Looking for mod suggestions on Sony NC650V or NS500V
Oct 3, 2002 at 9:39 PM Post #31 of 112
(he did check for offset puppyslugg) The headphone amp IC is IC207 on the main board schematic. It then feeds the schematic in the "headphone detail" schematic. It looks like it would be easy to improve the headphone output to the level of a basic headphone amp that alot of people build called "Cmoy" after the guy who first suggested it over on "Headwise". There are some coupling caps that might be eliminated there also. C290&291. Also some mute transistors and current limiting resistors etc. That can probably goo too. Do the same check for dc on those caps before eliminating them.

http://headwize.com/
Take a look at headwise, in the "library" for a link to the "projects" section for the basic headphone amp using the opa2134 for some ideas.

As I was going to bed last night it pop-ed into my head, that this is a dvd player, and they almost always have switch mode power supply's, and the schematic you have shows a switch mode power supply. That adds a layer of extra caution that must be used as the main input to the power supply is from the 110vac power line. Are you comfortable with the chance of getting a nasty shock?
 
Oct 3, 2002 at 9:57 PM Post #32 of 112
The front end of the switching supply consists of a noise filter that then feeds the line voltage directly into a bridge rectifier made up of diodes D11,12,13,14. which output dc to the main filter cap C5 which is a 220mfd 200volt electrolytic. That cap charges to somewhere around 180volts dc, which then feeds the rest of the power supply. I will go poke around and see if the other mods people are doing include changing the diodes out, as I don't know if low noise diodes are avaliable at that voltage. The C5 cap is a good canidate for an upgrade to a audio grade filter cap and the only one I would try here is the "Blackgate Power Tank". I think that cap is around $35 or $40 bucks. (I don't think changing these diodes will be worth doing, if you look at the schematic, you will see alot of noise filters throughout the power supply low voltage circuits that will most likely take care of any diode switching noise. If you had a linear style supply like most cd players, the diode change is a bigger issue.)
 
Oct 3, 2002 at 10:14 PM Post #33 of 112
I just reviewed the thread and I think that cap (Blackgate Power Tank) is not going to be such an improvement that it would be worth blowing your modification budget on it. I think the difference would be subtle compared to the mods we talked about. Output caps and output chips are the biggest difference. As for the resistors, I would suggest radio shacks 1/4 watt assortment (part # 271-309) as being the budget alternative to Holco's (etc.) It is my opnion that the difference in the sound of resistors is also a subtle change. Somne people are happy with stock carbon film resitors. One change that may be worth considering is the filter caps located at the power supply rails of the ouput IC's. Blackgate or Cerafine caps in the voltage needed are not too espensive.

(in Batman announcers voice-"Later that day in the Budgie Cave")

Weird- I don't see local decoupling caps on the power supply rails near the opamps, There are three filter caps in the main power supply low voltage circuits C101,301,401 that could be changed out to better caps. But I am used to seeing them also right at the IC's.
Anybody else see them?
 
Oct 5, 2002 at 4:20 PM Post #34 of 112
Sorry for the long delay--I had an exam *AND* my speakers got here! (Diva Swan m200) At first they sounded very wooden but now they sound much better (9 hours of break-in so far). The silver wire seemed to especially help bass clarity.

Back to business...

I had overlooked the headphone opamp & circuit! I'll be sure to clean it up sometime soon... maybe I'll order my MDR-7506 headphones sooner, because it's much cheaper without a separate amp! The circuit does seem to resemble the CMoy pocket amp quite a bit, just different resistor values...

Both my AD8620 and OPA2134 samples are here! Unfortunately, I haven't yet gotten the decoupling caps I need alongside the AD8620, like the ones here (taken from this thread: )

attachment.php


About the power supply:

Those 3 capacitors--I looked at them, and found out that two of them are 35V 330uF and one 25V 680uF. Can all three be safely replaced with 25V 1000uF caps, or must I get a >35V cap to replace the 35V one? I am asking because the 50V Black Gate costs ~$10 more than the 25V version.

About the resistors--I already had ordered a collection of Holco resistors, so that's no longer a problem.
 
Oct 5, 2002 at 7:23 PM Post #35 of 112
The schematic shows voltages of 12 to 13 volts at those points in the power supply so I would think 25vdc would be fine. It might be worth double checking the voltage with a meter, if possible, just to be sure. That seems a little low, so it might be a minimum allowed or something elde. Better to be safe!

Let us know what your impression of the chips are when you get them changed.

So there are decoupling caps at the opamps. I must have been looking on the wrong page of the schematic.Those will be the main ones to replace with better caps. (Bang for the buck)

I finally went and read the whole thread over on the DIY forum and think that is an excellent set of tweeks for this player. Sounds llike these guys know there stuff and have access to tools the rest of us dream about.
 
Oct 5, 2002 at 7:35 PM Post #37 of 112
Hmm, this thread is very interesting. Do you think I should look under the hood of the Denon DCM-370 and replace the opamp with an AD8620 and remove the output capacitors?
 
Oct 5, 2002 at 7:49 PM Post #38 of 112
I would, if it were my Denon 370 .
biggrin.gif


I am not implying that the 370 is crappy or anything, I just feel you could probably improve the sound of the 370 with some realtively easy mods. Opamps and caps are the first thing I would look at, FWIW.
 
Oct 5, 2002 at 8:04 PM Post #39 of 112
Just make sure that you can remove the caps without having dc present in the output. It will depend on the design of the circuit and not all caps can safely be removed.

Leo- looks like what he did was remove the factory decoulpling caps and solder in some 1 mfd blackgates. Not sure I agree with that, but it worked for him.

You know, it looks to me like it would be easy to eliminate the switch mode side of the power supply. You would need three supply voltages +12/-12 and +5 vdc. Eliminate transformer T1 and feed the new raw supply to the secondary connection points. You would probably need to put the new supplies in an outboard chassis and use an umbilical to connect them. Makes me think I should buy the service manual for my dvd player. Interesting.
 
Oct 6, 2002 at 7:40 AM Post #40 of 112
About Panasonic SMT FK caps, Digi-Key only seems to sell 1000+ quantities, so I'll stick with Black Gate.

A couple capacitor questions:

1) Can I get the cheaper "standard" Black Gate ones (on the left here), or the "non-polar" ones, like on the right? Standard 1000uF ones are much cheaper.

2) I am thinking of biasing the AD8620 into Class A when I install it, maybe per these instructions (posted by the SACDmods guy, I believe). Alternatively, I could try this this more challenging method. Will this require me to install coupling capacitors once again (the ones I already bypassed)?

Final tidbit regards power supply modding:

SACDmods mentions the same exact 3-cap replacement. But they furthermore mention the following one:

Quote:

3-API/DELVAN 330uH Inductors installed on the power supply board to form a LC filter for high frequency noise emanating from the switching power supply. These and the Black Gate capacitors lower the noise in the DC power supplying components crucial to the sound such as the DAC and op-amps.


Would you say this is worth pursuing?

And about eleminating the switch mode part of the PSU... Budgie, you seem to essentially suggest creating a linear power supply? It sounds pretty hard, even the DIYaudio guys didn't actually get around to it.
Thanks,
Leo
 
Oct 6, 2002 at 8:30 AM Post #41 of 112
Quote:

About Panasonic SMT FK caps, Digi-Key only seems to sell 1000+ quantities, so I'll stick with Black Gate.


You are mistaken, they are sold in singles. I only suggested the Pana. FK's if you were looking a cost effective alternative.
Quote:

I am thinking of biasing the AD8620 into Class A when I install it, maybe per these instructions (posted by the SACDmods guy, I believe).


That is one way to do it, but not necessarily the best and only way.

http://www.tangentsoft.net/audio/meta42/tweaks.html

Quote:

And about eleminating the switch mode part of the PSU... Budgie, you seem to essentially suggest creating a linear power supply?


I assume Budgie and many others (dorkus) don't like switching supplies because they introduce noise and shouldn't be used for audio. FWIW, if it were mine NV500S I would do the linear psu for the audio section. But maybe it's just me.
 
Oct 6, 2002 at 8:57 AM Post #42 of 112
puppyslugg, would you recommend the cascaded-JFET biasing method, (link ppl's thread) instead? I think I'll need help calculating the resistor/JFET values for the AD8620...

I would be glad to do the linear-PSU mod, if you guys are confident it will help and willing to assist me along the way. As a way of contributing back/showing my appreciation, I will write up an article/post with pictures of all my mods. The only problem about the linear supply is, I would prefer to keep everything inside of the player.

By the way, how can I tell if I'm suffering from switching noise? I created a 3-minute silent CD track, and turned up the volume to 100% in my speakers. Holding my ear to them, I barely detect a noise (which does sound kinda electrical). This noise is probably well below -100dBA--at any reasonable volume, I don't hear it even holding my ear to the speaker.

PS: About my new speakers (Diva M200 ), I have broken them in for 20 hours now; silver signal tape actually improved the sound clarity. Very warm sound, to the point where highs and lows are (IMHO) understated--I'll need to get used to it more. The Divas work nicely with the NC650V, however.
 
Oct 6, 2002 at 3:49 PM Post #43 of 112
The regular polarized caps are what you want for the power supply.

The chokes may be an improvement. Depends on getting the filter value right for the freq. of the noise your filtering. Otherwise it's hit or miss. You can cause current limiting issues when you start sticking chokes in, though.

Right about here is were we start the discussion about the use of switch mode supply's for audio. Alot of people believe they must contribute noise and are not acceptable. I believe that the manufacturers use them because they cost less then linear supplies, but I don't believe they are always noiser then a linear supply. It would be a worthwhile experiment, but I think it is very hard to predict the results of such a mod.

You should be able to leave the coupling caps out, with the class A mod. The opamp will try to keep the output at zero, which is how this mod works. It keeps half the opamps output turned on all the time, to cancel out the dc from the class A mod. This shifts the operation point up to the remaining half of the output circuit, which will be running class A. I would check with a meter again to be certain all is okay!

The bias into class A sounds like a good idea to me, but I have not got around to trying it yet. People around here, who have done it, report good results, and I only remember one person who did not like it. I think I would try it with a simple resistor temporarily tacked in and see if it sounds like the dirrection you want to go with the mods. before going for the cascode jfets.

Damn them Diva's are good looking speakers!
 
Oct 6, 2002 at 7:52 PM Post #44 of 112
Quote:

would you recommend the cascaded-JFET biasing method, (link ppl's thread) instead?


It is the same one on tangent's site. With the ad8620, the benefits maybe marginal...they sound pretty good without class A biasing. But it will reduce switching noise and class A is the choice of audiophiles...right?
biggrin.gif
And as Budgie mentioned, it may help you remove output coupling caps, another plus.

As far as the psu is concerned, if you don't want to go 'outside the box', just do what you can with the stock psu via cap swaps, etc.
 
Oct 7, 2002 at 12:58 AM Post #45 of 112
I researched the AD8620/Class A biasing some more, and indeed it seems to offer few benefits like you (puppyslugg) said.

So I'll stick to what works:

(done) Bypass output coupling caps
(done) Remove muting transistors

1) Replace the ceramic decoupling caps (next to opamp) with 1uF Black Gates (non-polar)
2) Replace the line-out opamp with AD8620
3) Replace the caps in the power supply with Black Gates (polar) 1000uF 25V ones
4) Replace the 470-ohm output resistors (R239/R240 before the headphone tap, and R307/R309 after the tap) with Holcos, maybe 50 ohms or so? Chad (of DIYaudio) went with 10ohms and Marc went with 100. Any opinions on this?

Headphone mods: I'll postpone them by a few weeks (until I get the extra money for MDR-7506 headphones), but they will be very similar. It's only a question of deciding between OPA2134 versus AD8620 for the headphone opamp.

I'm going to go ahead and order the Black Gates... the Holcos are on their way.
Leo
 

Users who are viewing this thread

Back
Top