Hi,

If there is no noise when you have nothing connected but noise when you connect something, it is usually a grounding issue. Both channels? Is your signal source separately grounded or floatng? Did you try connecting a battery operated source, like an MP3 player or similar? Is your power supply earth connected or is the ground of your amp floating?

Which components, best to look at the schematics. Do you have it? But my guess is that it is not a component problem.

Jan

Thanks Jan for replying so quickly !
OK, that's good news, because I re-soldered many components that didn't looks good (specially yellow caps).
So I still have ground problem...
I have heavy wire type for gnd (last pgnd of AMP5), and for earth.

But there are copper type (brown wire), is it OK ?

I have tested without connecting the fourth PGND, it's the same. But now, if you're telling me that's a ground problem, I will double check that.

I remember that without connecting the fourth PGND (of the board) to chassis, and without connecting earth, I had the problem (so board "floating" and transformer too).

Now that you're asking about earth on power supply, here is what I have :

My power supply (transformer) isn't connected nowhere but only AC1<->transformer / AC2<->transformer / First PGND<->Center tap (two last wire of transformer)

I haven't made any other connection between the transformer and chassis/earth.
Must I connected something concerning the transformer ?

Hi,

Connecting center tap to chassis and / or earth ior not s a topic of much discussion and also a topic of electrical safety and I am no expert.
It also depends on how your signal source is done, grounded or floating. Some equipment have a float / ground switch so you can toggle.
If the signal source is grounded (to earth) but not the amp, you can have low level current through earth from the amp causing the noise.
If both amp and source are floating, you may also have strange effects. So how is your source, grounded or floating?
If both are grounded but at different points there can also be ground loops.

My source are the RCA's, with the ground part of RCA connected to g1 (or g2), (vol pot between). Then, if I'm connecting nothing to RCA's or connecting my computer, the result is the same.
Maybe it's coming from the pot ?
But in that case, the g2 will have the problem too.

I can test connecting usually wires In1 on In2 to see if it has noise too.

Jan, how about this post ? It's for AMP10, but I think it is similar to AMP5 ?
I think I will check again the decoupling caps (yellow caps).

Can you just give me the method to measure those caps or to know if they are working properly ?

Wait, maybe I made a mistake on inputs. I will test and post results here.
For, now I will take some pics.

OK here are the board HQ pics :

Overview
Inputs view (In1/g1 and In2/g2, then, mute section)
Near input caps (scratches)
Near input caps (other side)
My (beautiful :p) toroids inductors
Side view of toroid inductor
Back side of the board overview
Back side/Tripath section
Back side/Some yellow caps
Back side/Two other yellow caps
Back side/Speakers section
Back side/Relay/One toroid inductor
Back side/Input caps
Tripath back side view
Tripath side view

I'm trying to make some sketch.

OK sketch almost done. For now, I will not make any tests anymore.
It seems that there is noise on both inputs, and it's depending on how high the vol pot is.
As soon as sketch's done, I will post it here.

OK I have add a file for the sketch. The quality seems good, but I'm sorry if it's not very "clear", I'm not used to do such scheme.
Tell me if it's not clear enough, I will try to draw it better.

If problem is somewhere with connections on the amp, see attachments for a sketch.
If problem seems to be on the board, I would like to know which decoupling caps is used when input connected (maybe it's one of them, like this post for an AMP10).

What I can see from the pictures: many solder joints are not solderd through the board, solder is not showing on the top of the PCB when solderd from the bottom side. This indicates the soldering iron temperatire has been too low or that the solder periods have been too short, so there may be "cold joints". With perfect joint, both top and bottom side will look more or less the same. Now, there may be solderings that look OK from the bottom side, but actually have a bad or no connection. I also see lots of residuals from soldering. It may menas small, thin conductive streaks across where they should not be. I suggest you try brushing / cleaning this off after soldering. Do you know someone with more soldering experience that can help?

Measuring capacitors: you need a capacitance meter. Some DMMs have this. Better DMMs can measure both very large and very small caps. Simler DMMs can only measues intermediate sizes (like 1nF to 100uF) It is best to measure components off the board, as the circuit always influences measurements. If measurements look OK on the board, they may be OK, if they do not look OK, it may be other compoennts influencing the measurements. So it is hard to draw conclusions if they are still mounted in the circuit, unless you have lots of knowledge and experience and can draw conclusions from the schematic. Plastic and ceramic caps are seldom damaged unless badly abused, while electrolyths can be damaged if mounted with reverse polarity or overheated.

OK it will be a little hard to say if components are OK :(
I have some news : after resoldering almost all decoupling caps and putting the board in the case, I don't have anymore noise (just a little bit, but much much lesser than before), nevertheless I have a kind of ring in the high frequencies specially on one channel (called 'Hizzz' in the 41Hz FAQ ?).

I will investigate, but I think my amp is almost OK :)

Problem solved by V-bro. I was the axial choke that was not good :

See here.