Valve guitar amp fault -need help on one aspect of the overall circuit (specifically involving FETs)

I've a valve guitar amp that's crackling (Blackstar HT-5R - common fault apparently as outlined here -

)

There's one part of the circuit I'd like to understand (but it's a puzzle to me at least), here's the part...


(PA in = Power Amp input signal (in other words the input signal from the pre-amp)

FYI the overall (full) schematic is here - https://vianegativa.files.wordpress.com/2012/07/blackstar-ht5-schematic.pdf

HT_PS is 450V DC, but I'd like to know what typical DC voltages I could expect to see on the gate, drain & source of those FETs (TR2 & TR3)

In fact I have no idea what's going on with the circuit connected to the sources of TR2 & TR3 ....I'd imagine that the gate of TR3 sees a replica of the TR2 gate signal ...but how that's acheived is beond me...any light anyone can shed would be most welcome!

This circuit is a pretty classic differential long-tailed pair with a constant-current source. TR3, the non-inverting input, is biased to a constant 0V for a single-ended input with a differential output. The circuit with TR4 is set to draw a constant current of roughly 7mA through the MOSFET pair (~3.5mA per MOSFET).

For PAin = 0V, I believe you would probably expect to see about -3V at the sources, about +370V at the drains, and about 0V at the gates.

Just for reference, here is a simplified view of what is effectively going on, with the differential probes marking the differential output of the amplifier stage. As an additional note on the circuit above, D21 is a protection diode that prevents the voltage at TR4 from exceeding the maximum rated VCE for that transistor.

That's very kind of you - many thanks.

R38 /39 are overheating badly (apparently this is a known problem & happens due to voltage leakage from the drain to the gate - the solution is apprently to drill a hole in the pcb between the gate & the source (it's all in the video at the top of my opening post, but at least now I know the correct approximate DC voltages & (what to expect AC signal wise) & can progress acordingly.

Thanks once again!

Edit: Problem now fixed...that youtube video was bang on the money - leakage due to a bad pcb between the FET drain & gate ...drill a hole in between those pins - problem gone!

i would have scraped the PCB and painted it with clear enamel. if it was carbonization of the PCB from heat, it usually starts on the surfaces.