Denon DCD-2700 CD Player Mod Thread

Ok, so here are my first impressions. Purely subjective opinions.

After removing IC 309, 310 (the filters) there was an immediate improvement in mids and highs. BUT. I didn't need to turn the volume up as much to get the sale loudness, and there was a little bit of harshness present. Total guess was that without the filter present that IC313, 314 (the last op-amps in the chain) were providing gain that was no longer needed. I actually only listened to a few songs before I stopped and pulled those op-amps and replaced them with the jumpered empty plugs. And then I swapped the "tin can" HA/NOPB 49720's into positions into IC 311, 312.

This is much better. I need a little more volume to get the same loudness as before now. But there is more overall clarity, especially in the mids and highs. Vocals are clearer. I like this.

I think that I've effectively showed that at the very least those 2 op-amps per channel are unnecessary, and that removing them can be an improvement. I can say this, there's no way that I will be putting those 4 op-amps back in the player.

DSC03952.jpg DSC03953.jpg

From here I'm going to concentrate on the power supply for a while. I've been reading a lot of talk about the benefits of replacing the bridge rectifier diodes in the analog section (1SR35-200) with Schottky diodes or fast/soft recovery diodes. The 11DQ10 Schottky seems to have been a favorite but is no longer being made. As it happens I do have 4x Vishay SBH100-E3/54 diodes on hand, which are supposed to be a good sub for the 11DQ10's. These will probably be going into my player in the near future. And whenever I make my next Mouser order I'll be ordering what I need to do a voltage regulator upgrade. Then I'll start measuring what it would take to make a torroidal transformer or 2 fit. This is proving to be a lot of fun.

Cheers,
James

http://www.mouser.de/ProductDetail/.../SB1H100-E3-54/?qs=u91jd7aysvMXltu%2bMMJLmQ==
 
Last edited:
So I just listened to Grateful Dead's Workingman's Dead and American Beauty. Both albums sounded fantastic but listening to "Friend Of The Devil" I really got sucked in to the bass guitar. Realized that this is the first time that I can remember hearing the bass guitarist's finger work. Sound quality across the spectrum is better since removing the filter and final output op-amps. I like this change.

Cheers,
James
 
Last edited:
My basic English and lack of knowledge doesn't allow me to understand most of the informations you linked, but if its found that without filter the sound is better ( has no noise or level of the noise is acceptable,
at the same time we'll get an increase in dynamics, more natural sound etc.) That wil be very interesting and desired for me.

BTW, I decided that I'll begin improvements form the begining of the power path, the power cord and power supply , and this is a litle paninfull ;) I have to wait a month minimum.
I'm also interested LClock X03 and ZAP filter2 linked by Unfarilane.

Edit: Ooo I didn't notice your posts above

PS Interesting thoughts about Power Amplifier Supply
http://www.tnt-audio.com/clinica/ssps1_e.html
 
Last edited:
Well I don't understand much of what is in those documents either. :)

I think that concerning the op-amps the important thing is to install dip-8 sockets in each position. Then you have the flexibility to install or remove different op-amps very easily. That way if you find that you like the sound with the filters better it is easy to re-install them.

I have decided to install fast soft recovery diodes in the psu rather than the Schottky diodes. Mainly because the Schottky's have less voltage drop, which sends unneeded extra voltage through to the voltage regulators, which then just turn the extra voltage into heat. If the recommendation from Unifarilane and others is that a more powerful power supply is needed then it makes little sense to throw away power right at the start.

Last year Nashou66 recommended that I try out some Vishay SBYV27 ultrafast soft recovery diodes when I was working on my CX-600 project. I had completely forgotten that I bought 100 of them and still have lots of them. I bought the 100v reverse voltage version but I don't think that that should be a problem as a sub for the 1SR35-200. I think that if I get more than 100v reverse voltage on a 12v circuit that con only mean that something catastrophic has happened and there is full mains voltage coming backwards through the cdp. And considering that mains voltage is usually around 235v here a 200v reverse current on the diode will cause it to fail anyway.

Other than that the SBYV27 is 2A load rated instead of the 1A of the 1SR35, and has a 15ns recovery time rating, which if I understand correctly is very close to a Schottky. I'm going to try them out, hopefully on the coming weekend. If not then then at least as soon as possible. I think that they look really good and can hardly wait to do this.

Also, now that I looked at the psu more carefully I think that my choice of replacements for the 2x 2200µF/25v caps in the digital section of the pseu wasn't the best choice. Staying with 25v instead of going with 35v lets me choose a 2700µF cap wiith better ripple current specs, so when I make my next Mouser order I'll be looking into buying something in the 2700µF range.

Cheers,
James

http://www.mouser.de/ProductDetail/Vishay-Semiconductors/SBYV27-100-E3-54/?qs=sGAEpiMZZMtbRapU8LlZD0HbIjlpuZ44EG2VygAavVo=
http://www.mouser.de/ProductDetail/Vishay-Semiconductors/SBYV27-200-E3-54/?qs=sGAEpiMZZMtbRapU8LlZD0HbIjlpuZ445lIBCY%2bUM%2bs=
 
Last edited:
I think that concerning the op-amps the important thing is to install dip-8 sockets in each position. Then you have the flexibility to install or remove different op-amps very easily. That way if you find that you like the sound with the filters better it is easy to re-install them.

Thad's good idea, I'll follow this.

Other than that the SBYV27 is 2A load rated instead of the 1A of the 1SR35, and has a 15ns recovery time rating, which if I understand correctly is very close to a Schottky. I'm going to try them out, hopefully on the coming weekend. If not then then at least as soon as possible. I think that they look really good and can hardly wait to do this.

My Intuition says that soft recovey diodes are better choice.

Also, now that I looked at the psu more carefully I think that my choice of replacements for the 2x 2200µF/25v caps in the digital section of the pseu wasn't the best choice. Staying with 25v instead of going with 35v lets me choose a 2700µF cap wiith better ripple current specs, so when I make my next Mouser order I'll be looking into buying something in the 2700µF range.

My Intuition says also that It's worth to increase the capacitance of filter caps, 10000 uF for analog section and 4700 uF for digital ( there is place on the psu board to part the 2200/25 caps - I mean insert the caps with a larger diameter/capacitance and fixing them with hot glue ) I will try this.

PS
" Yesterday I listened my DCD 2700 first time.. btw This is my come back to CD because I prefer Vinyl and Tube pre/amplification; analog sound generally. DCD is connected with Philips AH 686 witch has soft/analog sound, my Turnable was connected with DIY full tube preamplifier and Mingda MC 845A tube amplifier and it's my point/source of reference. Returning to the point; comparing DCD 2700 and DCD 920 I used ealier, DCD 2700 has beautiful low/bass base which is very deep and varied, scene/ stereophony is similar in size but in DCD 920 scene is flat I mean sounds aren't three-dimensional, are short ( there isn't much space around them, I mean space in which they can and they do sound/flow, becoming three-dimensionall), no so ringing/juicy/full harmonics comparing to the tube sound. DCD 2700 is very close to this !, has very analog rich sound but... :) too rich in bass/low mids for me. In my opinion it obscures/covers higer mids and highs.Returning to the context of this thread I can not wait to move with moding :), cos I feel that this player has potential, btw , my Intuition chose it for me :)"
I'm not shure but I hope I expressed my thoughts understandable for english speakers.
 
Last edited:
I understood your thoughts quite well. You are talking about sound stage and depth.

Your impressions of the 2700 are very much like mine when I first listened to it. I felt that it was lacking in upper end presence and clarity when compared to my modded DCD-1290. But the 2700 has much better bass/midbass and has the potential to go much further than the 1290.

My intuition tells me that you will be able to install sockets for op-amps in your 2700 within a few days. ;)

Cheers,
James
 
Hi unfairlane (if you are still here),

What have you used or recommend for diode upgrades? Especially the bridge rectifier diodes but also in general any others also.

Cheers,
James
 
When I used to upgrade the CA 640P's I found that changing the diodes to fast switching Schottky types didn't make any difference I could detect YMMV. You have 0.1uf caps between pins 4 & 8 on the op amps, another thing you can try is putting a 4.7uf or 10uf quality electrolytic in parallel with them, I have been doing this and find it does make a worthwhile difference. Just make sure you get the polarity right.
 
I understood your thoughts quite well. You are talking about sound stage and depth.
***
My intuition tells me that you will be able to install sockets for op-amps in your 2700 within a few days. ;)

Thank you James for claryfying my thoughts in english :)

Intuition comes from higher Source/Self , I trust and follow her after made contact with her and I'm going to take advantage of her support in moding, She is priceless in everything also in selecting elements :)

Cheers
Slawek
 
Last edited:
When I used to upgrade the CA 640P's I found that changing the diodes to fast switching Schottky types didn't make any difference I could detect YMMV.

:) My mileage was actually different.

After listening to the cdp for over a week now without the filter 2 things stood out. There was a bit of upper end harshness, just enough to cause listening fatigue. So I guess I found the reason they added the filter. But they threw the baby out with the bathwater because without the filter the overall listening experience was a good deal better. Just have to deal with that harshness.

So today I did the diode swap. All 4 diodes in the analog bridge and also the 2 in the digital section. Actually, first I did it in a DCD-1290 yesterday and liked what I heard. And the results are pleasing. That upper end harshness is gone. The sound is just cleaner. I like this. It sounds better.

DSC03981.jpg DSC03984.jpg

And now I'm going to go out front and listen to some more music...

Cheers,
James
 
I did an extended listening session last night and am very pleased with the results so far. (Rory Gallagher, Slade Alive, Working Man's Dead) The sound is very smooth now and these last few mods have improved depth and clarity considerably. My son will be bringing his 2700 over today for retrofitting of the diodes and removal of the filter op-amps.

Cheers,
James
 
Very interesting. It's known that power supply is very important. If your longer listening sesion will confirm these positive changes. It shows that the power supply is very important in all audio equipments, not only in amplifiers and trought improvements of source of power we get a lot of good.

EDIT :
After few days of listening, I noticed that DCD 2700 heats up strongly so for sure it will be good to change transformer for the better/stronger. I'm thinking about using two toroidal TRs mounted in separate box to remove the source of heat. I have a question, might anyone know exact parameters of TR used in DCD 2700 ?.

My plan is:
1 Change analog filer caps on 4700/50 Elna Tonerex and listen ( thanks James)
2 Change rectifier diodes in analog section on SBYV27 and listen ( thanks James :) )
3 Change rectifier diodes in digital section on SBYV27 and listen
4 Change analog filer caps on 10000/50 or 63V and listen
5 Change digital filter caps on 4700/25 or 50V and listen
6 Change transformer on two toroidal TRs and listen

I'm curious If I will be able to listen differences on each step. I will aid with Denon DRM 700A, recording the same testing material on it after each change.
 
Last edited:
Hi

I chnaged my plan , but first changes behind myself

upload_2017-3-13_1-8-59.png

upload_2017-3-13_1-9-27.png
upload_2017-3-13_1-10-22.pngupload_2017-3-13_1-11-5.png

More info today during day :)

***
So.. for many reasons I merged several earlier planed steps in one.
Changes:
- All rectifiers diodes changed to SBYV27-200
- Big cap filters changed to 4700/50v Elna Tonerex and 4700/25v Nichicon KW
- C-705: to 47µF/100v Panasonic FC,
- C-706: to 0.1µF/63v Wima MKS2,
- C-403: to 330µF/25v UCC KYB,
- C-245 to 100µF/63v FC,
- C-351 to 220µF/35v FC.
- I aded ( more correctly my Intuition added) 2x10uf 63v to C707 and C708; "-" conected to GND [edit; this was wrong, I changed polarization of this added caps as c701 i c702) , in the future I'll change them for better ( I was suprised :) and I put what I had, I feel/Intuition says that is important to put better ones).

After these changes is much better :) My wfie said that highs were up, sound is brither, more clearer.
In my ears: stage is bigger, but more important is that is much more deeper.., now every intstrument has space with his individual/independent dynamic (more efficient power supply) and space is clearer, before sounds appeared and disappeared in a black hole / veil , now veil is smaller, sounds are longer/more audible in the upper and lower levels and connected togeteher, I mean they penetrate each other to a greater degree. Subtleties appeared and yes! there's more color/timbre in sounds, voices are more present :). It's very interesting that my vintage Philips AH 686 is no more so restful :); animates him dynamic of DCD 2700.

Next week more changes, among other I'll change 2x4700/25 to 4x4700/25 Nichicon KW and 2x4700/50 Tonerex to 6x4700/50 SAMSUNG SMS. I don't know how I'll do it yet, but my I... says that we can do it :).
Ah yes; one more, I aded some damping material under rear panel of cabinet and decided that I'll build separate power supply in the future for It

Edit:
Small changes

4x4700/25v Nichicon KW -> 4x4700/35v Nichicon FW
6x4700/50v Samsung SMS ->6x4700/50v Elna LAO

Cheers
Slawek
 
Last edited:
Very interesting Slawek. Looks like you are having fun. :beerchug: I'll be reading and enjoying your posts.

Today I ordered some Texas Instruments versions of LM317 and LM337 voltage regulators and some assorted Vishay-Dale 1% metal film resistors. When they get here I'll be building some voltage regulator replacements, +12v, -12v and +5v.

http://www.mouser.de/ProductDetail/Texas-Instruments/LM317KCT/?qs=TB/Q0sBK/Gcz1btw204Pbw==
http://www.mouser.de/ProductDetail/Texas-Instruments/LM337KCSE3/?qs=iSMark9AYDWpc5pQ5gDgRw==

Cheers,
James
 
Yes I have fun and much more :). Very interesting James, I'm waiting for your thoughts/expieriences about improved volatage regulators, in turn I was thinking about adding stabilization to +-10V for servo, which DCD 2700 doesn't has...
I dream to deal with this type of work totally :), unfortunately I have to move this weekend planed work for next week.

Cheers
Slawek
 
A small update. With my last Mouser order I included a pair of UCC KYB 3900µF/25v caps for the digital section of the psu. These are the largest capacity KYBs that I could find that would drop in without any mods. Compared to the 2200µF/35v they have a half amp more ripple factor, 3.64A vs 3.14A. ESR is basicly the same, 0.017 vs 0.018.
http://www.mouser.de/ProductDetail/United-Chemi-Con/EKYB250ELL392MK40S/?qs=sGAEpiMZZMtZ1n0r9vR22UhSjrTM10DEmv5al1zEgSts8sf2MS/6Sw==

Today I changed out those caps and while I was at it I also changed C707, 708 & 709 (0.1µF/25v) from the original ceramic to Wima MKS2 0.1µF/63v. And Slawek supplied me with 6 uprated SBYV27-200 diodes so I swapped them in in place of the 100v versions.

While I was hoping as always for some more sweetness I really wasn't expecting a lot. And once again I have been surprised. More Sweetness! There is a very nice and noticable improvement. Clarity and definition are again improved. Instruments and vocals have more separation. They stand out more individually. With the harmonized vocals in Graham Nash's Songs For Beginners and the Dead's American Beauty each singer is more distinct and separate from the others. A very gratifying upgrade for less than €5 in parts.

I have to go back and listen to some more...

DSC04033.jpg DSC04036.jpg DSC04037.jpg

Cheers,
James
 
I've been working on the voltage regulators. As I said before, I ordered the Ti versions of the LM317 & 337 from Mouser, as well as some Vishay/Dale 1% metal film resistors. The write-up at Ray's Audio Page recommends against using super low ESR caps. I used Panasonic FC Elcos and Wima MKS2 film caps, I hope that they aren't to low in ESR. He also recommends using 1N4002 diodes. I have a bunch of 1N4007's so that's what I used.

Mouser's pages for the regulators also has links to the data pdf's. And I went online to get the data pages for the original NJM78xxFA and NJM79Mxx regulators to compare the pin-outs. Anyone with a good eye will note that my first drawing is wrong. I believe that the 2nd one is correct. I also added some heat sinks to the little boards. The mounting pads for the new regulators are connected internally to the middle leg so I used sil-pads and bushings to isolate the heat-sinks.

DSC04039.jpg DSC04041.jpg DSC04047.jpg DSC04048.jpg

Cheers,
James
 
Here are some shots of the finished boards. +12v, -12v and +5v. They won't win any beauty contests but I believe that they will serve their purpose. While I feel confidant that I made them correctly I'm not about to install them without testing. So I'll be taking them with me when I go back to work next week and bench test them with our 24v bench power supply.

DSC04053.jpg DSC04054.jpg DSC04062.jpg DSC04071.jpg DSC04083.jpg

Also.
After I replaced those 3x 0.1µF ceramic caps in the PSU I tested them. They ranged from 125nF to 83nF. This got me looking at the schematic again. I counted 16 more of the 0.1µF, 8x 0.01µF and a single 0.47µF ceramic cap that I could easily replace. The next time that I have the player opened up I'm going to replace them all with Wima MKS2, which I have a lot of. In spite of being rated at 5% and 10% tolerance they all check within 1-2%.

Another thing that I looked at. There is a 1.3k ohm resistor (R385, 386) in the signal path right after the branch that inputs to those BA15218 filter op-amps that I removed. I'm speculating that they are there as a speed bump to cause (some of?) the signal to go into the filter by-pass. If so, with the filter op-amps removed they would no longer have a useful purpose and I'm contemplating removing and jumpering them.

Cheers,
James
 
I had some time on my hands yesterday so I opened up the box and replaced all of the 0.1µF and 0.01µF 25v ceramic caps with Wima MKS2 63v. While I was in there I went ahead and removed and jumpered R385 and 386.

I can't really say that this made any hear-able improvements. On the other hand I can state that it didn't make any hear-able problems. And it didn't really cost very much. There are now only a few ceramic caps left in the player, probably less than 20. I'll be making a list of the values and when I make my next Mouser order will be ordering some polystyrene caps to replace the last of the ceramic caps. Photos later.

Cheers,
James
 
I have some time to add some news.

-replaced c709 for 10µF/35v ES Muse Bi-Polar I've got from James

upload_2017-4-5_14-26-47.png

- replaced C-703, 704 with 4x 4x4700/35v Nichicon FW
upload_2017-4-5_14-39-49.png

-aded 2,2 nF 300 V on AC input pins

upload_2017-4-5_14-44-0.png

and the same cap on 24/12V secondary winding

upload_2017-4-5_14-49-4.png


These was intuitive inspirations and I was suprised during listening sessions....
 
Back
Top Bottom