We have to consider all possibilities and a scope check would be definitive.
Hard to see that any overload would harm the PSU cap's though. Depending on the current capability of the supply those parallel resistors might have been stressed. Perhaps even one of the diodes in one of the bridges has suffered.
With the offset at zero and bias set correctly check that both rails are equal value DC wise.
Just looking at the circuit and I think the amp should be DC stable without the input FET's in place. Why not remove the FET's, power up and having checked bias and offset (readjusting if needed) see if the hum is gone.
That sounds a bit odd... the frequency I mean. A DVM readout might be in error I suppose, particularly if it was a distorted kind of waveform. 100Hz is possible due to the full wave bridge rectifiers but not 60.
Would the signal transformer pick up stray hum with the FET's removed I wonder?
Without a scope we are guessing.
You could fit the FET's from the good channel as a test.
I'm thinking a bit deeper now... you say one of the 1000uF caps cooked. That means lots and lots of current has flowed through the gate of the FET concerned and also through the winding on the signal transformer. There is no other low impedance path. The 10k's isolate everything else so the power FET's and transformer took a hit.
I hate saying this when fault-finding
but I think you are coming down to a process of elimination.
Does the PSU also power another channel? If so and that is OK then the PSU is essentially proved OK.
Removing the input FET's might have been a bad call... can you swap them with known good ones.
After that its a case of output FET's or worst case the signal transformer. There is nothing else left.
Would the signal transformer pick up stray hum with the FET's removed I wonder?
Without a scope we are guessing.
You could fit the FET's from the good channel as a test.
I'm thinking a bit deeper now... you say one of the 1000uF caps cooked. That means lots and lots of current has flowed through the gate of the FET concerned and also through the winding on the signal transformer. There is no other low impedance path. The 10k's isolate everything else so the power FET's and transformer took a hit.
I hate saying this when fault-finding
but I think you are coming down to a process of elimination.Does the PSU also power another channel? If so and that is OK then the PSU is essentially proved OK.
Removing the input FET's might have been a bad call... can you swap them with known good ones.
After that its a case of output FET's or worst case the signal transformer. There is nothing else left.
OK and 
the transformer is good.
Just thinking aloud now, uncharted territory.. with the JFET's removed could we not short out the two secondary windings on the transformer. Just a quick dab across each with wire.
The DC conditions would be unchanged and any resulting noise on the output would have to be a result of noise on the rails.
the transformer is good.Just thinking aloud now, uncharted territory.. with the JFET's removed could we not short out the two secondary windings on the transformer. Just a quick dab across each with wire.
The DC conditions would be unchanged and any resulting noise on the output would have to be a result of noise on the rails.
OK, worth a try.
Also if you are an experimenter then how about a series cap, resistor and speaker across each rail to see if there is excess ripple there. Have a listen and see if you hear excess noise.
Say 100 to 1000uF in series with a 100 ohm and a speaker. Once the cap is charged the resistor can be reduced right down.
Also if you are an experimenter then how about a series cap, resistor and speaker across each rail to see if there is excess ripple there. Have a listen and see if you hear excess noise.
Say 100 to 1000uF in series with a 100 ohm and a speaker. Once the cap is charged the resistor can be reduced right down.
So.... a small update.
I tried so silence the transformer by putting a bridge from source to gate of one of the jfets (Soldering points on the pcb... jfets not in place).
The AC voltage on the speaker terminals was way lower, but still at 80mV AC (my DMM says AC but i can't tell the wave form excactly).
The hum in the speaker is still louder than with the jfets in place.
For the moment, the PSU is the main focus.
After that, i will replace the mosfets.
I tried so silence the transformer by putting a bridge from source to gate of one of the jfets (Soldering points on the pcb... jfets not in place).
The AC voltage on the speaker terminals was way lower, but still at 80mV AC (my DMM says AC but i can't tell the wave form excactly).
The hum in the speaker is still louder than with the jfets in place.
For the moment, the PSU is the main focus.
After that, i will replace the mosfets.
I am happy and appreciated that I could start to build F6 even don't really understanding electronics and audio.
I have 2 questions.
1) My transformer have ground line.
Should I need to connect transformer ground to chaiss or AC in my house?
Here in Japan don't have earth ground on house so may be no connect anywhere?
2) I have sic-sbd 650v 10A. Can I use them?
I have 2 questions.
1) My transformer have ground line.
Should I need to connect transformer ground to chaiss or AC in my house?
Here in Japan don't have earth ground on house so may be no connect anywhere?
2) I have sic-sbd 650v 10A. Can I use them?
