541A R1 DAC.
I need only one good player, will try to get the best from these 3 guys and get away the 2 loosers to my ears. And the CD951 can do a good transport seems, even if right now an integrated CDP is all I want.

As for now, virgin CD951 but some years old (caps dead?), it's heavier than my CD-53, bigger (nice headroom for adding stuff), at first I tought the front panel was plastic but no, Philips made the aluminium front looks like plastic...
Inside, it's like they applied the Special Edition treatment from Marantz to the CD950: Cerafine aoround opamps and DAC. The caps on other ICs are so small... they deserve to leak in peace. Opamps are JRC bugs. The top lid is a bit dempened and the TX humms... bad news. And last thing they've discarded the 2 lines dot display from 940 for a single line one under the large screen.

The sound, compared to my CD-53 (state of the CD-53 right now, smoothing caps @ opamp PSU, PFM Flea clock on clean test-board, Servo reclocked from /2 Flea, Regulated drivers and Rubycon ZLH caps around digital ICs, analog stage still factory but 6.8µF MKP and removed transistors).
Stock CD-53 is warm but dull (boomy in fact) and in the right now state more open and clearer than the Philips but the main point is the lack of the half down (again, compared to the CD-53 as modded now) the lack of bass and mids. Or too bright if you want, with maybe clearer highs than the CD-53.
But it's a bit unfair as it still features cheap lytics as DC blocking caps, tiny decoupling caps and such nasty things.


First step, I can free the standard sound with LM4562 or OPA2132, fit 470/35 or 220/50 Rubycon ZLH at ICs but the DAC (or the DAC?), try to use bigger standard caps (from the CD-53 removal) in the PS and bypass here and there with 1µ MKS, maybe 4x MBR1100TR schottky left and MKP caps for the output. It should already sing more.


And then we'll go for real work (unless I really don't like the cheap mods I'll try tomorrow),
I want to:
Spy what have been done on the best DAC7 products, guys with Deltec DACs please open the lid (mister MC? ;) )

• avoid opamps -if simple solution is possible- even the internal TDA1547 ones,
use of the Ray's DOS dedicated to the CD-63, have already been done by a member,
or http://digilander.libero.it/paeng/a__tda1547_tweak.htm
but having in mind the technical point of view from another great dutch boy "The internal opamps buffer the switched cap network (...) but with the TDA1547, the opamp can be bypassed, because the connections are made outside the IC. It's a good thing, because the internal opamps are crap. But now that the SC network is 'exposed' to the outside world, you have to be careful not to load it too much, or distortion will be unavoidable."

• build some dedicated "not so" noisy regulated PS,
like that "sixpack": Fikustransport also great about the wood plinth.

• fit a clock, a Flea or an AudioGD one (the AudioGD may go with a discarded player),

• the right and definitive choice of caps on PCB (SEPC & BlackGates?) with no overkill values but efficient ones,

• the ability to use it as very good transport...

And last but not least... I'm a bit lazy for ready and understanding, but I will so please... help me!

Matthieu
 

Attachments

  • DSC00206.jpg
    DSC00206.jpg
    191.9 KB · Views: 1,482
Great!

Hi Matthieu,

I'll be watching this thread closely as I am currently in the middle of a personal modding battle between a Philips CD951 and a Marantz CD50. :headshot:

So far I've added os-cons (from Fikus' modding list) to the digital part of the CD951, that makes the player sound good, shocking in fact.

For the CD50 I've only weighted the chassis and put it to NOS (with correct mute) playback.

I am falling more and more in love with the DAC7 sound compared to the good old TDA1541A. The mids are really amazing after the os-con change (it needs some time to burn in though) and the music is really alive already.

The CD50 is just okay at the moment, a lot better than stock but not superduper yet.

Next up is paralleling the decoupling caps on the DAC chips and do the reverse weightening and os-cons so the comparison becomes better. After that, individual PSU's to the digital consumers. By then, they should be unbeatable, and obviously I am curious in the sound difference.

I've put the mods as Fikus has made to both players in a nice Word document. If you're interested in those, PM me.

Cheers,

Mark.
 
Too late to edit...
I'm fitting all what I have left in, no matter what it is ;)
Wonder if low ESR caps everywhere is really needed, e.g. smooting caps? And do bypass capacitor at rectifiers still needed with the schottky? I'll leave them right now anyway. They fit bypass caps everywhere!

Right now until a better solution I plan to use the internal op amp config, I mean the parts wich are out of the DAC now for them, use a single outside opamp there and then the same Discrete Output Stage as in the CD-63.

But that's later anyway...

Matthieu
 
Any help welcome...
Yesterday in test purpose I've tuned it on for hours but had the opportunity to listen to it later. It was warm, but when pressed play aid a couple of time error. Some more press and it was OK. I thought it was an RC problem as it was another one... but in fact that was stupid as it worked fine sometimes and with the Marantz.

But today, after opamp changed and many caps changed for bigger (that's all) it refuse to turn the disc, it spins backward en a bit normal but "error" or "loading" messages. I've checked the flex ribbbon a couple of times without success.

I don't know where to check and I'm not familiar at all with the pcb, a bit lost.
Maybe bigger caps can harm there??? Sounds strange anyway...

Matthieu
 
Any help welcome...
Yesterday in test purpose I've tuned it on for hours but had the opportunity to listen to it later. It was warm, but when pressed play aid a couple of time error. Some more press and it was OK. I thought it was an RC problem as it was another one... but in fact that was stupid as it worked fine sometimes and with the Marantz.

But today, after opamp changed and many caps changed for bigger (that's all) it refuse to turn the disc, it spins backward en a bit normal but "error" or "loading" messages. I've checked the flex ribbbon a couple of times without success.

I don't know where to check and I'm not familiar at all with the pcb, a bit lost.
Maybe bigger caps can harm there??? Sounds strange anyway...

Matthieu

Bigger caps shouldn't hurt, as long as you stay in the 2x or 3x range...... Maybe you shorted something with some solder? :confused:
 
Fixed!
Stupid at iron... I've seen a small cap at 2107... 2.2µF not for supply, changed it for a 470/35 :D back to 2.2 it works just fine! Will change this to film.
Now I'll have an ear but with cheap non polar lytics at dc blocking (and MKS to bypass them, won't be worst...).

Thank for support Mark ;)

No worries mate :spin:

I'll be doing some mods in the coming days..... keep you posted.
 
How is your analog stage right now?

The analog stage is currently standard, thus the 5532 opamps in the CD951. 5534 in the Marantz if I am correct. I am a little in doubt wether to go for the lampizator stage or upgraded opamps.

Obviously the lampizator can be used in both players enabeling a really good comparison of the sound and quality of the player. However I have no experience with tube buffer amps and need a very detailed description in order to make a good one.

Upgrading to (for instance) Burr Brown 2904's is another option, companied with nice parralleling caps etc. Low cost, simple and often with interesting results.

Any recommendations from your side? :bulb:

Cheers, Mark.
 
Regarding op-amps, I really do not like any of them, but when experimenting with them, I´ve always ended up with 5534 and 5532 which often are the originals.
They have very low noise and very good current delivery, all the rest is pretty ordinary, but told to be sufficient, and most important to me, they are bipolar.
They are through their lifetime modified and developed further, so you might find newer ones sounding different. Btw. they are used in almost all pro audio gear and also Accuphase use them in their high end players @ only 8V.

Carefull decoupling is a sport that can provide substantial improvements. Try polyphenylene sulfid capacitors from i.e. Evox or Wima. They are exellent for that purpose, and also deliver very good sound quality, when used for feedback caps also.
The Evox is called SMR and comes in 5mm packages and up.
 
My first player, the CD-53 is waiting for its Discrete Output Stage back, as seen on Ray's website http://www.raylectronics.nl/ it's just more neutral, more fluid than any opamp I've tried in this one. I also want to avoid opamps and also want to have the best decoupling work and these PPS are recommended very often, thanks for the Evox tip.

The CD951 will also get one, the problem being the internal opamp doing the I/V thing and the buffer for the switched capacitors. I may use the DOS after the DAC opamps, as it's balanced voltage output or not by reuising the former PCB traces of the output ones to use better opamps. Maybe LME49720HA, maybe other solutions if possible as I've found 2 cases wich seams to load the switched capacitors network without buffer.
Right now the LM4562 does a good neutral job but I can't tell much as the output caps are still OEM. If I were to keep the opamp circuit I would change tiny regulators and even smaller caps after.

And the CD-40 (wich is a CD-50 without a daughter board that's it?) is planned to be fitted with AD8620 or another fast IC like AD8066. I don't know what cheap option is avalaible for an non-opamp output.

But I agree with you, such "I have to choose" game is complicated, how to know if the work done and still to be done is the same? Some enjoy modding just for modding, I like it but just to choose, when done it's over for me! I'm scared my heavily modded CD-53 is "better" than other right now because it's recloked 2 times and has no opamp but only DOS and top output caps combo... and what about price? I've not planned to loose money in trying, at least get my money back on modded players and why not make money... that's all but uncertain...
 
Regarding op-amps, I really do not like any of them, but when experimenting with them, I´ve always ended up with 5534 and 5532 which often are the originals.
They have very low noise and very good current delivery, all the rest is pretty ordinary, but told to be sufficient, and most important to me, they are bipolar.
They are through their lifetime modified and developed further, so you might find newer ones sounding different. Btw. they are used in almost all pro audio gear and also Accuphase use them in their high end players @ only 8V.

Carefull decoupling is a sport that can provide substantial improvements. Try polyphenylene sulfid capacitors from i.e. Evox or Wima. They are exellent for that purpose, and also deliver very good sound quality, when used for feedback caps also.
The Evox is called SMR and comes in 5mm packages and up.

Hi Kurt,

What do you recommend for an analogue stage?

Cheers,

Mark.
 
My first player, the CD-53 is waiting for its Discrete Output Stage back, as seen on Ray's website http://www.raylectronics.nl/ it's just more neutral, more fluid than any opamp I've tried in this one. I also want to avoid opamps and also want to have the best decoupling work and these PPS are recommended very often, thanks for the Evox tip.

The CD951 will also get one, the problem being the internal opamp doing the I/V thing and the buffer for the switched capacitors. I may use the DOS after the DAC opamps, as it's balanced voltage output or not by reuising the former PCB traces of the output ones to use better opamps. Maybe LME49720HA, maybe other solutions if possible as I've found 2 cases wich seams to load the switched capacitors network without buffer.
Right now the LM4562 does a good neutral job but I can't tell much as the output caps are still OEM. If I were to keep the opamp circuit I would change tiny regulators and even smaller caps after.

And the CD-40 (wich is a CD-50 without a daughter board that's it?) is planned to be fitted with AD8620 or another fast IC like AD8066. I don't know what cheap option is avalaible for an non-opamp output.

But I agree with you, such "I have to choose" game is complicated, how to know if the work done and still to be done is the same? Some enjoy modding just for modding, I like it but just to choose, when done it's over for me! I'm scared my heavily modded CD-53 is "better" than other right now because it's recloked 2 times and has no opamp but only DOS and top output caps combo... and what about price? I've not planned to loose money in trying, at least get my money back on modded players and why not make money... that's all but uncertain...

What about the Discrete Output Stage back, as seen on Ray's site for the CD50? I mean, why do you plan it for the 53 and not the 50?
 
That won't work with the CD50, it's an unbalanced DAC output and the DOS is meant for balanced.
It will anyway work fine with our CD951, evein if right now I don't know if filter has to be recalculated and parts values changed. In worst case I'll keep the DAC internal opamp and get rid of external analog stage in the CD951.
You also may get a DOS PCB from Ray if you plan to do it, check this link to its post about it:
http://www.diyaudio.com/forums/digi...ntz-cd63-cd67-mods-list-1348.html#post1974993
I'll report soon, maybe in few days if filter is ok as in the CD53 =) (my CD53 ones are out waiting for BlackGates :) )
 
Hi Kurt,

What do you recommend for an analogue stage?

Cheers,

Mark.
If one have the option of choosing one self, I´d recommend a discrete design.
And to me that means NFB design, DC coupling, carefull filtration, low output impedance, high current capabillity, very low noise supplies and so on.
But this is hardly obtainable these days, almost only very expensive high end products feature this kind of design.
But next to that, I´d go for bipolar op-amps, and be very carefull about choise of surrounding components and supplies.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.