Arcam AVR280 damaged amplifier board

[tony359]

Hi all
While repairing an old Arcam for fun (a TON of bad capacitors) I  unintentionally damaged one of the two amplifier boards on the unit. I didn't realise the V+V- caps were still charged and I plugged the main amplifier PCB connector - likely misaligned as it was quite tight in the bunch of cables. I saw sparks. Now I have V- (-53V) out of three out of four channels of that board.

Can I pick your brain on what could likely be the cause of this? I have been doing some preliminary testing and so far I've found Qx08 with emitter and collector shorted on all the affected channels. The main transistors seem to be ok.

Since I have voltage on the outputs, the PROTECT line goes HI and the amp shuts down.

I also believe I have found some Mute transistors not working (??) but I don't think that was me!

Schematics attached. The connector in question is P103A. That has V+, V- and all the outputs. The connector can be partly plugged misaligned but... I might have pushed -53V (V-) into BSRO. Still not sure how it's possible that three channels died. Could have been the sudden flow of current - to the correct inputs -  from a charged capacitor to damage the components?

And yes, I did check that the caps were empty before plugging it back. Unfortunately I didn't realise that without the amp connectors plugged in, that board has no ground. So I referenced my multimeter to the wrong ground and it was telling me that it was ok. It wasn't...

Thanks all!

[tony359]

Also, could someone kindly help me in finding a replacement for an RN1241 muting transistor? They don't seem to be available anywhere. Happy to add the bias resistor myself if that's easier.

[tony359]

No replies but I'm going to follow up on myself for future reference.

I found Qx02 and Qx08 shorted on three channels out of four.

I've powered the amp board with external power and it works - even though I am getting 50mV offset on those channels while the good channel is 0mV. That said, the good channel is a Front channel and the one I broke are surrounds and I think they're not identical. I'll double check once I put everything together.

Any ideas of where to look if I found that the offset is higher than the other board?

Thanks!

[David_AVD]

Check the resistors around are of the shorted transistors. Also, Qx04 & Qx05 need to be gain matched for minimal offset on the output.

[tony359]

Thanks David,

I checked them all and they are all good - and matching a working channel.

I did not touch Qx04 and Qx05 so I hope those are fine!

The board works, it's back in the AVR. However, the offset at the outputs has slightly risen. Working channels from the other board (and one from the one I fried) read 0.005V, the channels I have repaired read 0.036V. The channels seem to be working fine.

I feel it's ok but at the same time I feel there is something I left behind!

[David_AVD]

Most manufacturers will set the acceptable limit as +/- 50mV but I try and have them less than +/- 10mV.

[tony359]

The weird thing is that only the channels which failed show an elevated offset.

All the other, including a channel which "survived" on the affected board, show a lower value.

As you say, 36mV is likely acceptable but it feels as if something happened to those channels. I doubt it's a coincidence!

I did not replace the matched transistors so - unless they got "smacked" by my clumsiness  when plugging the board in - it could be something else. Any ideas?

[David_AVD]

Check all of the low value resistors (say, under 1K) to make sure some haven't gone high in value.

[tony359]

To be honest I think I checked them all and more than once. The unit works, I pushed it and all channels deliver what ARCAM specs. There is no noise or any weird behaviour so I am a bit lost.

[David_AVD]

It could still be a gain difference in the Qx04 / Qx05 differential pairs. It may be coincidence and those channels always had a slightly higher offset than the others, or that they were stressed by the other events and changed?

Either way, a transistor tester that measures gain is very handy to have. Just last week I fixed an amplifier where one transistor had gone very low gain. A simple diode junction test with a DMM didn't show that.

[tony359]

I have one of those testers, it's one of the many clones "M-Tester", I think it's Arduino based. But I read online that the gain read by those devices is not very accurate - maybe it's accurate to compare to another component? What do you think?

As you say, maybe it was just a coincidence. Or maybe it could be that sea of poor quality capacitors the unit features - I replaced two bulged ones I had on the PSU section, causing trouble and then I started testing the others adjacent to the PSU area and oh boy, most of them had gone out of spec. I had to stop or I would have had to replace them all.

[David_AVD]

I have a couple of the BSIDE ESR02 units and find them very useful. Even if the absolute gain is not correct, you only need to match them to each other.

[tony359]

Amazing! I was led to believe that those were useless for that purpose. As you say you're just matching them so the absolute value's accuracy is not so important. Thanks!

[tony359]

I made a video of my repair (the amp board repair will be in the second part though).

Just for reference, I do not make money out of my videos, it's just a hobby.

[tony359]

Bringing up this old thread to share Arcam's Firmware utility and Firmware I used in the video (second part actually which I never shared!).

One of my viewers asked for it and I didn't save it so I acquired the files again and I thought of sharing them here for everybody to use. A link to this thread will be on the video description. The Firmware file contains firmware for the AVP700, AVP701, AVR250, AVR280, AVR300, AVR350. Plus the DSP software 1.12.

I hope this helps!