Cheap TPA3118D2 boards, modding them and everything that comes with it

Hello together ...

I also ordered two of theses sanwu TPA3118 clones. After first tests I really like their sound. But I somehow have a problem with the "mute-function" of the boards. Actually nothin happens, when I shorten the two pins labeled with mute. As I understand from the TI datasheet, this normally shoud immediately mute the output.

Am I doing something wrong, or does somebody else came across this problem?

Thanks, Lars

You have to pull the mute pin high to mute.
 
So you'd say, that these chinese copy-cats did it better than the guys at the small company called "Texas Instruments". Hmm, and they didn't noticed this yet? Well, interesting speculation.

You haven't defined "it" yet. What should they have made better in terms of performance you're speculating about.

When you can define that, it's possible to define a test procedure to check for this.

Initial starting point, get 2 fake Sanwu boards with "fake" 3118. Change one of them for a genuine 3118 sourced by Texas Instruments. Run the defined tests on both and compare the results. Finally discuss it.

Any other options?
 
Which comparison are you referring to? For now i only did optical comparision. Irribeo was so kind to send me one of these boards - mine didn't arrived yet.

Did we had defined any additional tests yet?

The other thing i checked was the AVCC fact - due to the new TPA3128 which states different AVCC levels to work with.
 
There are many ways to copy a chip. I don't know which technology TI uses for these devices, but there are ways to reverse engineer devices with relatively large features size and small number of metal layers, though it requires quite a bit of engineering investment. A much easier way to copy a chip is to "acquire" (with some insider help from the fab) the masks that are used for fabrication and use the same fab/process to manufacture copies.
 
my fakes sound really really great.
i have also purchased a batch of 3118 with certs for a new layout.

its possible the fake is a fraken part, completely different but with packaging adjustment.

I have thd plots at 3.5khz on my fakes. Do you have plots on yours?

-
 
Nope, i don't have. Wouldn't make much sense if not measured in the same setup anyway. If you want to compare to a genuine 3118, why not just desoldering the "fake" one and do the same measurement again, so you have both chips in exactly the same setup. 🙂

Then show both THD plots and we all can have a look at it. 🙂

(I don't have a setup to measure THD here.)
 
Last edited: