There's lots of advice on modding Philips/Marantz TDA1541 machines, but very little on Sony unfortunately.
Sony swapped between A and non-A chips in some machines. CDP337ESD, for another example. I assume the circuit is happy with either.
On more up-market machines, they used film caps for DEM decoupling (the two rows of seven caps either side of the DAC). If your ceramic caps are NP0 (same as C0G) I would leave them as they are. If not, consider replacing with film. There's masses of historical debate about the best values to use.
If output coupling caps are polar, you might find non-polar electrolytic an improvement. If the machine is working reliably, I would leave most other caps alone. Those in the analogue power supplies (DAC and audio opamps) are likely to make the most difference if you change them. Extra 100nF ceramics from opamp power pins to ground may be an improvement.
The large-value cap is for memory, and possibly supports some "favourite track" function. Check the user manual.
You are lucky in my view to have a proper plastic loader and laser mech. The lasers are easily and cheaply replaceable, and the loaders are reliable and endearingly clunky. The only weakness is the belts don't last long. Cleaning and re-greasing the cam drum and ball can make a worthwhile difference if you have some suitable 100% synthetic grease handy.
P.S. Just seen Captouch's mention of OP275. Try them please, Captouch, I'd like to know what you think. They made the biggest difference I've ever heard!
Sony swapped between A and non-A chips in some machines. CDP337ESD, for another example. I assume the circuit is happy with either.
On more up-market machines, they used film caps for DEM decoupling (the two rows of seven caps either side of the DAC). If your ceramic caps are NP0 (same as C0G) I would leave them as they are. If not, consider replacing with film. There's masses of historical debate about the best values to use.
If output coupling caps are polar, you might find non-polar electrolytic an improvement. If the machine is working reliably, I would leave most other caps alone. Those in the analogue power supplies (DAC and audio opamps) are likely to make the most difference if you change them. Extra 100nF ceramics from opamp power pins to ground may be an improvement.
The large-value cap is for memory, and possibly supports some "favourite track" function. Check the user manual.
You are lucky in my view to have a proper plastic loader and laser mech. The lasers are easily and cheaply replaceable, and the loaders are reliable and endearingly clunky. The only weakness is the belts don't last long. Cleaning and re-greasing the cam drum and ball can make a worthwhile difference if you have some suitable 100% synthetic grease handy.
P.S. Just seen Captouch's mention of OP275. Try them please, Captouch, I'd like to know what you think. They made the biggest difference I've ever heard!
Last edited:
The caps around the TDA1541 DAC are the thin, circular, green ones like the OP has shown in the early postings on this thread - not sure what kind of caps they are (NP0 or not), but I believe they were used on lower end machines and polypropylenes often of higher value were used for 1541 decoupling on higher end machines. Please correct me if I'm wrong.
Regarding 1541 non-A vs A, most everyone seems to prefer the A, and I see they're available on eBay from a lot of Chinese sellers for $15-20. If I knew they were genuine A's, I would consider it, but am wondering if they could be remarked non-A's in some cases. The Lampozator guy seems a fan of the non-A though.
I was only planning on changing the cluster of six caps that contain 4 Sony Duorex and 2 Muses - I believe they're all past the DAC output and near the opamps - I have the schematic from the Sm, but am on my tablet so can't post them here. Because of reports that the Duorex caps to bad and corrosive, I thought I best change them! There are 40-50 other caps around the board, but yeah, I was planning on leaving those alone.
The player is quite plasticky and once you remove the metal cover and bottom metal plate, quite lightweight, but it mechanism is indeed fast and it reads discs faster than any other player I have. So, even for a ~25yr old player, replacement laser assemblies are available? Or only from donor units? The PCB is of high quality though.
Yes, I will report back on OP275, OPA2604, and LM4562 vs the original 5532 (assuming they survive my desoldering). I had the OP275 in my Hifiman EF5 headphone amp and have since used AD797 and LM49990 in it, so the OP275 (and a spare I bought) are available. Only have one of the AD797 and LM49990 (they're single channel opamps anyway), so even if those were compatible, I could only do a single channel on the CD player - though that could be interesting as well! But I won't risk it unless you guys think it would be stable. Right now, two single channel chips are on both sides of a Browndog adaptor since the EF5 only uses a single dual channel opamp socket.
Thanks for all the replies for a newbie modder - I appreciate it. The new parts won't ship until early this week, so I won't get to the mods until later in the week.
Regarding 1541 non-A vs A, most everyone seems to prefer the A, and I see they're available on eBay from a lot of Chinese sellers for $15-20. If I knew they were genuine A's, I would consider it, but am wondering if they could be remarked non-A's in some cases. The Lampozator guy seems a fan of the non-A though.
I was only planning on changing the cluster of six caps that contain 4 Sony Duorex and 2 Muses - I believe they're all past the DAC output and near the opamps - I have the schematic from the Sm, but am on my tablet so can't post them here. Because of reports that the Duorex caps to bad and corrosive, I thought I best change them! There are 40-50 other caps around the board, but yeah, I was planning on leaving those alone.
The player is quite plasticky and once you remove the metal cover and bottom metal plate, quite lightweight, but it mechanism is indeed fast and it reads discs faster than any other player I have. So, even for a ~25yr old player, replacement laser assemblies are available? Or only from donor units? The PCB is of high quality though.
Yes, I will report back on OP275, OPA2604, and LM4562 vs the original 5532 (assuming they survive my desoldering). I had the OP275 in my Hifiman EF5 headphone amp and have since used AD797 and LM49990 in it, so the OP275 (and a spare I bought) are available. Only have one of the AD797 and LM49990 (they're single channel opamps anyway), so even if those were compatible, I could only do a single channel on the CD player - though that could be interesting as well! But I won't risk it unless you guys think it would be stable. Right now, two single channel chips are on both sides of a Browndog adaptor since the EF5 only uses a single dual channel opamp socket.
Thanks for all the replies for a newbie modder - I appreciate it. The new parts won't ship until early this week, so I won't get to the mods until later in the week.
I believe the KSS150A is interchangeable with KSS210A. Both are available from several vendors new. Possibly some are better than others, depending on origin.
It's kinder on the pcb to use sockets for opamps until you decide what you like best.
The Arcam Alpha used single opamps, OP27 and 5534. Probably unlike the Sony, it does not use the I/V stage as part of the filter. I think I'm using OPA27 and LME49710 at the moment. I don't recall hearing much difference.
It's kinder on the pcb to use sockets for opamps until you decide what you like best.
The Arcam Alpha used single opamps, OP27 and 5534. Probably unlike the Sony, it does not use the I/V stage as part of the filter. I think I'm using OPA27 and LME49710 at the moment. I don't recall hearing much difference.
CDP M95 is very similar. Smaller case, and audio on separate board. Same loader and laser. Ceramic DEM caps replaced with 220nF and 470nF plastic. Just a few electrolytic swaps, including the three around the TDA1541 and all on the audio board. NJM2114 opamps replace original 5532.
Attachments
yup, CDP M95 board looks similar to CDP 950, just bit different part & layout on upper right corner there. Modded my CDP 950 to the max and it's now used as transport only 
if I'm going to do it again I'll have discrete common gate jfet I/V a'la ECdesign and a B1 buffer after that. Should be very transparent, no caps in the line
if I'm going to do it again I'll have discrete common gate jfet I/V a'la ECdesign and a B1 buffer after that. Should be very transparent, no caps in the line
My CDP-950 doesn't have any digital outs, so analog out is the only option. My parts shipped yesterday, so hopefully I'll be getting them in soon.
BTW, I will be using sockets for the opamps, not desoldering/soldering the new ones each time.
Any of you have the CDP-707ESD/557-ESD? Just curious. There's one locally near me, but since the laser is NLA, I'm hesitant to pull the trigger. Not a TDA1541 DAC, but a BB PCM64P - seemingly not very common.
BTW, I will be using sockets for the opamps, not desoldering/soldering the new ones each time.
Any of you have the CDP-707ESD/557-ESD? Just curious. There's one locally near me, but since the laser is NLA, I'm hesitant to pull the trigger. Not a TDA1541 DAC, but a BB PCM64P - seemingly not very common.
CDP557ESD has the epic KSS190A laser mech and all-metal sash-driven loader. Two transformers and all the top Sony hi-fi candy.
Not nearly enough plastic in my view. Moving so much weight, I think the mechanics in the head get sloppy eventually, and they can be mechanically noisy.
I've seen KSS190A fetching silly prices so I guess good ones are rare. If the laser itself is working, you can fix most other things and it could be an excellent machine if you don't mind low-level fidgeting noises from the mech.
Not nearly enough plastic in my view. Moving so much weight, I think the mechanics in the head get sloppy eventually, and they can be mechanically noisy.
I've seen KSS190A fetching silly prices so I guess good ones are rare. If the laser itself is working, you can fix most other things and it could be an excellent machine if you don't mind low-level fidgeting noises from the mech.
Would I spend that on a 557 ? For me it would have to be in pristine out of the box condition to contemplate it, and it would be just as a restoration project... and the big question would be what would I do with it afterwards, because it is outclassed by other players on the sound quality front.
So only you can answer that one
If its what you want then go for it
I did a little Sony restoration here but I wouldn't have considered it but for the fact it was 100% unmarked.
http://www.diyaudio.com/forums/digital-source/226288-sony-cdp790-kss240-restoration-project.html
I did a little Sony restoration here but I wouldn't have considered it but for the fact it was 100% unmarked.
http://www.diyaudio.com/forums/digital-source/226288-sony-cdp790-kss240-restoration-project.html
I have a couple of Sony ES pieces from roughly that period (TA-E77ESD preamp and ST-730ES tuner), so the rosewood would fit in. But yeah, it's non-trivial $ and I could spend that money on a desoldering iron and get to work recapping all my old vintage receivers/integrateds.
We'll see, but I'll probably content myself with the CDP-950 tweaks instead.
But yeah, it's non-trivial $ and I could spend that money on a desoldering iron and get to work recapping all my old vintage receivers/integrateds.We'll see, but I'll probably content myself with the CDP-950 tweaks instead.
Well, I finished putting in the opamp sockets, 0.47uF MKP caps on pins 13/18, 0.22uF MKP caps on other TDA1541 pins, and replaced 4 of the 6 caps I intended to replace.
I made a mistake and ordered 4.7uF instead of 47uF for two of the Duorex caps, which kind of sucks, but I replaced the 470uF and 1000uF ones.
To get a baseline for the cap only changes, I put the 5532DD opamps back into the sockets and gave it a listen. The biggest change I noticed was there was much more bass after the mods. Seemed smoother as well.
Will let the caps burn in a bit with the 5532 opamps, then try the OPA2604, LM4562, and OP275 opamps to see what kind of changes they bring.
Before:
After: The MKP caps are much bigger, so 4 went on top, and 3 went on the bottom.
The new 1000uF Muse caps are really big and tall, and kind of get in the way of the opamp sockets and the cover may touch them as well, though I think I can manage. I knew they'd be big, but thought I could bend them into a convenient position. But it turns out they're constrained on one side by the opamp socket, another by the MKP caps, and on the others by each other or the other caps.
I made a mistake and ordered 4.7uF instead of 47uF for two of the Duorex caps, which kind of sucks, but I replaced the 470uF and 1000uF ones.
To get a baseline for the cap only changes, I put the 5532DD opamps back into the sockets and gave it a listen. The biggest change I noticed was there was much more bass after the mods. Seemed smoother as well.
Will let the caps burn in a bit with the 5532 opamps, then try the OPA2604, LM4562, and OP275 opamps to see what kind of changes they bring.
Before:
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.
After: The MKP caps are much bigger, so 4 went on top, and 3 went on the bottom.
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.
The new 1000uF Muse caps are really big and tall, and kind of get in the way of the opamp sockets and the cover may touch them as well, though I think I can manage. I knew they'd be big, but thought I could bend them into a convenient position. But it turns out they're constrained on one side by the opamp socket, another by the MKP caps, and on the others by each other or the other caps.
No, but that's a reasonable retail price if it's local and in good condition and working order. Depends on what you want. I'm happy to pursue TDA1541 to its limit whilst preserving its character. After that good character disappeared, with few exceptions. Later still, all character disappeared and all CD players became equally acceptable. IMHO
The other reason is that the very wonderful KSS190A is a liability. I have a CDP337ESD. The typical Sony head is an overcomplicated design with a basic flaw. The heavier they are, the worse they rattle eventually. Also, there's nothing special about the laser diode apart from being irreplaceable.
Attachments
Did some more comparisons between the different op amps.
As far as the rest of the system, the CD player went into a Conrad Johnson PV-5, which fed a B&K EX442 amp driving IMF RSPM IV's.
In order of preference, I liked:
LM4562: The most dynamic, detailed, and among the largest presented soundstage. On a relative basis, it could be said to be on the brighter side, but overall I didn't find it fatiguing. This is the one I'll keep in.
JRC5532DD: Actually not too far from the LM4562, but it seemed a bit harsher at times and less natural sounding.
OP275: Seemed to accentuate the mid-range a bit more than was natural, so vocals seemed more "in-your-face" and at times a bit overbearing to me.
OPA2604: The most polite, non-fatiguing, but also the least dynamic and exciting with the smallest soundstage. Things seemed more compressed/veiled with this opamp in this particular player.
I wasn't able to verify (with lab equipment) that all opamps were running completely stable, though I assume at least the 5532 was since this was the original opamp used in the player. I just plugged them into the sockets I installed.
After I settled on the LM4562, I did a final comparison against the analog outputs of my Denon DVD-3910 universal player, as well as the digital out of the same Denon player going into a Schiit Bifrost Uber DAC.
The Bifrost was more punchy and dynamic compared to the analog outputs of the 3910 - so it made more of an impression, though more punch and prominent high end doesn't necessarily mean the most natural. The modded Sony CDP-950 really didn't give up anything against either other option - had plenty of bass weight (though not as much as the Bifrost) and detail and sounded very natural and pleasant. I didn't have a strong preference for any particular option of the three - it would probably depend upon the source material.
As far as the rest of the system, the CD player went into a Conrad Johnson PV-5, which fed a B&K EX442 amp driving IMF RSPM IV's.
In order of preference, I liked:
LM4562: The most dynamic, detailed, and among the largest presented soundstage. On a relative basis, it could be said to be on the brighter side, but overall I didn't find it fatiguing. This is the one I'll keep in.
JRC5532DD: Actually not too far from the LM4562, but it seemed a bit harsher at times and less natural sounding.
OP275: Seemed to accentuate the mid-range a bit more than was natural, so vocals seemed more "in-your-face" and at times a bit overbearing to me.
OPA2604: The most polite, non-fatiguing, but also the least dynamic and exciting with the smallest soundstage. Things seemed more compressed/veiled with this opamp in this particular player.
I wasn't able to verify (with lab equipment) that all opamps were running completely stable, though I assume at least the 5532 was since this was the original opamp used in the player. I just plugged them into the sockets I installed.
After I settled on the LM4562, I did a final comparison against the analog outputs of my Denon DVD-3910 universal player, as well as the digital out of the same Denon player going into a Schiit Bifrost Uber DAC.
The Bifrost was more punchy and dynamic compared to the analog outputs of the 3910 - so it made more of an impression, though more punch and prominent high end doesn't necessarily mean the most natural. The modded Sony CDP-950 really didn't give up anything against either other option - had plenty of bass weight (though not as much as the Bifrost) and detail and sounded very natural and pleasant. I didn't have a strong preference for any particular option of the three - it would probably depend upon the source material.
Interesting results. The 4562 is what might be considered the "true" successor to the 5532. The 2604 can be one of the most "musical" or lets say subjectively pleasing devices in many set ups... but its all a matter of degrees.
There shouldn't be any stability issues in the typical Sony arrangement with any of those devices.
There shouldn't be any stability issues in the typical Sony arrangement with any of those devices.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.