Aiyima TPA3251 Modification Build Thread!

Member
Joined 2019
Paid Member
Touched the tracks? I swapped all 9 of the caps and have soldered ( a few different times) caps on the legs of the socket so I assume that counts as "touching".....lol

And yes, these 1656's are so loose you can basically "rattle" them around in the socket. Zero force needed to put them in, they just fall to the bottom of the socket. For sure they would fall out if I turned it over. There are all sorts of weird audio artifacts using them as is. Not even sure how to explain it, like RF noise maybe?

I can hear when one of them isn’t making full contact into the old, stretched socket...generally one channel is louder than the other when the sound should be equal..

Pete
 
Nah, yes, they had heatcycles, but as long as you didn't put enormous heat on the caps (and you didn't as otherwise the tracks would have been damaged there) it should be OK a few cm away at the OPA, so no worries :)
By touching I meant really directly in situ... but then as you said it is for experiment, so nothing to lose :)

OPA that rattle in the socket, ausio artifacts... stop! Go Turion, there isn't a proper contact! I was amazed how much abuse they could take on my rig, even trying to push them while playing, but that was for experiment only... I wouldn't advise it.

What about double socket as per Turion to start with, and depending on development you may or not decide to remove the existing socket and make it permanent then.

I hope at least the Brown Dogs fit tight in your new socket...
 
Quote: No soldering is involved unless I interpreted your sentence incorrectly

Indeed not. Bound was perhaps not a wise choice of mine.

In my case in such situations, I pushed hard with new sockets but can only do that a certain number of times, it is not meant to be. But it is quite solidely attached then... too solidely perhaps.

On older sockets assembly, that have taken some if the above abuse, they are just getting slightly loose and you may have the OPA rocking very slightly usualy in the length and not making perfect contact on all pins... although still making contact on all pins. It is then better to hold it in a given position but forcing won't help as too loose. I found that using a touch of Blu Tack between the sockets on say one side was adequate to avoid the small rocking and keep the asssembly in its intented position. Not great, but well...
 
Hopefully solved then, thanks to Pete's advice!

50 to 60 OPA swaps... no wonder the socket is dead LOL!

I guess you guys understand why I did a test rig to do a preliminary selection on my builds, as we are talking dozens of OPAs swapped in all directions... and tested as are plus with additional Class A bias (with a preliminary socket into socket :)
 
Nah, yes, they had heatcycles, but as long as you didn't put enormous heat on the caps (and you didn't as otherwise the tracks would have been damaged there) it should be OK a few cm away at the OPA, so no worries :)
By touching I meant really directly in situ... but then as you said it is for experiment, so nothing to lose :)

OPA that rattle in the socket, ausio artifacts... stop! Go Turion, there isn't a proper contact! I was amazed how much abuse they could take on my rig, even trying to push them while playing, but that was for experiment only... I wouldn't advise it.

What about double socket as per Turion to start with, and depending on development you may or not decide to remove the existing socket and make it permanent then.

I hope at least the Brown Dogs fit tight in your new socket...

This was always intended to be a test rig for me. Abuse this one figuring out what I like then build a new one fresh.
 
Member
Joined 2019
Paid Member
Quote: No soldering is involved unless I interpreted your sentence incorrectly

Indeed not. Bound was perhaps not a wise choice of mine.

In my case in such situations, I pushed hard with new sockets but can only do that a certain number of times, it is not meant to be. But it is quite solidely attached then... too solidely perhaps.

On older sockets assembly, that have taken some if the above abuse, they are just getting slightly loose and you may have the OPA rocking very slightly usualy in the length and not making perfect contact on all pins... although still making contact on all pins. It is then better to hold it in a given position but forcing won't help as too loose. I found that using a touch of Blu Tack between the sockets on say one side was adequate to avoid the small rocking and keep the asssembly in its intented position. Not great, but well...

These sockets were not really meant for the rolling abuse we have been giving them. ;)

BTW, traditional DIP8 op amps, e.g. Muses02, will fit into these newer style (round holes) dip8 adapters PCB mounts (with a little effort) as well but the old style dip8 receptacles (square looking holes) don’t work very well with the round pinned adapters.

Pete
 
Member
Joined 2019
Paid Member
Yeah, I'm not sure what type of solder they used on this but it's really hard to undo.

The 3eAudio boards use, what someone has described as, cement solder. The soldering iron temperature one needs to desolder components is close to the temperature one destroys the pads with. I just use chip-quik solder and melt it into the soldered item. It creates a slight mess but it stays liquid a lot longer (with lower heat) and the part comes out. Just a lot more solder to solder-wick away..;)

Pete
 
+1, that's what I do, and sometimes even that doesn't work great because both solders struggle to mix despite unbelievable temps

We came to the conclusion Aiyima used today's cheap solder plus a layer of glue to keep some components in place. On top, most components are probably soldered in say 2 "go", and some are part of additional processes, or by hand, or whatever but solder joints are defo not consistant (well taking already into account that some large ground tracks dissipate the heat and make iron heat transfer more difficult on some locations).

Pot board... hours on it, we succeded on one board with 4 pins out of 6 and with 5 out of 6 without damaging anything. And at the end never succeeded going further LOL. Some blue output caps came easy on both amps, but they were not the same across both amps and both amps had ther difficult legs LOL
 
Lester’s SOIC8 adapter (that he ships with his TPA3255 amps) uses really big pins. Anywhere I’ve used his SOIC8 adapter the socket has been stretched out and I’ve had to use other sockets as intermediary mounts so the browndog adapters fit in tightly.

Pete

Weird... tried the Jlester OPA1656 DIP8 with my Aiyima and Yuan Amp, nothing to complain with )
 
Member
Joined 2019
Paid Member
but solder joints are defo not consistant (well taking already into account that some large ground tracks dissipate the heat and make iron heat transfer more difficult on some locations....

Yeah, I’ve found this to be nerve-racking experience as well, especially with 3e’s amps. The combination of that with the solder type used makes for easily damaged solder pads.

Pete
 
Member
Joined 2019
Paid Member
Weird... tried the Jlester OPA1656 DIP8 with my Aiyima and Yuan Amp, nothing to complain with )

daniboun: the problem arises when you start using the browndog adapters (and their slightly thinner pins) AFTER using Lester’s DIP8 opamp adapter. Lester has even made the comment, somewhere, that he finds his DIP8 PCB-mounts to loosen up over time after so much opamp rolling. I just swap out the DIP8 mounts if I need to.
 
Colleagues, your op-amp swap games are very exciting. Note that the gain is 1 and the maximum feedback. This is a poor mode and oscillation at very high frequencies is possible. It is very difficult to detect generation; when the probes of the device are connected, the generation may disappear or, on the contrary, arise. A way out of this problem must be sought.
 
Member
Joined 2019
Paid Member
Colleagues, your op-amp swap games are very exciting. Note that the gain is 1 and the maximum feedback. This is a poor mode and oscillation at very high frequencies is possible. It is very difficult to detect generation; when the probes of the device are connected, the generation may disappear or, on the contrary, arise. A way out of this problem must be sought.

UAN, those of us w/o oscilloscopes live life on the edge with our sloppy habit of op-amp rolling and the risk of oscillations clouding our subjective evaluations of each one of the opamps (and possibly damaging our speakers). It’s definitely not like tube rolling. I just depend on the more technically learned folk weighing in and warning us of possible problems. Until then I just depend on what my ear hears and hope that ignoring proper electrical engineering concepts don’t bite us in the ****....;)
 
Member
Joined 2019
Paid Member
turion64,
Yes, the selection of operational amplifiers by ear also has a reason. I don't see any other way yet. Now I am fighting with the bad sound of the FX 502s pro, there must be chips whistling there too.

Have you read this forum? TPA3250 somebody is listening?

TPA3250 somebody is listening?

Seems like it might be helpful with your battle..

Cheers,

P