Long post, bear with me. I just thought it was about time to say something after all I've been learning here by lurking.
I had never done much in the way of modding till recently when my Philips 963sa's transport gave out. An old Sony ns500v came out of its closet retirement as a replacement, but I was surprised at how muddy and unsatisfying it sounded by comparison. Another (not entirely compatible) transport pulled from a cheap DVD player made the 963sa able to play music again, but by then I was ready to make it a goal to see if I could get the Sony to match it.
With a little email tutoring from a more experienced modder, I ordered parts from Digikey - AD8066 and OPA2134 opamps, a couple of OP275s more or less just to practice SMD soldering with, and a handful of capacitors. Concentrating on the output end, I put in the OPA2134 for the stereo opamp and two pairs of Panasonic SU bipolars for coupling (head-to-front in parallel, 33uf each for 66uf). The panasonic caps might be a quirky choice if they were serious, but in fact I just got a bag of them because they cost less than nothing; they got me past digikey's handling charge. I added a 5.6kohm resistor from each of the outputs back to the -11v line for cheap class-A, fired the thing up, and got an immediate comment on the sound from my normally inattentive wife, of all people. As much as modders need to look out for placebo effects, if someone who doesn't know what you've been doing wants to know why all of a sudden it sounds so good, I think it's comfirmation of a step forward.
A Percy Audio order got me some decent audio-grade capacitors and stealth diodes. I put the hexfred diodes on the +/- 11 volt lines in the power supply secondary, and added 330uf muse KZ caps in place of the 47uf stock caps at c513 and c213. It seems to have tightened up the bass and lower midrange, but not in a slap-up-side-the-head obvious way. Also had a hexfred in place of (I think it was) D311 but had to pull it back out and replace it with the stock diode, as that line gets fed back through the optocoupler and I guess the hexfred drops voltage a little bit - the whole supply compensated and ran high on other lines as a result. The LCD display had some inactive element bleed through, and there were times the player would suddenly go into standby while playing. Restoring the stock diode made those symptoms go away. Back on the output end, replacing the panasonic coupling caps with black gates didn't seem to make an immediate improvement - more on that later.
The plan next was to try the AD8066 in place of the 5.1 front channel output. Unfortunately I left a solder bridge (not between pins, I'd checked that with a meter, but from one of the corner pins to a nearby pad) that burned the chip up - it ran hot and sounded distorted. After fixing the solder, the chip still didn't work, so I tossed it. Oh well. Digikey was out of stock on the AD8066 when I tried to get another, so having heard intriguing things here and there about the OPA2107, I got one of those. Now that's in place, also with resistors from outputs to the -11v rail, and playing through BG coupling caps.
So I can A/B on the two sets of outputs under the assumption that depending on the menu settings in the sony, the 5.1 front output is close to identical to the stereo signal (don't really know whether that's true). At first I thought I still liked the sound from the much less expensive OPA2134 - it's really sweet, very nice on strings, everything well delineated and clean - but think I'm coming to prefer the OPA2107, which is also nice on the top end but plays with more authority and presence throughout. At some point I may get around to trying the OPA2134 through the black gates again (and/or the OPA2107 through the panasonic caps, for the heck of it), because that's another variable and I don't really know yet what effect if any it is having. It could be the BG caps sound better now because they are breaking in, or they're a better match for that chip, I don't know.
It didn't take a lot of money to improve the Sony to the point where it sounds to my ears quite a bit better than the stock Philips player - of course, experimenting (instead of making one decision and sticking with it) means spending a little more, but not nearly as much as on an expensive player, or a professional mod. For those who feel too nervous to start, I want to say that apart from a little surface-mount soldering, there's nothing scary about this - I did manage to roast one IC, but the board survived my clumsiness. I disobeyed the standard advice about keeping the other mods (extra capacitance, bias resistors) physically close to the opamp, because I wanted to keep the surface mount work to a minimum. The results are Good Enough For Me For Now. And there's still room to do more if I want to get into the area of the clock and D/A conversion, which I haven't thought about at all yet.
I kind of wish that Percy carried surface mount op-amps (he says he would if there was more demand), or Digikey carried specialty capacitors. As it is, working on one player in this particular way can mean falling a little under the minimum order at both places, so you end up picking up a few more parts you don't know whether you'll need. Is there anybody who carries all this stuff in one place?
Anyway, really a good experience - I'm happy with the sound, at least as a first-timer dealing with a player I didn't know what to do with a couple of years ago and was ready to give away. I'm enjoying my music more again.
Go on, tell me everything I did wrong.
I had never done much in the way of modding till recently when my Philips 963sa's transport gave out. An old Sony ns500v came out of its closet retirement as a replacement, but I was surprised at how muddy and unsatisfying it sounded by comparison. Another (not entirely compatible) transport pulled from a cheap DVD player made the 963sa able to play music again, but by then I was ready to make it a goal to see if I could get the Sony to match it.
With a little email tutoring from a more experienced modder, I ordered parts from Digikey - AD8066 and OPA2134 opamps, a couple of OP275s more or less just to practice SMD soldering with, and a handful of capacitors. Concentrating on the output end, I put in the OPA2134 for the stereo opamp and two pairs of Panasonic SU bipolars for coupling (head-to-front in parallel, 33uf each for 66uf). The panasonic caps might be a quirky choice if they were serious, but in fact I just got a bag of them because they cost less than nothing; they got me past digikey's handling charge. I added a 5.6kohm resistor from each of the outputs back to the -11v line for cheap class-A, fired the thing up, and got an immediate comment on the sound from my normally inattentive wife, of all people. As much as modders need to look out for placebo effects, if someone who doesn't know what you've been doing wants to know why all of a sudden it sounds so good, I think it's comfirmation of a step forward.
A Percy Audio order got me some decent audio-grade capacitors and stealth diodes. I put the hexfred diodes on the +/- 11 volt lines in the power supply secondary, and added 330uf muse KZ caps in place of the 47uf stock caps at c513 and c213. It seems to have tightened up the bass and lower midrange, but not in a slap-up-side-the-head obvious way. Also had a hexfred in place of (I think it was) D311 but had to pull it back out and replace it with the stock diode, as that line gets fed back through the optocoupler and I guess the hexfred drops voltage a little bit - the whole supply compensated and ran high on other lines as a result. The LCD display had some inactive element bleed through, and there were times the player would suddenly go into standby while playing. Restoring the stock diode made those symptoms go away. Back on the output end, replacing the panasonic coupling caps with black gates didn't seem to make an immediate improvement - more on that later.
The plan next was to try the AD8066 in place of the 5.1 front channel output. Unfortunately I left a solder bridge (not between pins, I'd checked that with a meter, but from one of the corner pins to a nearby pad) that burned the chip up - it ran hot and sounded distorted. After fixing the solder, the chip still didn't work, so I tossed it. Oh well. Digikey was out of stock on the AD8066 when I tried to get another, so having heard intriguing things here and there about the OPA2107, I got one of those. Now that's in place, also with resistors from outputs to the -11v rail, and playing through BG coupling caps.
So I can A/B on the two sets of outputs under the assumption that depending on the menu settings in the sony, the 5.1 front output is close to identical to the stereo signal (don't really know whether that's true). At first I thought I still liked the sound from the much less expensive OPA2134 - it's really sweet, very nice on strings, everything well delineated and clean - but think I'm coming to prefer the OPA2107, which is also nice on the top end but plays with more authority and presence throughout. At some point I may get around to trying the OPA2134 through the black gates again (and/or the OPA2107 through the panasonic caps, for the heck of it), because that's another variable and I don't really know yet what effect if any it is having. It could be the BG caps sound better now because they are breaking in, or they're a better match for that chip, I don't know.
It didn't take a lot of money to improve the Sony to the point where it sounds to my ears quite a bit better than the stock Philips player - of course, experimenting (instead of making one decision and sticking with it) means spending a little more, but not nearly as much as on an expensive player, or a professional mod. For those who feel too nervous to start, I want to say that apart from a little surface-mount soldering, there's nothing scary about this - I did manage to roast one IC, but the board survived my clumsiness. I disobeyed the standard advice about keeping the other mods (extra capacitance, bias resistors) physically close to the opamp, because I wanted to keep the surface mount work to a minimum. The results are Good Enough For Me For Now. And there's still room to do more if I want to get into the area of the clock and D/A conversion, which I haven't thought about at all yet.
I kind of wish that Percy carried surface mount op-amps (he says he would if there was more demand), or Digikey carried specialty capacitors. As it is, working on one player in this particular way can mean falling a little under the minimum order at both places, so you end up picking up a few more parts you don't know whether you'll need. Is there anybody who carries all this stuff in one place?
Anyway, really a good experience - I'm happy with the sound, at least as a first-timer dealing with a player I didn't know what to do with a couple of years ago and was ready to give away. I'm enjoying my music more again.
Go on, tell me everything I did wrong.
This Sony is the same as the 900v .
I tried many things.
Now I like best the set up with the dac going direct to a transformer.
The AKM sharp filter along with the transformer roll-off is enough filtering. See.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=52449&perpage=10&pagenumber=2
Also beef up the power supply caps, local bypassing at each chip, steel plate to corner the switcher from the unit and get the caps in series w/ the audio out.
Lewis
I tried many things.
Now I like best the set up with the dac going direct to a transformer.
The AKM sharp filter along with the transformer roll-off is enough filtering. See.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=52449&perpage=10&pagenumber=2
Also beef up the power supply caps, local bypassing at each chip, steel plate to corner the switcher from the unit and get the caps in series w/ the audio out.
Lewis
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.