hi brian,
i've seen this fellow's mod pages. his mods are ok but i feel like they only scratch the surface... e.g. with the NS500V he tries to remedy the ills of the power supply with different tweaks but the real solution of course is to replace the whole bloody thing.
also i don't know why he replaces the coupling caps with Black Gates when they can be bypassed altogether (at least for the 500V - the 222ES i just got does have some offset). the LC Audio clock mod seems like a worthwhile improvement, but we can of course buy the clock module directly from LC.
incidentally, the 222ES is an excellent player, and the parts quality inside is actually quite nice... the beautiful Elna Cerafine and Nichicon Muse caps on the analog board show some care in the parts selection, and it is reflected in the sound quality as well. so far it does not quite get the incredible inner detailing, texturing, and imaging of the 500V but is better in pretty much every other area... we'll see how it breaks in. i recommend it highly, i hope you are able to get one - they are getting harder to find.
i've seen this fellow's mod pages. his mods are ok but i feel like they only scratch the surface... e.g. with the NS500V he tries to remedy the ills of the power supply with different tweaks but the real solution of course is to replace the whole bloody thing.
also i don't know why he replaces the coupling caps with Black Gates when they can be bypassed altogether (at least for the 500V - the 222ES i just got does have some offset). the LC Audio clock mod seems like a worthwhile improvement, but we can of course buy the clock module directly from LC. incidentally, the 222ES is an excellent player, and the parts quality inside is actually quite nice... the beautiful Elna Cerafine and Nichicon Muse caps on the analog board show some care in the parts selection, and it is reflected in the sound quality as well. so far it does not quite get the incredible inner detailing, texturing, and imaging of the 500V but is better in pretty much every other area... we'll see how it breaks in. i recommend it highly, i hope you are able to get one - they are getting harder to find.
yeah, I wasn't implying getting mods done by him. I was just showing what he had done. The LC clock mod would probably be a good idea.
I am trying to find a 222es now. It sounds like a better project, and doesn't have the video out for the dvd like the 500 does.
The SCD-CE775 might also be a good option if I cannot find a 222es.
How is the power supply in the 222es, should it be replaced as well? I noticed that it had dual transformers.
I wonder the effect of FREDs for the power supply.
I found a 222es for $399, but not $325, as you did. I am wondering if you can modify the 500v based on the differences between it and the 222es.
--
Brian
I am trying to find a 222es now. It sounds like a better project, and doesn't have the video out for the dvd like the 500 does.
The SCD-CE775 might also be a good option if I cannot find a 222es.
How is the power supply in the 222es, should it be replaced as well? I noticed that it had dual transformers.
I wonder the effect of FREDs for the power supply.
I found a 222es for $399, but not $325, as you did. I am wondering if you can modify the 500v based on the differences between it and the 222es.
--
Brian
hi brian,
yeah, i was not implying that you were implying to get the mods done by him either. thanks for pointing it out tho, sometimes i forget there are other people reading the threads around here...
the 222ES is probably a nicer machine to mod, since there is a little more room inside and the basic infrastructure is better to begin with. i will add however that the NS500V sounds mighty nice still and has decent mod potential... i switched the opamp to OPA2134 last night and it sounds very good, a bit smoother and warmer. the player is still brighter than the 222ES but i won't know its full potential until i replace the power supply. i still like the AKM DACs in the 500V an awful lot too, the BB chips in the 222ES sound more "common" to me.
HEXFREDs in the power supply will probably reduce switching noise, and the improvement is usually audible. bypass caps across diodes can do the same thing. when practical i like to use both techniques. there are 2 transformers inside but the 2nd one is a pretty dinky unit so upgrading the entire supply might still be a worthwhile project... a good outboard supply in particular would definitely improve the performance, and the analog rails could stand to be juiced up a little (they are either +/-11V or +/-7V i think).
i'm not sure if one could mod the 500V to sound like a 222ES. the DACs are different as i mentioned (not sure which is better), but the digital boards look very similar. the advantage of the 222ES is that the analog board is cleaner (no video), and the DAC chips are on it so there is more control over the entire DAC/output stage circuit. you could even replace the entire analog board if you wanted, maybe swap DACs or try a discrete circuit if room permits. interestingly, there is a separate clock crystal right on the DAC/analog board - i guess this is the "Direct Digital Sync D/A Clock Converter" feature Sony advertises. the transport mechanism itself seems to be higher-grade in the 222ES as well, the 500V is all plastic while the 222 has a metal sub-assembly (although the rest of the mounting is still plastic). it could make a great basis for a top-loader but that is a whole different project altogether probably.
the CE775 is a good bet if you want to save some money or can't find a 222ES. it is essentially the same w/o the 2nd transformer and maybe some cheaper parts here and there. i may actually pick one up myself, so i can maybe build an all-new player from scratch based on the spare parts but not have to hack up this ES unit. if you want to keep the basic player intact though the 222ES is slightly more convenient to mod as it has the provision for separate analog supply etc... it's also 3/4" taller than the 775, but not sure how much that translates into actual room inside.
yeah, i was not implying that you were implying to get the mods done by him either. thanks for pointing it out tho, sometimes i forget there are other people reading the threads around here...
the 222ES is probably a nicer machine to mod, since there is a little more room inside and the basic infrastructure is better to begin with. i will add however that the NS500V sounds mighty nice still and has decent mod potential... i switched the opamp to OPA2134 last night and it sounds very good, a bit smoother and warmer. the player is still brighter than the 222ES but i won't know its full potential until i replace the power supply. i still like the AKM DACs in the 500V an awful lot too, the BB chips in the 222ES sound more "common" to me.
HEXFREDs in the power supply will probably reduce switching noise, and the improvement is usually audible. bypass caps across diodes can do the same thing. when practical i like to use both techniques. there are 2 transformers inside but the 2nd one is a pretty dinky unit so upgrading the entire supply might still be a worthwhile project... a good outboard supply in particular would definitely improve the performance, and the analog rails could stand to be juiced up a little (they are either +/-11V or +/-7V i think).
i'm not sure if one could mod the 500V to sound like a 222ES. the DACs are different as i mentioned (not sure which is better), but the digital boards look very similar. the advantage of the 222ES is that the analog board is cleaner (no video), and the DAC chips are on it so there is more control over the entire DAC/output stage circuit. you could even replace the entire analog board if you wanted, maybe swap DACs or try a discrete circuit if room permits. interestingly, there is a separate clock crystal right on the DAC/analog board - i guess this is the "Direct Digital Sync D/A Clock Converter" feature Sony advertises. the transport mechanism itself seems to be higher-grade in the 222ES as well, the 500V is all plastic while the 222 has a metal sub-assembly (although the rest of the mounting is still plastic). it could make a great basis for a top-loader but that is a whole different project altogether probably.
the CE775 is a good bet if you want to save some money or can't find a 222ES. it is essentially the same w/o the 2nd transformer and maybe some cheaper parts here and there. i may actually pick one up myself, so i can maybe build an all-new player from scratch based on the spare parts but not have to hack up this ES unit. if you want to keep the basic player intact though the 222ES is slightly more convenient to mod as it has the provision for separate analog supply etc... it's also 3/4" taller than the 775, but not sure how much that translates into actual room inside.
How about the C555ES? I am wondering how it compares with the C222ES. I would be affaid to mod it, since it costs over $500.
It was posted that the SCD-C555ES features a better PC board with more discrete circuitry as well as an advanced Twin R-Core Power supply as compared to the C222ES.
How about an op-amp swap? Oade Brothers sells a modded C555ES, and the only modification is swapping the op-amps. They say provides a significant improvement in performance.
Is it possible to swap out the dacs? I have these dacs laying around here (samples friends and I acquired from Analog Devices):
2 - AD1862N
6 - AD1865N-J
6 - AD1865N-K
2 - AD1866N
and I have some opamp samples that I ordered. I was going to play around with dac chips with friends, but we never got around to it.
Fairchild has some new fast/soft stealth diodes that might be interesting:
http://www.fairchildsemi.com/news/2002/0206/stealth_diodes.html
(they have smaller ones then the 1200V ones, but I ordered the 1200V for my amp power supplies)
I requested some free samples from them, which should come soon.
Is there anyway to get service manuals for these cd players?
--
Brian
It was posted that the SCD-C555ES features a better PC board with more discrete circuitry as well as an advanced Twin R-Core Power supply as compared to the C222ES.
How about an op-amp swap? Oade Brothers sells a modded C555ES, and the only modification is swapping the op-amps. They say provides a significant improvement in performance.
Is it possible to swap out the dacs? I have these dacs laying around here (samples friends and I acquired from Analog Devices):
2 - AD1862N
6 - AD1865N-J
6 - AD1865N-K
2 - AD1866N
and I have some opamp samples that I ordered. I was going to play around with dac chips with friends, but we never got around to it.
Fairchild has some new fast/soft stealth diodes that might be interesting:
http://www.fairchildsemi.com/news/2002/0206/stealth_diodes.html
(they have smaller ones then the 1200V ones, but I ordered the 1200V for my amp power supplies)
I requested some free samples from them, which should come soon.
Is there anyway to get service manuals for these cd players?
--
Brian
the C555ES is a very nice player. vs. the 222ES:
- same DAC chips (BB DSD1702)
- larger, better DAC/analog board w/bigger decoupling caps
- larger analog PS transformer (555 uses R-cores for both)
- more sturdy "Frame and Beam" construction on 555 (222 chassis is still a little flimsy), with copper plating (should sound better than raw steel on 222)
if you don't mind spending the extra cash, i think the 555 is definitely a nicer player. having said that though, a modded 222 should easily out-perform a stock 555. but a modded 555 will be even better, just not sure how much. the better chassis is definitely worthwhile and the bigger analog PCB will give you more room to work with.
i guess the value judgement is up to you. some would argue the 222 is basically a CE775, so either get the 775 or 555. i can't argue with this but i like my 222. op-amp swap is easy enough to do in either player so i'm sure whatever improvement Oade gets w/the 555 you can replicate on the 222. just be very careful when removing the original SMT chips, damn Sony glues them down so i ripped a lot of traces up when removing them from my NS500V. not catastrophic though, you can do a decent patch job with some fine wire.
i have no idea if the AD chips are pin-compatible, somehow i doubt it but check out the DSD1702 spec sheet to confirm.
those new rectifiers look cool. maybe put up a new post asking people what they think about it?
p.s. i just read a quote of Dan Wright, a well-respected modifier, saying that the 555 uses 12 opamps total (6 dual ICs), with one chip per channel. my 222ES has only 3 dual ICs, so i'm not sure what's going on in the 555... 2 opamps per channel would suggest balanced outputs but i don't believe the 555 has them.
- same DAC chips (BB DSD1702)
- larger, better DAC/analog board w/bigger decoupling caps
- larger analog PS transformer (555 uses R-cores for both)
- more sturdy "Frame and Beam" construction on 555 (222 chassis is still a little flimsy), with copper plating (should sound better than raw steel on 222)
if you don't mind spending the extra cash, i think the 555 is definitely a nicer player. having said that though, a modded 222 should easily out-perform a stock 555. but a modded 555 will be even better, just not sure how much. the better chassis is definitely worthwhile and the bigger analog PCB will give you more room to work with.
i guess the value judgement is up to you. some would argue the 222 is basically a CE775, so either get the 775 or 555. i can't argue with this but i like my 222. op-amp swap is easy enough to do in either player so i'm sure whatever improvement Oade gets w/the 555 you can replicate on the 222. just be very careful when removing the original SMT chips, damn Sony glues them down so i ripped a lot of traces up when removing them from my NS500V. not catastrophic though, you can do a decent patch job with some fine wire.
i have no idea if the AD chips are pin-compatible, somehow i doubt it but check out the DSD1702 spec sheet to confirm.
those new rectifiers look cool. maybe put up a new post asking people what they think about it?
p.s. i just read a quote of Dan Wright, a well-respected modifier, saying that the 555 uses 12 opamps total (6 dual ICs), with one chip per channel. my 222ES has only 3 dual ICs, so i'm not sure what's going on in the 555... 2 opamps per channel would suggest balanced outputs but i don't believe the 555 has them.
FWIW
I believe the C555ES uses the stereo Burr Brown PCM1738. I did a search awhile back on the Hi-Res Asylum.
It is different than the stand alone DSD1702 (in which the DSD data can be directly input). It requires an external DSD digital filter to convert the DSD data to PCM that is then input to the PCM1738.
It's spec sheet looks great, similar to the PCM1730 that I was going to play with (until I got the itch for a Jocko board). The '1738 has differential current output and no opamps, so you can mess around a bunch with the I/V stage. Lotsa versatility.
I believe the C555ES uses the stereo Burr Brown PCM1738. I did a search awhile back on the Hi-Res Asylum.
It is different than the stand alone DSD1702 (in which the DSD data can be directly input). It requires an external DSD digital filter to convert the DSD data to PCM that is then input to the PCM1738.
It's spec sheet looks great, similar to the PCM1730 that I was going to play with (until I got the itch for a Jocko board). The '1738 has differential current output and no opamps, so you can mess around a bunch with the I/V stage. Lotsa versatility.
I have 3 samples of the PCM1738E. The only problem is that the only package that it comes in is the 28-SSOP, so it is not very useful without the converters.
You can get free samples from www.ti.com
They also carry digital filters that you can use with it. (I picked up the DF1706 and the DF1704, which are both said to work with it)
The chip is also used in the XA777ES, and is said to use the same circuits as specified in the datasheet:
http://www-s.ti.com/sc/ds/pcm1738.pdf
pages 22-24 describe the application with the relevant circuit
Relevent circuit diagrams screen captured from pdf:
http://brian.darg.net/dac/
Do you know which opamps, these sacd players use? Might be worth an upgrade.
Interesting bit about XA777ES (from Hi Rez):
In Reply to: So does the XA777ES have three PCM1738's or six PCM1738's posted by michi on December 12, 2001 at 12:29:51:
SIX dual-channel, differential output DAC chips. For stereo, three chips are paralleled (all six channels) for each of left and right. For multichannel (6-channel), each dual DAC has its two channels paralleled for each output.
The C555ES has only 3 such chips.
--
Brian
You can get free samples from www.ti.com
They also carry digital filters that you can use with it. (I picked up the DF1706 and the DF1704, which are both said to work with it)
The chip is also used in the XA777ES, and is said to use the same circuits as specified in the datasheet:
http://www-s.ti.com/sc/ds/pcm1738.pdf
pages 22-24 describe the application with the relevant circuit
Relevent circuit diagrams screen captured from pdf:
http://brian.darg.net/dac/
Do you know which opamps, these sacd players use? Might be worth an upgrade.
Interesting bit about XA777ES (from Hi Rez):
In Reply to: So does the XA777ES have three PCM1738's or six PCM1738's posted by michi on December 12, 2001 at 12:29:51:
SIX dual-channel, differential output DAC chips. For stereo, three chips are paralleled (all six channels) for each of left and right. For multichannel (6-channel), each dual DAC has its two channels paralleled for each output.
The C555ES has only 3 such chips.
--
Brian
hmm, interesting. are you sure you need an external filter for DSD on the PCM1738? figure 11 on page 26 of the spec sheet seems to show a direct DSD input, unless i am misinterpretting it.
i did not realize the PCM1738 was a 24-bit ladder DAC. i would have thought to take full advantage of SACD you'd want to use a delta-sigma converter and convert the DSD directly. however, it appears that even the DSD1702 performs some level of amplitude quantization in its DAC.
so the 222ES and 555ES are fundamentally different after all. i'm curious what the difference in sound quality is... someone on the Hi-Rez asylum has owned both so hopefully they will respond to my question.
i'm going to take a look at some AD and Crystal chips as well for comparison.
i did not realize the PCM1738 was a 24-bit ladder DAC. i would have thought to take full advantage of SACD you'd want to use a delta-sigma converter and convert the DSD directly. however, it appears that even the DSD1702 performs some level of amplitude quantization in its DAC.
so the 222ES and 555ES are fundamentally different after all. i'm curious what the difference in sound quality is... someone on the Hi-Rez asylum has owned both so hopefully they will respond to my question.
i'm going to take a look at some AD and Crystal chips as well for comparison.
I was under the initial assumption that it was delta sigma, but reading more into it, I found this:
Read page 29, it gives a better diagram and description, which seems to state that it is delta sigma. It is at least part delta-sigma.
Read this thread on Hi Rez:
http://www.audioasylum.com/forums/hirez/messages/69617.html
The conclusion is that is mostly delta-sigma, but could be called a "hybrid", taking the best of both worlds.
"...and by the way if you read the data sheets carefully the BB1738 is not a delta sigma DAC ... it is a "hybrid" converting the most significant bits using a ladder and the lower bits using delta sigma - in an attempt to get the "best of both worlds"."
--
Brian
Read page 29, it gives a better diagram and description, which seems to state that it is delta sigma. It is at least part delta-sigma.
Read this thread on Hi Rez:
http://www.audioasylum.com/forums/hirez/messages/69617.html
The conclusion is that is mostly delta-sigma, but could be called a "hybrid", taking the best of both worlds.
"...and by the way if you read the data sheets carefully the BB1738 is not a delta sigma DAC ... it is a "hybrid" converting the most significant bits using a ladder and the lower bits using delta sigma - in an attempt to get the "best of both worlds"."
--
Brian
222ES vs. 555ES
FYI someone just posted this on the Asylum:
http://www.audioasylum.com/forums/hirez/messages/82779.html
FYI someone just posted this on the Asylum:
http://www.audioasylum.com/forums/hirez/messages/82779.html
Dorkus, I may have misinterpeted it.
I gave the PCM1738 a quick look over as I was comparing it to the PCM1730. It seems that both are similar with the '1738 having more input and digital filter slope options than the '1730.
Trouble was (for me) that the greater options available to the '1738 are only accessable via SCI. Jeez, I have a hard enough time trying to figure out converting I2S to MSB first, right justified, 24 bit data. I don't have a clue on how to manage a SCI. So I nixed that DAC as a standalone project for me.
On that first look awhile back, it appeared that the external PCM digital filter ports also accepted DSD from a "DSD Decoder". I assumed that it was "decoded" to PCM prior to input to the '1738.
On a second closer look, the '1738 does appear to accept straight DSD data. According to the datasheet (page14): "This configuration allows for direct interface to a DSD decoder for SACD applications".
I gave the PCM1738 a quick look over as I was comparing it to the PCM1730. It seems that both are similar with the '1738 having more input and digital filter slope options than the '1730.
Trouble was (for me) that the greater options available to the '1738 are only accessable via SCI. Jeez, I have a hard enough time trying to figure out converting I2S to MSB first, right justified, 24 bit data. I don't have a clue on how to manage a SCI. So I nixed that DAC as a standalone project for me.
On that first look awhile back, it appeared that the external PCM digital filter ports also accepted DSD from a "DSD Decoder". I assumed that it was "decoded" to PCM prior to input to the '1738.
On a second closer look, the '1738 does appear to accept straight DSD data. According to the datasheet (page14): "This configuration allows for direct interface to a DSD decoder for SACD applications".
back to the NS500V...
yuck! so, i was looking at the output circuit board again, and i figured out what those transistors in series with the +/-11V analog rails were for. if you look a few posts back to my circuit analysis i mentioned that there was basically a diode drop across them but i wasn't sure if it was a regulator of some sort or not.
turns out they're just simple FET or BJT switches. the reason they're there is to ensure that the +/-11V supply to all the output circuitry (output opamps and video buffer) does not ramp up until well after the unit has fully powered up and the muting circuit has had a chance to kick in. there is a fairly large capacitor (330uF 6.3V) tied to the gate/base of the switch devices which along with some resistors sets a time constant for the circuit on the order of a couple seconds.
this delayed power-up is only relevant for initial powering of the unit when used with the muting circuits. since i've killed the muting circuits anyway, it is completely useless and is only increasing the impedance of the supply, i imagine significantly so. out they go! i'm going to remove all the switch transistors and replace with straight-wire bypasses, also upgrade some decoupling caps with a few 120uF 25V Panasonic HFU's i have lying around... then we'll see how things sound. i'm not sure how audible this change will be but i expect at least slightly improved dynamics and bass.
yuck! so, i was looking at the output circuit board again, and i figured out what those transistors in series with the +/-11V analog rails were for. if you look a few posts back to my circuit analysis i mentioned that there was basically a diode drop across them but i wasn't sure if it was a regulator of some sort or not.
turns out they're just simple FET or BJT switches. the reason they're there is to ensure that the +/-11V supply to all the output circuitry (output opamps and video buffer) does not ramp up until well after the unit has fully powered up and the muting circuit has had a chance to kick in. there is a fairly large capacitor (330uF 6.3V) tied to the gate/base of the switch devices which along with some resistors sets a time constant for the circuit on the order of a couple seconds.
this delayed power-up is only relevant for initial powering of the unit when used with the muting circuits. since i've killed the muting circuits anyway, it is completely useless and is only increasing the impedance of the supply, i imagine significantly so. out they go! i'm going to remove all the switch transistors and replace with straight-wire bypasses, also upgrade some decoupling caps with a few 120uF 25V Panasonic HFU's i have lying around... then we'll see how things sound. i'm not sure how audible this change will be but i expect at least slightly improved dynamics and bass.
DC offset warning!
i just re-measured the dc offset at the 6-channel outputs of the player, and it is not zero on every channel... front L and R are in the tens of millivolts but the other channels have more, around a few millivolts. more importantly though the offset is DANGEROUSLY high when the unit is in soft power-off mode (i.e. red power light) - well over a volt! so, if you are going to run it into a direct-coupled amp/preamp, be sure to keep the volume completely down when the unit is turned off!!! or to be really safe, it looks like we will need to leave the blocking caps in after all... even if i build my always-on power supply, which takes care of the power-off offset and turn-on thumps, there's still no guarantee of reasonable offset, particularly with the differential amp topology employed.
interestingly, i am using a OPA2134 on front channels, where offset is near zero; OPA2604 on the rears, where offset is a couple millivolts; and the stock opamp for center/sub, where the offset is highest. does the OPA2134 have better DC performance than the OPA2604?
i just re-measured the dc offset at the 6-channel outputs of the player, and it is not zero on every channel... front L and R are in the tens of millivolts but the other channels have more, around a few millivolts. more importantly though the offset is DANGEROUSLY high when the unit is in soft power-off mode (i.e. red power light) - well over a volt! so, if you are going to run it into a direct-coupled amp/preamp, be sure to keep the volume completely down when the unit is turned off!!! or to be really safe, it looks like we will need to leave the blocking caps in after all...
even if i build my always-on power supply, which takes care of the power-off offset and turn-on thumps, there's still no guarantee of reasonable offset, particularly with the differential amp topology employed. interestingly, i am using a OPA2134 on front channels, where offset is near zero; OPA2604 on the rears, where offset is a couple millivolts; and the stock opamp for center/sub, where the offset is highest. does the OPA2134 have better DC performance than the OPA2604?
hmm... you may not want to decrease the impedance between the noisy power supply and your analogue circuits, mainly because, if nothing else, the extra impedance will provide some measure of noise isolation. Of course, you can always ensure a low impedance supply for your analogue circuitry with a large amount of decoupling capacitance, and if you do remove these transistors just to get rid of the voltage drop, you may want to consider replacing them with either ferrite beads and/or inductors. Also, are these transistors providing power-up muting for the video circuitry as well? If so, that may be a concern...
I've been a little sidetracked the past few days, but I'll be resuming my work on the ns500 here fairly soon... I've got some parts on the way (OPA2134 and AD8620s), so I'll give those a try and report back on the results. One thing I do like about the stock analogue board is the extensive use of guard traces. This is a very nice feature, and I'm glad to see that the Sony engineers did a good job in that respect.
As far as power supply goes, I'm not sure if it will really be necessary to have an external chassis for the replacement. I think it's possible to fit one in where the existing board goes. It's not a tremendous amount of space, but then again, the player doesn't consume much power, so a large transformer shouldn't be needed... a small toroidal unit like one of the Telema series carried by Digi-Key should do nicely.
Anyhow, let us know how the transistor removal goes.
PS - I haven't had to deal with the SMD parts glue yet (I really hate that stuff), but if you're willing to sacrifice the component, the glue is generally degraded by soldering temperatures, and should soften or at least weaken if you can heat it up for a period. This should occur before you start to burn the board itself, and will make your life easier in terms of preventing traces from lifting etc. I'm going to search for some more information online about removing that glue safely, so if I find anything out, I'll be sure to post it here.
Oh yeah, and one more thing: here's a useful tool that might help with this SMD work: http://www.usbmicro.com/apps.html If Rat Shack is open tomorrow, I might try building one of these for myself.
I've been a little sidetracked the past few days, but I'll be resuming my work on the ns500 here fairly soon... I've got some parts on the way (OPA2134 and AD8620s), so I'll give those a try and report back on the results. One thing I do like about the stock analogue board is the extensive use of guard traces. This is a very nice feature, and I'm glad to see that the Sony engineers did a good job in that respect.
As far as power supply goes, I'm not sure if it will really be necessary to have an external chassis for the replacement. I think it's possible to fit one in where the existing board goes. It's not a tremendous amount of space, but then again, the player doesn't consume much power, so a large transformer shouldn't be needed... a small toroidal unit like one of the Telema series carried by Digi-Key should do nicely.
Anyhow, let us know how the transistor removal goes.
PS - I haven't had to deal with the SMD parts glue yet (I really hate that stuff), but if you're willing to sacrifice the component, the glue is generally degraded by soldering temperatures, and should soften or at least weaken if you can heat it up for a period. This should occur before you start to burn the board itself, and will make your life easier in terms of preventing traces from lifting etc. I'm going to search for some more information online about removing that glue safely, so if I find anything out, I'll be sure to post it here.
Oh yeah, and one more thing: here's a useful tool that might help with this SMD work: http://www.usbmicro.com/apps.html If Rat Shack is open tomorrow, I might try building one of these for myself.
too late... already took the transistors out. =p
sounds ok. actually not as different as i thought, and i think the difference in the power supply is sort of swamped by the other change i made... i'm bypassing all the extraneous stuff after the opamp circuit (470 ohm resistor, RC filter, traces from the muting circuit) with a hard-wired 100 ohm Holco resistor. the resistor needs to break in but it sounds nice, a bit cleaner and rounder. the difference is not as large as i thought it would be though.
i'll probalby need an external case for the new power supply still because i want to use two transformers (digital/analog). the external case can be very small though as it'll only have the transformers, rectifiers, and main filter bank... regulation and so forth will go inside still. btw, i was going to use the talemna mini toroidals, but then i read in another thread that for digital circuits split-bobbin are preferrable due to the higher isolation. so i will probably try one of the other xformers in the digikey catalog.
sounds ok. actually not as different as i thought, and i think the difference in the power supply is sort of swamped by the other change i made... i'm bypassing all the extraneous stuff after the opamp circuit (470 ohm resistor, RC filter, traces from the muting circuit) with a hard-wired 100 ohm Holco resistor. the resistor needs to break in but it sounds nice, a bit cleaner and rounder. the difference is not as large as i thought it would be though.
i'll probalby need an external case for the new power supply still because i want to use two transformers (digital/analog). the external case can be very small though as it'll only have the transformers, rectifiers, and main filter bank... regulation and so forth will go inside still. btw, i was going to use the talemna mini toroidals, but then i read in another thread that for digital circuits split-bobbin are preferrable due to the higher isolation. so i will probably try one of the other xformers in the digikey catalog.
I bought one of these units last November and ripped it apart shortly thereafter. How much trouble did you find soldering in the SMT OPA2134? I have a few of these parts laying around from previous all-SMT headphone amp builds, and was figuring I could try them out.
I replaced the crappy 47uF output caps with Cerafines about a week after I bought the player. I don't recall the muting/whatever transistors that you've been talking about. Maybe I'll have to get out the old magnifying glasses...
-b
I replaced the crappy 47uF output caps with Cerafines about a week after I bought the player. I don't recall the muting/whatever transistors that you've been talking about. Maybe I'll have to get out the old magnifying glasses...
-b
hi b,
the hardest part about soldering SMT is getting rid of the old part, esp. cuz of the glue sony uses under the chip. instead of trying to desolder the opamp properly i highly recommend snipping all the leads off first, then trying to pry it off with a small flathead screwdriver, separate from the leads... you may still want to heat up the device beforehand though to weaken the glue as hifizen suggested... you want to almost burn off the glue, but be careful not to ruin the board with too much heat.
it is annoying, but not impossible. i actually broke quite a few traces in the process of removing the chips (they are super delicate, you can basically touch them w/the iron and they'll break off), but with some fine wire (i used some individual wires from a stranded cable, 32 gauge or so) and a steady hand you can patch things together quite nicely. when you put the new device onto the pads (assumming they're still there!), wet the pads with a little solder beforehand, then position the device with pliers and just get one of the pads to stick by applying your iron.
things i used: fine desoldering wick, needlenose pliers, one of those small flush-cut wire clippers, and a continuity meter to figure out where to patch when i broke the traces. my iron is 35W w/pencil tip but a varitemp iron is highly preferrable (i need to get one). do NOT use a solder sucker, as it can suck the traces right off the board... i would know.
the hardest part about soldering SMT is getting rid of the old part, esp. cuz of the glue sony uses under the chip. instead of trying to desolder the opamp properly i highly recommend snipping all the leads off first, then trying to pry it off with a small flathead screwdriver, separate from the leads... you may still want to heat up the device beforehand though to weaken the glue as hifizen suggested... you want to almost burn off the glue, but be careful not to ruin the board with too much heat.
it is annoying, but not impossible. i actually broke quite a few traces in the process of removing the chips (they are super delicate, you can basically touch them w/the iron and they'll break off), but with some fine wire (i used some individual wires from a stranded cable, 32 gauge or so) and a steady hand you can patch things together quite nicely. when you put the new device onto the pads (assumming they're still there!), wet the pads with a little solder beforehand, then position the device with pliers and just get one of the pads to stick by applying your iron.
things i used: fine desoldering wick, needlenose pliers, one of those small flush-cut wire clippers, and a continuity meter to figure out where to patch when i broke the traces. my iron is 35W w/pencil tip but a varitemp iron is highly preferrable (i need to get one). do NOT use a solder sucker, as it can suck the traces right off the board... i would know.
rickpt said:
whoops!
oh well... too late now.
actually i am fairly certain they are just a simple power delay switch. they are clearly triggered by a RC circuit with a long time constant, which is necessary for the muting circuit to work in intial power condition. but maybe they do some capacitance multiplication function as well. can't be sure, but it seems logical here - the main power supply has very little filter capacitance. incidentally, the player still sounds ok with them out, but not a whole lot different and maybe a tad noisier... they may have been providing some noise isolation. no matter, i am going to be feeding the board with my own clean DC power supply soon anyway.
p.s. the transistors i removed were really tiny SMT devices. would they really have low output impedance, esp. w/o any sort of feedback?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.