The only mod I did except replacing the 5532 was adding a shield to the IC:s and grounding it. I have no idea if it did any audible difference but I could clearly see on the scope that the output noise level (with shorted inputs) was lower than without the shield.
Maybe there can be a better ground point than what I choose but it clearly makes a difference on measurements.
Maybe there can be a better ground point than what I choose but it clearly makes a difference on measurements.
Welcome on board Radelius, glad you are happy 
TBH, as posted I suspected the culprit to be the PS given your comments. If you read the spec sheet it is clearly mentioned that this OPA should be bypassed with 0.1uF... in some cases even 0.01uF depending on application. As you perhaps know already some some OPA are very sensitive to a careful layout, some requiring a proper separate grounding, and some a bit less (often the ones that perform average LOL). The OPA1656 is crying for good quality current. You have been there, you know all that... and you didn't give up and shared : thanks.
Of course Aiyima should know all this aswell, but we don't see any 10-30uF lytic around the OPA. No disaster, I do without on other units, fitting SMD C0G caps of 22uF... but did Aiyima went the trouble in such a non expensive unit?
Bottom line, adding a 0.1uF C0G (in our case) - or MKS or even better MKP as you did - directly at the legs makes a great improvement, outlining something was not perfectly executed originaly. Further, adding a 10uF lytic + 10uF X7R did also help a lot... shame we don't have the exact schematic of the this amp, but at the end we found what is missing... and have still a 1uF WIMA MKS to try on top, but TBH I believe we have corrected the main flows re OPA PS.
Would you mind sharing the exact reference of the Wima MKP you used (is it MKP10 etc.?) and perhaps a pix of your installation of the cap on the board?
Thanks for your feedback and well done keeping digging
Claude
TBH, as posted I suspected the culprit to be the PS given your comments. If you read the spec sheet it is clearly mentioned that this OPA should be bypassed with 0.1uF... in some cases even 0.01uF depending on application. As you perhaps know already some some OPA are very sensitive to a careful layout, some requiring a proper separate grounding, and some a bit less (often the ones that perform average LOL). The OPA1656 is crying for good quality current. You have been there, you know all that... and you didn't give up and shared : thanks.
Of course Aiyima should know all this aswell, but we don't see any 10-30uF lytic around the OPA. No disaster, I do without on other units, fitting SMD C0G caps of 22uF... but did Aiyima went the trouble in such a non expensive unit?
Bottom line, adding a 0.1uF C0G (in our case) - or MKS or even better MKP as you did - directly at the legs makes a great improvement, outlining something was not perfectly executed originaly. Further, adding a 10uF lytic + 10uF X7R did also help a lot... shame we don't have the exact schematic of the this amp, but at the end we found what is missing... and have still a 1uF WIMA MKS to try on top, but TBH I believe we have corrected the main flows re OPA PS.
Would you mind sharing the exact reference of the Wima MKP you used (is it MKP10 etc.?) and perhaps a pix of your installation of the cap on the board?
Thanks for your feedback and well done keeping digging
Claude
Hi Strongbow60,
Interesting post of yours. I did experiment with shielding OPAs in the mid 90s, was back then "the tweak" to try amongst insiders. Did awfull things with sheets of copper, being probably lucky at the end to not short anyting, LOL! Your work on the contrary looks absolutely great: congrats.
Yes, people who measured back then reported less noise pick up and the inside of my Pioneer Elite (PD-75 of 1992) of has many of those shields per factory. Now, to very honest, I have to admit I never heard any difference when fitting them... nor when retiring them inside the Pioneer to go for better OPAs. So I forgot about all this, thanks for refreshing memories
It would be nice to read if someone notices audible sonic improvements going that route: it is always great to have additional paths, eventhough they turn out to be small improvements.
Claude
Interesting post of yours. I did experiment with shielding OPAs in the mid 90s, was back then "the tweak" to try amongst insiders. Did awfull things with sheets of copper, being probably lucky at the end to not short anyting, LOL! Your work on the contrary looks absolutely great: congrats.
Yes, people who measured back then reported less noise pick up and the inside of my Pioneer Elite (PD-75 of 1992) of has many of those shields per factory. Now, to very honest, I have to admit I never heard any difference when fitting them... nor when retiring them inside the Pioneer to go for better OPAs. So I forgot about all this, thanks for refreshing memories
It would be nice to read if someone notices audible sonic improvements going that route: it is always great to have additional paths, eventhough they turn out to be small improvements.
Claude
Carrying on with the 0.1uF bypasses : next on the list we have our two 3000uF PS caps. Remember these are already bypassed with four 2.2uF X7R caps.
Funny test, Gilles did fit them temporaly (we are short on caps to fit both amps with all the coming mods), closed the lid… and was interrupted by a friend. He had to perform his listening tests several hours later and completely forgot which unit was tweaked and which wasn’t. He took notice of everything he could hear while performing a kind of poor AB blind test (OK, no ABX and who knows what his subconscient might have remembered, but all that is neither our goal nor our interest TBH ) and at the end he had a small but very recurrent winner. It is nice to have 2 identical units for all these tweaks. But big sweat: which unit was having the edge ?
Opening it, it turns out the unit with the extra bypasses won. On very demanding and well mixed music, all kinds of including latest Yello, the sound was a little more precise, more detailled, and the bass line defo more textured. Enjoyable, but a very small gain. Perhaps up to +1. On « normal / average » records though, no real difference was audible to Gilles. Note when it was audible, it did make remarkably more difference than the 3 bypass caps together we played with yesterday.
Bottom line we give this tweak say 0.8 : this is really very fine tuning, but it adds up with many similar steps and it cost very little anyway - but you shouldn’t start here. Even more so because we are yet not completely done with the big PS caps: we have planned to swap them with presumably better ones (and also possibly in a second step we might increase capacity but that last one is not very promising and only because we have the parts lying around).
The bypass test caps we fitted will now be removed, so both amps are identical again, and we have still 3 other places to put them, so stay tuned.
Claude… and Gilles
Funny test, Gilles did fit them temporaly (we are short on caps to fit both amps with all the coming mods), closed the lid… and was interrupted by a friend. He had to perform his listening tests several hours later and completely forgot which unit was tweaked and which wasn’t. He took notice of everything he could hear while performing a kind of poor AB blind test (OK, no ABX and who knows what his subconscient might have remembered, but all that is neither our goal nor our interest TBH ) and at the end he had a small but very recurrent winner. It is nice to have 2 identical units for all these tweaks. But big sweat: which unit was having the edge ?
Opening it, it turns out the unit with the extra bypasses won. On very demanding and well mixed music, all kinds of including latest Yello, the sound was a little more precise, more detailled, and the bass line defo more textured. Enjoyable, but a very small gain. Perhaps up to +1. On « normal / average » records though, no real difference was audible to Gilles. Note when it was audible, it did make remarkably more difference than the 3 bypass caps together we played with yesterday.
Bottom line we give this tweak say 0.8 : this is really very fine tuning, but it adds up with many similar steps and it cost very little anyway - but you shouldn’t start here. Even more so because we are yet not completely done with the big PS caps: we have planned to swap them with presumably better ones (and also possibly in a second step we might increase capacity but that last one is not very promising and only because we have the parts lying around).
The bypass test caps we fitted will now be removed, so both amps are identical again, and we have still 3 other places to put them, so stay tuned.
Claude… and Gilles
This is what it looks like even for my 1556
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Well I'm playing around with OP's again and landing in the fact for the day that I still prefer 1642 with MKP2 decouppling caps. I was playing some classical vinyl's that I like and didn't enjoy them as much as I used to. switched the 1642 back in and the soundstage and the recording room came back.
There is something with the 1656 that my system and ears don't like.
Sorry guys but give 1642 a second chance
Have a nice weekend/S
There is something with the 1656 that my system and ears don't like.
Sorry guys but give 1642 a second chance
Have a nice weekend/S
Taste and coulours, nothing wrong with that as long as you are happy
Just out of curiousity...Vinyls... do you have the same findings with CD?
I found sometimes that I liked something better because it compensated for some other flows. For example I love my B1 Korg, a wonderfull... (admittely) soundmachine, in my main system. I prefer it slightly to SOA preamps, passive or active, but could understand people loving it the other way round, close call. BUT I would tend to say that the B1 Korg would probably play any SOA preamp (bar tube ones) to the wall when associated to Class D, because it complements so well what most Class D amps are missing. I don't have Class D at home, but if I would I would consider that kind of unit upfront somewhere in my system.
Of course, it is alchemy and at the end of the day you are compensating flows and flavours, to have a top result you like, but it won't give you back what you lost or compensated.
I still have in my 20y old DAC my AD825 OPA biased in Class A with a pair of genuine 2SK170 (LOL!). The OPA 1656 is better, I compared on another output stage, but I didn't bother trying it in my DAC as that ultimate transparency and a tad less euphonic effect wouldn't probably do that very old DAC any favour - it is simply more homogeneous with all its limits, no need to put slicks on a SUV. Having said that, hmm, I could try...nah, let's just build a new one from scratch if needed, LOL
Enjoy music the way you like it most, that's priceless
Claude
Just out of curiousity...Vinyls... do you have the same findings with CD?
I found sometimes that I liked something better because it compensated for some other flows. For example I love my B1 Korg, a wonderfull... (admittely) soundmachine, in my main system. I prefer it slightly to SOA preamps, passive or active, but could understand people loving it the other way round, close call. BUT I would tend to say that the B1 Korg would probably play any SOA preamp (bar tube ones) to the wall when associated to Class D, because it complements so well what most Class D amps are missing. I don't have Class D at home, but if I would I would consider that kind of unit upfront somewhere in my system.
Of course, it is alchemy and at the end of the day you are compensating flows and flavours, to have a top result you like, but it won't give you back what you lost or compensated.
I still have in my 20y old DAC my AD825 OPA biased in Class A with a pair of genuine 2SK170 (LOL!). The OPA 1656 is better, I compared on another output stage, but I didn't bother trying it in my DAC as that ultimate transparency and a tad less euphonic effect wouldn't probably do that very old DAC any favour - it is simply more homogeneous with all its limits, no need to put slicks on a SUV. Having said that, hmm, I could try...nah, let's just build a new one from scratch if needed, LOL
Enjoy music the way you like it most, that's priceless
Claude
How effective is that bypass cap?
Thinking about adding one to my inbound OPA1656 for my TPA3255
How effective is that bypass cap?
Thinking about adding one to my inbound OPA1656 for my TPA3255
Hi ! Have a look at page 11, post by Claude. Might give you an idea...
Gilles