Help! strange phenomenon in 6 channel system, partially self-built

Hi all!

I need your help once again, I didn't find similar topics on the forum yet, and I hope you have the patience for this :)
I have a strange phenomenon going on with my 6 channel amplifier. Spoiler: 50hZ noise, but with surprises.

I recently built a 6 channel amplifier after many years of planning and plans changing for a 3 way system with active crossover. Basically I used a stereo 3e audio tpa3255 and a 4 channel tpa 3251. These boards have 20dB gain. and I need more, so i tried to design myself 3 stereo preamps. They are based on a OPA1656, i attach the schematics (amp__section and supply section, split for ease of reading). The supply section of each of them is connected to a transformer, the centrals of the transformers and the grounds of the PCBs are connected together by thick wires. The ground of the PCBs is a thick ground plane.

The connections are the most obvious you can imagine (see layout): basically, wires from the external RCA connectors to the IN terminals, and wires from the OUTs of these boards to the INs of the 3eaudio boards. I tried also some variants, that didn't make any change, including connecting together the grounds of the 3E audio boards.

If I exclude the preamps from the chain, the 50 Hz noise is much lower, but still there and not a pleasure for the ears.

So, there is a 50 Hz noise. The source is my laptop with an Asus Xonar u7 usb soundcard, and i use the surround speaker outputs. They have an output impedance of roughly 700 ohms, and this might be important!!!

What follows, is with my self designed preamps included in the chain:

I tried to disconnect all speaker drivers from the amplifier except by one to make some experiments. Obviously, if disconnecting 5 speaker drivers is the only change i introduce, the noise is still there on the remaining speaker driver.

BUT: i tried to use the headphone out of the xonar instead of the speaker out: no noise at all!
Then i tried to use as a source a portable mp3 player: again, no noise!
And then, i put a portable headphone amplifier between the speaker out of the Xonar and the self designed preamp. again, no noise!

The headphone out of the xonar, the mp3 player and the portable headphone amp all have a very low output impedance, but i am not sure that this is the reason, even though i strongly suspect it is. And if so, why is the output impedance of what goes to my preamps making such a difference? My preamps are also systems with a high input impedance and a low output impedance.


Does anybody have an explanation for all of this? Should i be worried about some possible major flaws in the design of the system, and try to change something, or should i just buy, for instance, a few headphone amplifiers to solve the problem? I was thinking about going for some affordable Aiyima H1.

Thank you a lot in advance for any input!

Cheers,

Sergio
 

Attachments

  • amp_section.png
    amp_section.png
    16.7 KB · Views: 51
  • supply section.png
    supply section.png
    20.2 KB · Views: 53
  • layout.png
    layout.png
    59.1 KB · Views: 48
Yes, i guess it is a ground loop with the laptop, but why if i still use laptop and usb soundcard, but with the lower output impedance headphone-out (or if i put a headphone amp in the middle), then the noise is gone? I suspect there must be some circuit design knowledge i am missing here.
 
Here is a drawing of the system connections!
I hope the colors make clear what i don't write.
i simplified the drawing by considering just 2 preamp boards, and considering them mono.
Both configurations gave exactly the same 50Hz noise.

in the experiments i made, just consider portable headphone amps in the middle between the xonar and the preamp, or a portable player instead of the xonar, or the xonar's headphone out instead of the speaker out

edit- i forgot to draw the class D supply - it is just asingle 32V SMPS connected to both of the class D boards
 

Attachments

  • amplisystem.png
    amplisystem.png
    27.4 KB · Views: 23
  • amplisystem variant.jpg
    amplisystem variant.jpg
    124.7 KB · Views: 28
Last edited:
One of the outputs (- most likely) is connected to mains earth probably. This goes back to the earth of the laptop and so to the Xonar. Big loop outside of what you have drawn. In cases like this, you also need to draw all the power supplies, also from the laptop and how thoses are connected to mains earth.

Have you tried the laptop on batteries?
 
Here is a more complete drawing, here there is really everything.
Using the laptop on batteries or on wall plug doesn't make any difference, and i am not surprised because if i am not wrong, laptops run on galvanically isolated supplies.
 

Attachments

  • amplisystem.png
    amplisystem.png
    32.8 KB · Views: 20
Strange that it doesn't make a difference on the laptop with batteries. Galvanically isolated does not mean that earth isn't connected. Negative is normally connected to earth in the psu. I know it is the wace in the laptops I do have (all 4 of them from different brands).