Need some pointers with finishing touches on mosfet amp repair

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi Welcome, Yes. The capacitor missing from the schematic is important.

If it, the coupling capacitor, can be removed to open the input, that might say if there is a leakage path through that input coupling cap. If the offset goes away after opening the input, it sounds like a DC leakage path through the cap. The DC offset circuit is very low current, so shorting the input to ground wipes out any/most influence that it has on DC offset.

The 500mV on the good and bad channel bugs me. That sounds weird to me.
....but they do have the DC offset control at the input.
Hi again, thanks for your replies. I will come back with the voltages AndrewT asked for later, but first...

There is actually a capacitor at the input, but it's on the earlier stages of the amp, not on the schematic. When I measure between ground and input, I get about 500 mV offset: on both channels. This voltage is present pretty much everywhere except after R54, where it is gone on the good channel, but even higher (again, about 600) on the bad channel.

I tried to short the input. This happens now: 1.1V offset, and the adjustment pot no longer does much of anything (before it was going from 900-600 mV one end to the other).

Does this say anything useful?
the input grounding resistor (1M0, or 2M2) should be first after the RCA/Phono input.
Then the DC blocking cap.
After that the DC output offset adjustment tapping.
and finally the RF attenuating R and C to ground.
This blocks DC from a faulty Source and filters excessively low LF and filters Radio Frequency interference.
The offset is allowed to work properly.

I would alter the front end to do all that first and then troubleshoot. The alterations will do no harm to performance and can only improve it.
To clarify what I just did: I cut off the outer leg of the 2K2 resistor at the input, and grounded it. The offset now shot off to 1.1 V (not mV) and the offset adjustment pot no longer does much of anything. A few mV here and there.

I'm pretty sure this is not part of the fault, but if anyone's interested, this is the earlier part of the amp, which has now been cut off entirely. As you can see, there is a DC blocking capacitor right at the output. C47/48:

Edit: Interestingly, it seems the offset is -600 mV with the resistor left floating, and -1.1 V with it grounded. When it is connected to the earlier stages of the amp, it is 600 mV. Not negative.
Last edited:
Joined 2004
Paid Member
Hi Welcome,
Those transistors are very low beta. You really need to see if they are matched to each other. Have you checked the gate protection zener diodes? You may also have a transistor or two that has gone leaky. MPSA42 and MPSA92 aren't exactly suited to drive those gates that way, and they sure as heck don't make happy diff pairs. This amp was designed on a severe budget. You should look into using better transistors suited for a signal stage, and maybe some TO-126 types for the drivers. They really could have used another current stage in there before the Mosfet gates. That would have allowed you to use signal transistors in two locations and proper driver transistors.

I would try a matched pair of 2N5401 for the input diff pair.

Founder of XSA-Labs
Joined 2012
Paid Member
Try getting some new MPSA92 on the LTP input and match them for Hfe. A bag of 50 costs $3 and should have enough for a perfect match. Use good high quality 1% resistors on R56 and R55. That should help with your offset. I have an amp with single pair of same MOSFETs and matched the inputs and offset is 1mV with no potentiometer to adjust.
Joined 2004
Paid Member
Hi xrk971,
The 2N5401 devices are higher gain, lower noise and rated for 150 Vceo. This circuit screams out for real signal devices in the diff pair. Try them yourself. Given that they are still current parts, you can buy real ones. You need to buy any transistors through the proper distribution chain to be certain they are real. Otherwise, everything you posted is correct. The bag of 50 might cost you more, depending on where you buy your parts from (not Ebay).

...if you do swap out the diff pair for some higher gain, matched, devices, maybe consider chucking that whole DC offset circuit. Just a thought. They've got that NP cap bypassing the AC path there. If it shorts, you'll have another DC offset problem.
Just hammer the input to ground with a 50k~100k resistor. Properly ground referencing the input makes more sense to me. Then you, at least have a proper reference. With a higher gain, better matched diff pair, you might not need any DC adjustment. If you still need a DC offset adjustment, I'd think a better place would be between the two emitters of the diff pair. One step at a time. Fix the offset first, then find out what makes sense for you.
Thanks dudes!

I replaced the input pair as well as that other pair, I just took four out of the packaging and all devices had a beta of 78... interesting how consistent modern parts are. Originals were anywhere from 40-80.

DC offset = fine!

So this is now a closed chapter. Your suggested mods are very nice, but this is just a PA amp for blaring loud music so I guess I'll settle with its original circuit. Worked well enough before it burned up. ;)

Thanks again, all, I really appreciate your help.
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.