Having said that, a few quick points as I am out of this now...
- When you bypass caps, it is important to keep distance / wires as short as possible. Locating elsewhere the remote bypass can be fine under some circumstances but seldom so, it needs consideration (ESR & Co, time constant). Apart from the 2 main PS caps, it is IMHO for this unit best in most cases to bypass at the pins directly (somainly under the board). It also enables quick fitting and removing if you don't have 2 identical amps to try to assess the sonic variations.
- The DC blocking caps around the OPA make little difference. In some cases, depending on what you did elsewhere, you can do without some of them and as no cap is best cap we sadly noticed even in these circumstances little difference. Not saying none, saying not priority mod. The latest units seem to have a gain of 1 only at the OPA so possibly all caps can be taken away in this area (??? All from memory and at your own risk to try).
-"Quote" : Previously i had completely removed the cap in that position and replaced it with Wee WME 2.2uF film cap. I saw some weird behaviour, the sound quality changed for very good initially then after some time the sound changed to mono like (like am listening to bluetooth speaker) with no seperation. And occasionaly right channel will stop working. I almost got a mini stroke. Changed back to NIchicon FW 100uf 35V. Everything back to normal ! Big Relief !!!
=> I bet so, LOL!
Well, one needs to understand what the caps are doing and what value can be used. I don't know what cap you mean and have really no time for it, but you were lucky if you replaced a 100uF with a 2.2uF without damage. Not saying it can't be done, again all depends what the cap stands for initialy, but they can be some issues re bandpass (RC filter), buffering, instability due to wrong value... behaviours where time constant plays a role and it works and then no more after a certain time. Having said that, what you describe could also be due a a bad connection / soldering point, so who knows... important is everything is back in place. Please try to refer to my comments in the past on this cap: IF (??) I did some back then or touched it and it was senstivie I for sure put some comments.
- Quote "The four caps on the output side of OPA i am contemplating to change but am confused between Nichicon ES, Nichicon FG or Elna Silmic II. But thinking of doing this once i have my OPA 1656 in the place since that would define a better route in which would i like to go depending on their sonic signature (also as per your report no significant benefits). No bypass on these four caps as of yet."
=> Wise so, this is really fine tuning, honing the sound. The OPA1656, wih its right PS feed / bypass caps, is likely to make such a difference to YOUR ears, but of course it all depends (taste and colours etc.)
- Quote "Prior to these I changed the coils to Coilcraft 10uH (since i had them with me & they made significant upgrade in my TPA 3116 amp when i did it). Its pretty neat & is very firm. I am not able to compare the improvements made since most of the mods were done in a go & it's hard to imagine how the amp sounded in stock. but change must be very positive.
=> We changed the coils to quality ones and it made a small improvement. The emphase is on small, IMHO not a priority and the initial coils were doing an OK job so lot of money and work for little return. ON our units of course, this might have changed, and well noting we need only a few Watts to drive our LS so your mileage wight vary with more current hungry LS of course!
- Quote" I am contemplating the idea to bypass those 4 small ceramics close to the chip itself with some FKS2 (or FKP2 if slim enough) since they fit inside the cavity of heatsink just fine so will be much beneficial than heavy bypassing on the power caps lead out. Also FKP2 or FKS2 have a much lower DF compared to MKT or MKP types being a foil cap and are much suited for pulse duty. So even a smaller value can outperform a larger value MKT or MKP."
=> Some is true of course but I would advise you look at the chip's sheet to make sure you don't overdo it re min requirements, just in case...
- You did all the complicated mods but not the easiest and most rewarding one IMHO, so my bet is your amp is more than ready to welcome the OPA1656 and you should indeed notice a difference on your well tweaked unit!
Have fun and thanks for sharing all your hard work
Claude
- When you bypass caps, it is important to keep distance / wires as short as possible. Locating elsewhere the remote bypass can be fine under some circumstances but seldom so, it needs consideration (ESR & Co, time constant). Apart from the 2 main PS caps, it is IMHO for this unit best in most cases to bypass at the pins directly (somainly under the board). It also enables quick fitting and removing if you don't have 2 identical amps to try to assess the sonic variations.
- The DC blocking caps around the OPA make little difference. In some cases, depending on what you did elsewhere, you can do without some of them and as no cap is best cap we sadly noticed even in these circumstances little difference. Not saying none, saying not priority mod. The latest units seem to have a gain of 1 only at the OPA so possibly all caps can be taken away in this area (??? All from memory and at your own risk to try).
-"Quote" : Previously i had completely removed the cap in that position and replaced it with Wee WME 2.2uF film cap. I saw some weird behaviour, the sound quality changed for very good initially then after some time the sound changed to mono like (like am listening to bluetooth speaker) with no seperation. And occasionaly right channel will stop working. I almost got a mini stroke. Changed back to NIchicon FW 100uf 35V. Everything back to normal ! Big Relief !!!
=> I bet so, LOL!
Well, one needs to understand what the caps are doing and what value can be used. I don't know what cap you mean and have really no time for it, but you were lucky if you replaced a 100uF with a 2.2uF without damage. Not saying it can't be done, again all depends what the cap stands for initialy, but they can be some issues re bandpass (RC filter), buffering, instability due to wrong value... behaviours where time constant plays a role and it works and then no more after a certain time. Having said that, what you describe could also be due a a bad connection / soldering point, so who knows... important is everything is back in place. Please try to refer to my comments in the past on this cap: IF (??) I did some back then or touched it and it was senstivie I for sure put some comments.
- Quote "The four caps on the output side of OPA i am contemplating to change but am confused between Nichicon ES, Nichicon FG or Elna Silmic II. But thinking of doing this once i have my OPA 1656 in the place since that would define a better route in which would i like to go depending on their sonic signature (also as per your report no significant benefits). No bypass on these four caps as of yet."
=> Wise so, this is really fine tuning, honing the sound. The OPA1656, wih its right PS feed / bypass caps, is likely to make such a difference to YOUR ears, but of course it all depends (taste and colours etc.)
- Quote "Prior to these I changed the coils to Coilcraft 10uH (since i had them with me & they made significant upgrade in my TPA 3116 amp when i did it). Its pretty neat & is very firm. I am not able to compare the improvements made since most of the mods were done in a go & it's hard to imagine how the amp sounded in stock. but change must be very positive.
=> We changed the coils to quality ones and it made a small improvement. The emphase is on small, IMHO not a priority and the initial coils were doing an OK job so lot of money and work for little return. ON our units of course, this might have changed, and well noting we need only a few Watts to drive our LS so your mileage wight vary with more current hungry LS of course!
- Quote" I am contemplating the idea to bypass those 4 small ceramics close to the chip itself with some FKS2 (or FKP2 if slim enough) since they fit inside the cavity of heatsink just fine so will be much beneficial than heavy bypassing on the power caps lead out. Also FKP2 or FKS2 have a much lower DF compared to MKT or MKP types being a foil cap and are much suited for pulse duty. So even a smaller value can outperform a larger value MKT or MKP."
=> Some is true of course but I would advise you look at the chip's sheet to make sure you don't overdo it re min requirements, just in case...
- You did all the complicated mods but not the easiest and most rewarding one IMHO, so my bet is your amp is more than ready to welcome the OPA1656 and you should indeed notice a difference on your well tweaked unit!
Have fun and thanks for sharing all your hard work
Claude
I totally understand. Thought, i should add my findings here since it wasn't mentioned before in this thread.
This is exactly how i have done on my unit other than the PS caps which have an extension with decently thick wires.
The cap i was talking about was the middle one out of the 5 before the OPamp. I felt since bypass works great on this position (as per your finding), why not to replace it completely with a quality film cap. But that was a blunder. No bipolar for that position now as per my terrible experience. It was changed from 10uF electro polar to 2.2 uF bipolar to finally 100uF electro polar. Ans everything is fine now.
The value of ceramics used on the board is lesser than as on TPA 3255 datasheet. So, i guess i will be fine. I have done this successfully on my TPA 3116 boards.
Ha ! I wanted to make sure that i get genuine OPA 1656, so order is on mouser. The adapter & the pins are in mail as well.
It's been a great thread read here. Thanks for your contribution as well. Very thorough and detailed.
Last edited:
Thanks to you for reading us and posting your feedbacks, keeping this thread thus alive.
Yep, genuine OPA1656 are the key, you are right, read so many horror stories since with fakes...
Let us know your findings and of course you can tweak its PS bypass caps, as it appears this amp suffers quite on the OPA PS side... at least the 3 we had did.
Enjoy music
Claude
Yep, genuine OPA1656 are the key, you are right, read so many horror stories since with fakes...
Let us know your findings and of course you can tweak its PS bypass caps, as it appears this amp suffers quite on the OPA PS side... at least the 3 we had did.
Enjoy music
Claude
Sure, will be updating my experience with OPamp upgrade. Of course PS upgrade on adapter is a must.
Hello,
I had A04, a green pcb that came with fake NE5532 and 4 gold Nichicon 10uf no serial(not FG FW) before the heatsink also fake.
I changed for LME49720 and those 4 fakes for Nichicon UES and the foggy and bad stock sound became clear, transparent and neutral, I changed back for Elna silmic and it took several weeks for the clear and transparent sound to appear but softer and warmer than UES , with more delicate and open highs.
With that experience I now have A07, and I changed all the electrolytic capacitors before the heatsink also with Panasonic but FR.
I use opa1656 with 4 Nichicon UES and 5 Elnas before the opa.
Now it's sharper and more transparent, it still took weeks for the sound to open up.
For another A07 I would use 4 Elnas in that position and UES or Elna before the opa for a totally different sound
I had A04, a green pcb that came with fake NE5532 and 4 gold Nichicon 10uf no serial(not FG FW) before the heatsink also fake.
I changed for LME49720 and those 4 fakes for Nichicon UES and the foggy and bad stock sound became clear, transparent and neutral, I changed back for Elna silmic and it took several weeks for the clear and transparent sound to appear but softer and warmer than UES , with more delicate and open highs.
With that experience I now have A07, and I changed all the electrolytic capacitors before the heatsink also with Panasonic but FR.
I use opa1656 with 4 Nichicon UES and 5 Elnas before the opa.
Now it's sharper and more transparent, it still took weeks for the sound to open up.
For another A07 I would use 4 Elnas in that position and UES or Elna before the opa for a totally different sound
Attachments
Nice ! I have my order of Elna Silmic II 4.7 uF 50V on mouser as well for those 4 caps in the coupling position. Elna Silmic II's have usually upto 150 hr of burn in time as reported by many who have used them in their gears. But it's nice to hear that your Elna's are now sounding clear & transparent. I havn't tried Elna's before but so thought of giving it a try. I have tried Nichicon ES in my other amplifier build but it didn't had that hi-fi highs. Even bypassing it with wima didn't helped much. My 0.5$ MKP cap was much better than that so i replaced them with MKP ones. ES ones have great vocals but highs are pretty rolled off. On the other hand Nichicon KZ has airy & very nice high end. It doesn't feel that I am loosing any detail with them. I have used myself those in other builds and in this as well on the power decoupling caps position. I guess those will be a perfect combo for those 5 elnas before the OPA. Best of both worlds.
For the caps before heatsink on power side , i guess Panasonic FR are excellent choice. I have used FM myself since i had those beforehand but FR should be little bit better. For this position FM performed better than Nichicon FW, FG, KZ, Kemet A750 polymer caps and many others.
Since FR upgarde is already in place, i don't think you will gain as much as reported in this thread after OPA PS bypass. But still a very simple mod & worth trying. Even a simple Wima 0.1 uF MKP with a small 4.7-10 uF electrolytic will play nicely.
While looking closely i noticed that couple of your ES are mounted in opposite direction as opposed to other two. Even though they are bipolar they have different lead length to denote the outer foil. I don't know how much it will make a difference to the sound but ideally they should be mounted in same orientation.
I might be completely wrong but i have found that even though Panasonic FM & FR in decoupling position (or bypass at power pins of OPA) provided an instant upgrade in terms of richness of sound the sound was rather flat & 2d like as opposed to Nichicon KZ (more 3d stage). I might swap my Panasonic FM out of the chain and rather replace them with KZ or PW in future if i felt like this for so long.
Attachments
Hello,
That's why I changed the UES to Elna silmic on the A04, the treble never opened up. With Elna it was the opposite, although it took time.
Outside of that, I think UES sounds great, and I took a chance on using it on the A07, but in conjunction with Elna, and it worked great.
I think it's because of the influence of Elna or the opa1656 combined, or maybe A07 has better treble than A04.
I use the 10uf 25v UES and put the shorter leg as negative. Elna's value is 10uf 35v, I wanted to try KZ but they are 8mm in 10 and 22uf.
That's why I changed the UES to Elna silmic on the A04, the treble never opened up. With Elna it was the opposite, although it took time.
Outside of that, I think UES sounds great, and I took a chance on using it on the A07, but in conjunction with Elna, and it worked great.
I think it's because of the influence of Elna or the opa1656 combined, or maybe A07 has better treble than A04.
I use the 10uf 25v UES and put the shorter leg as negative. Elna's value is 10uf 35v, I wanted to try KZ but they are 8mm in 10 and 22uf.
I had similar experience with ES before so not much keen on using them. Elna for this position is in order. If it dsnt synergizes well with my rest of the system i will see for FG, FW or KZ (if i can make them fit). As of now the 5 caps before OPamp on my unit are Nichicon FW 10 uF 63V except the middle one which i have changed to 100 uF 35V. I tried changing it to 2.2 uF film cap but it behave weirdly in that position. Strictly no bipolars in this position ! (ref post #717).
For the 4 input coupling caps, i have changed the value to 4.7 uF since i noticed no loss of bass with 1 uF MKT caps which were connected via extended wire. Lowering on capacitance gives chance to bump up the voltage value which is now 50V. This means lower distortion figure compared to low voltage rated cap (i dont know if we can hear that or not). Also a lower value capacitor is supposedly faster than a larger value one (very precievable once i used KZ 330 uF 50V in this position).
I am waiting for my OPA 1656 to arrive. Once i have that in place. The sound signature will better dictate me in which direction i will like to go with those 9 caps surrounding the OPA.
For the 4 input coupling caps, i have changed the value to 4.7 uF since i noticed no loss of bass with 1 uF MKT caps which were connected via extended wire. Lowering on capacitance gives chance to bump up the voltage value which is now 50V. This means lower distortion figure compared to low voltage rated cap (i dont know if we can hear that or not). Also a lower value capacitor is supposedly faster than a larger value one (very precievable once i used KZ 330 uF 50V in this position).
I am waiting for my OPA 1656 to arrive. Once i have that in place. The sound signature will better dictate me in which direction i will like to go with those 9 caps surrounding the OPA.
I had a similar experience with the Aiyima A04 where the right channel started to sound weird/lower than the left channel.
I previously added bypasses (0.1uF MLCC C0G) to all the coupling capacitors (the 5 before Opamps and the 4 after the Opamps).
In the end I decided to remove the bypass from the center capacitor of the 5 before the Opamps and now everything sounds nice again.
Then I decided to trace all the connections(Opamp pins and capacitors) and I realized how they work:
The Positive RCA Signal (L/F) passthrough a 1K resistors before getting into the POT input. The Pot outputs go into Capacitors 2 and 5 (positive lead) and then go into Opamp Pin 2 (1IN-) on each Opamp.
The negative(ground) RCA signal go into Capacitor 3 (positive Lead) and from there it goes into both Opamps Pins 3 and 5 (1IN+ and 2IN+). The 1OUT(pin 1) from each Opamp go into Pin6(2IN-) via a 100ohm resistor, and also go to Post-Opamp(amplifier coupling) Capacitor 1 and 3(from the second Opamp).
The 2OUT pin 7 from each opamp go into Post-Opamp Capacitors 2 and 4.
Basically what the Opamps do is to take the Input signal and invert it and then get that inverted signal into the second section of the Opamp to produce the "positive" signal.
1Out=(1IN+) - (1IN-) = 0-Input Signal = Input Signal Inverted which goes into 2IN-
Then:
2Out= (2IN+) - (2IN-) = 0 - (-Input Signal) = Input Signal.
Both Outputs go into the Chipamp coupling capacitors for the Positive and Negative Differential inputs into the Amplifier.
I don't know if there is some kind of extra gain on the Opamp section, but even if there is then it would be done on the first section (where the first inversion is done). Other way the signal amplitude for both Output 1 and 2 won't be equal.
I previously added bypasses (0.1uF MLCC C0G) to all the coupling capacitors (the 5 before Opamps and the 4 after the Opamps).
In the end I decided to remove the bypass from the center capacitor of the 5 before the Opamps and now everything sounds nice again.
Then I decided to trace all the connections(Opamp pins and capacitors) and I realized how they work:
The Positive RCA Signal (L/F) passthrough a 1K resistors before getting into the POT input. The Pot outputs go into Capacitors 2 and 5 (positive lead) and then go into Opamp Pin 2 (1IN-) on each Opamp.
The negative(ground) RCA signal go into Capacitor 3 (positive Lead) and from there it goes into both Opamps Pins 3 and 5 (1IN+ and 2IN+). The 1OUT(pin 1) from each Opamp go into Pin6(2IN-) via a 100ohm resistor, and also go to Post-Opamp(amplifier coupling) Capacitor 1 and 3(from the second Opamp).
The 2OUT pin 7 from each opamp go into Post-Opamp Capacitors 2 and 4.
Basically what the Opamps do is to take the Input signal and invert it and then get that inverted signal into the second section of the Opamp to produce the "positive" signal.
1Out=(1IN+) - (1IN-) = 0-Input Signal = Input Signal Inverted which goes into 2IN-
Then:
2Out= (2IN+) - (2IN-) = 0 - (-Input Signal) = Input Signal.
Both Outputs go into the Chipamp coupling capacitors for the Positive and Negative Differential inputs into the Amplifier.
I don't know if there is some kind of extra gain on the Opamp section, but even if there is then it would be done on the first section (where the first inversion is done). Other way the signal amplitude for both Output 1 and 2 won't be equal.
Attachments
Thank you for your comments on the Center Capacitor(Capacitor 3 of the 5 before the opamp). It made me realized the solution for the issue I was having after I installed the bypasses.
Now everything sounds perfect, and the audio from the right channel does not drop anymore.
Back in the days I posted about the signal path and what the op-amps were doing etc.
Search is your friend for more details.
In short, there is on our (eg older) amps a small gain stage, x2 if I remember - it has been ages.
Due to that and the way the PS works in this unit, a constant DC is sent between the stages, hence the mandatory cap to supress it and keep only AC (eg music) signals between Op amps, which is wayyyy preferable. That 5th cap is mandatory...
It seems later units are unity gain and could hence possibly do without that cap, but it appears it is still there. Later units have less gain, half of it, people complained about dynamic loss but my bet is they rather suffered from being upset by the pot's position. Or whatever, who knows...
IF interested, for more details about how that unit works and where to find a schematic or close to it (always better than words), please refer to the early posts: they are there to avoid typing again or reinventing the wheel.
As of me, I am out since years of this thread but couldn't resist peeping in... I am out again.
Have fun guys with this little unit
Claude
Search is your friend for more details.
In short, there is on our (eg older) amps a small gain stage, x2 if I remember - it has been ages.
Due to that and the way the PS works in this unit, a constant DC is sent between the stages, hence the mandatory cap to supress it and keep only AC (eg music) signals between Op amps, which is wayyyy preferable. That 5th cap is mandatory...
It seems later units are unity gain and could hence possibly do without that cap, but it appears it is still there. Later units have less gain, half of it, people complained about dynamic loss but my bet is they rather suffered from being upset by the pot's position. Or whatever, who knows...
IF interested, for more details about how that unit works and where to find a schematic or close to it (always better than words), please refer to the early posts: they are there to avoid typing again or reinventing the wheel.
As of me, I am out since years of this thread but couldn't resist peeping in... I am out again.
Have fun guys with this little unit
Claude
Glad it helped ! My main purpose of posting my findings was due to the reason that it was completely opposite of what was suggested in the thread (bypassing the middle one gave most benefits). Since then i removed all the bypass and re-did one by one making sure i do it based on my liking rather than just copying some tried formula. I am on A07 but i don't see why Aiyima will deviate from these fundamental design choices they used on A04. Also as per your findings in the previous post, you said the Cap 2 & 5 before the OPamp are the input coupling caps while the thread mention these as Cap 1 & 4 !! I guess some of the earlier statements in the thread were made prematurely or maybe there were very different batch samples which is very unlikely. I will reopen my unit tommorrow again to check on this. Thanks for pointing out.
Well, caps 2 and 5, and 1 and 4 could be the same caps depending on where you start counting:
1 2 3 4 5
5 4 3 2 1
Ahh I see
I followed the numbering as per the attachment.
So are the input coupling caps 2&5 or 1&4 ?
Attachments
LOL. I counted from Left to Right. So, according to your image they are 4 and 1Ahh I see
I followed the numbering as per the attachment.
So are the input coupling caps 2&5 or 1&4 ?
🤦♂️
Thanks for updating. My OPA 1656 is due to be delivered in couple of days. I hope that upgrade plus its power supply will be a significant step up.
I am currently pairing it with a cheap tube-pre with Voskhod 6J1P-EV, the sound is truly amazing.
Yes. I'm confident that the upgrade from a fake NE5532 to a OPA1656 will be noticeable.
I also have a pair of FX Audio Tube 03 and Tube 01(both heavily modded as well and Tube rolled). But I'll need to constate if they add something positive to current setup:
SMSL SU-6 => SMSL SH-6 => amplifier(I'm retaking the Aiyima A04, but I use multiple budget amplifier from all classes A, AB and D from time to time). I realized that constantly changing among my Amplifiers or Speakers makes me appreciate them more.
I am contemplating to purchase Aiyima Tube A3 ( Fx Audio Tube 03 design clones ) since they are only available via amazon in my country. I have read good things about them (Tube 03) when used without opamps alongwith cap & tube rolling.
Mine Fiio K3 DAC is ESS 9038Q2M based as well, and they get a bit etchy with my very resolving diy TPA3116 based amp mostly with Nichicon KZ onboard. I find tubes are certainly helping me in achieving the synergy i am looking for. Switching between amps is lot of fun indeed. I find myself juggling with my DIY dual TPA 3116 amp, Allo Volt + D & newly accquired A07.
Mine Fiio K3 DAC is ESS 9038Q2M based as well, and they get a bit etchy with my very resolving diy TPA3116 based amp mostly with Nichicon KZ onboard. I find tubes are certainly helping me in achieving the synergy i am looking for. Switching between amps is lot of fun indeed. I find myself juggling with my DIY dual TPA 3116 amp, Allo Volt + D & newly accquired A07.