McIntosh MC2100

[xtian]

Owner brought McIntosh MC2100 with a short and blowing the 5A mains fuse. I found one shorted power FET and shorted rectifier. I replaced all the power FETs (12) and supporting 5W emitter resistors, rectifier, and replaced all electrolytics. I got the amp to power up successfully, tested voltages and everything was running OK, drawing about 20 watts at idle.

Then, while running, I switched from Mono to Stereo mode, and 5A fuse blew. So I put my 75w bulb limiter back in-line, back into Mono mode, and powered up successfully. But then I could not recreate the issue by switching into Stereo mode. No issues with the switch. Shut down and started up a few more times, and I was seeing about 50/50 failure (bulb remains bright), with no correlation with Mono/Stereo setting.

I pulled the output cards, and no fault, amp idles at about 12 watts. Put each card in individually, and was getting faults fairly regularly. During fault, disconnecting rectifier for +/-35v rails solves the issue, so fault is downstream of the main rails.

I was not able to measure shorts in any of the transistors on the output cards, but lacking any other good ideas, I replaced the four PNP and NPN transistors on the cards--only the small ones with TO92 package...three other transistors are in metal cans, and were not supplied with my upgrade kit. This did not change the behavior. Now I am reliably getting a fault with either or both cards installed, but no issue with cards removed.

I'm at a roadblock. I don't know how to further test the output cards out-of-circuit. Any suggestions?


MC2100 Service Manual.pdf

[Helmholtz]

I don't see any (power) Fets.
Did you check transistor pin voltages?
Schematic voltages are provided as they are useful for troubleshooting.
Which of the replaced transistors were bad?
No sense replacing good transistors (they don't age).
Original parts may even work better because they have correct Hfe.

It may make sense to disconnect the autotransformer for troubleshooting (it's a DC short).
Check for DCV at output without transformer.
Is one of the channels good?

[g1]

Mono is a bridged mode for this amp. I think switching while powered up could be trouble.

[xtian]

Thank you, gentlemen. I was napping, and the idea of oscillation popped into my head. I was running tests with no load connected! With a load connected, and a 75w LBL in-line, IT WORKS! Passes signal. No runaway. Drawing 25W from wall thru LBL.

However, I get random startup fails in Stereo mode--still plays, but draws full 75w from wall with bulb glowing brightly. No issues in Mono.

If this is indeed oscillation, where's the best place to scope that? And, I'm very wary of clipping my earthed probe to the chassis (this amp has no earth connection) so please advise.

[xtian]

BTW, there very low (~25mV) DC at the outputs.

[Helmholtz]

BTW, there very low (~25mV) DC at the outputs.

If that's with the autotransformer connected it doesn't mean much as the transformer shorts DC to ground.

Without transformer it would be fine.

[Helmholtz]

If this is indeed oscillation, where's the best place to scope that? And, I'm very wary of clipping my earthed probe to the chassis (this amp has no earth connection) so please advise.

A SS amp should be stable without a load.

Oscillation can be scoped at the output.

Do not connect your scope ground to any of the hot output terminals.

[g1]

Are the random start-up fails on bulb with load connected, or no load?

[xtian]

I no longer have my variac/autotransformer in line, only a 40w LBL. It will power up stable in Mono mode, reliably. Stereo mode is troublesome. I can sometimes get it to work, but often it wants to run away. I am using a pair of 8-ohm speakers as loads. When in fault mode, the audio stays clean and clear.

I swapped the output cards L-to-R, same issue.

I wonder if my lead dress is to blame? There are longish wires connecting the power FETs to the card connector, and when you pull the heat sinks to access the FETs, these wires bend out of shape, and I've done my best to tuck them in when reassembled.

[g1]

When in mono mode, do you have the output configured as per owners manual? Or still stereo loads?

[xtian]

Also, re: oscillation, I replaced the 0.02u caps in Zobel network on the OTs secondaries. Didn't help.

[Helmholtz]

I no longer have my variac/autotransformer in line, only a 40w LBL.

I meant the autotransformer(s) at the power amp outputs.
You might also call them OTs or impedance matching transformers.
If these are connected, DC fault voltage at the output would be shorted to ground possibly causing excessive amp currents.

There are longish wires connecting the power FETs to the card connector, and when you pull the heat sinks to access the FETs, these wires bend out of shape, and I've done my best to tuck them in when reassembled.

What FETs??

Still waiting for answers.

[xtian]

When in mono mode, do you have the output configured as per owners manual? Or still stereo loads?

In Mono mode, I still have one 8-ohm speaker connected to each channel's 8-ohm output. But this is stable. It's Stereo mode I'm having trouble keeping stable. It's been playing a stereo track for the past 20 minutes just fine. But each time I start up is a throw of the dice.

[xtian]

What FETs??

Still waiting for answers.

In the schematic at top, Q15,17,19,21,23,25. Six per channel. I replaced all 12, because I found one shorted when amp arrived from client.

[pdf64]

In the schematic at top, Q15,17,19,21,23,25. Six per channel. I replaced all 12, because I found one shorted when amp arrived from client.

That’s confusing, the device symbols on the schematic, and the operating conditions, indicate those should be regular bipolar junction transistors.
What was the original device type part reference, and what did you use to replace them?

A 40W bulb may be excessive, reduce the mains voltage a bit?