AMP11 blown up. Any ideas?

[V-bro]

That is highly peculiar, I've built so many AMP11 amplifiers and none have blown in such a way, let alone twice!! I'd be tempted to think that there is some component on this particular board causing this... Hard to imagine which one though...

[MickB]

It's a completely new board, V-bro. It's one I bought from 41Hz a few months ago and assembled myself. The first one I bought from you and was several years old so it would have been a different batch of components anyway. So logic suggests that's it's nothing to do with the amp itself but must be down to one of the other parts of the system or something to do with the way it's been used.

Skating indoors does tend to build up big static charges, enough for it to be painful when you touch somebody. I've not noticed it outdoors, but is it possible that touching the heatsink while carrying a static charge would set off some sort of chain reaction? I isolated the heat slug of the TP2050 from the heatsink but I've noticed that it is connected to ground on the board through a couple of the mounting screws.

For the other parts of the system:

On the input side the preamp has changed since last time. The first time I used a pair of Neutrik audio transformers as a passive preamp. This time I fitted an active opamp preamp as I wanted to add an adjustable bass boost/cut control (which worked very well, BTW!). Overall gain was about the same though.

On the output side there's just the crossover and the speakers. Is there any way the crossover could cause this? Such as an intermittent short or something?

[V-bro]

Well that makes it even harder to lay a finger on then... I would recommend to always isolate the whole board from the heatsink, even though some 41hz designs have it connected to ground.

Static discharge, well that has never caused this kind of disaster as far as I know... Was the previous disaster already with the active preamp/filter? I wouldn't be surprised it is down to some turn on DC peak or shutdown peak from that circuit...

[MickB]

No, the first time it was with the passive preamp with the Neutrik transformers. Both times it failed while it was on and had been running for a while, it wasn't on turning on or off. The only parts which are the same for both failures are the heatsink, crossover and speakers. Also the battery and power switch of course, but I can't see how that could affect anything when it's all been running OK, either it works or it doesn't.

[V-bro]

It's a bizarre thing, if it were input overloading it should just damage the input opamps, not ruin the whole power stage if you ask me... It's like it's been severely oscillating somehow. Are these AMP11 with the torioidal output filter chokes or the ready made ones?

[MickB]

First one was toroidal, second one was ready made. It seems determined to maintain the mystery

Is it possible for the crossover to have anything to do with it? Or could a short on the outputs have this effect?

And continuing to scratch around for ideas without really knowing what I'm talking about: the 3 tweeters (DT-28N) are connected in parallel and they're 8ohm each so that gives a nominal impedance of 2.7ohms. Could that be overloading the outputs when running off 24V (nom) supply? Crossover is at 2.2kHz

[V-bro]

I don't suspect the tweeters, they are also filtered and at the operating frequency are not a dead steady impedance, usually higher (look at the impedance curve) Tweeters don't really eat much of power, so usually are not a problematic load...

Too bad I don't have the board design plans of the AMP11, would make it lots easier to figure out what's happened. What I can do is go by the schematic, see where the overload has taken place... Let's dig in to this!! Can you give some more precise directions what pins (numbers) are blown??

[MickB]

Thanks, really appreciate your efforts yet again V-bro. I know how busy you are.

Yes, on both amplifiers pins 5 & 6 disappeared completely. According to the TP2050 datasheet that's GRD2B and GRD2A (negative power supply for Channel 2). On the second amplifier pin 14 vanished as well and pin 13 was hanging by a thread, they are GRD1B and GRD1A (the negative power supply for the Channel 1).

Here's the schematic from the TP2050 datasheet, hope it's big enough to read...

TP2050 outputs.jpg

[V-bro]

What could be the case is that it is simply due to overcurrent. Mind you the protection circuitry isn't really fast enough for massive transienst. The pins that are toast are the power stage ground reference pins, these have to carry the highest current in the whole amplifier.

Maybe it help to use an STA516B power stage next time...

[MickB]

That does seem to fit well with all the circumstances. It might also explain why it failed more quickly this time as I've just fitted the bass/boost cut which would mean higher current when the bass is turned up.

So would running it with the volume too high be enough to cause overcurrent? When I use it myself I turn it down when it starts to distort as I want it to sound good, but most of the time it's being used by other people and sometimes the volume gets wound right up and they don't seem to notice. Plus it's being used in a noisy environment so the distortion is less noticeable. On the other hand, both times it failed it was being carried by sensible people who I don't think would just wind it up full and not notice

Or do you think the transients might be spikes caused by noise on the input or something? There have been problems with the wireless transmitter which has meant the signal cutting in and out which sometimes produces loud pops and crackles.

Is there any way of protecting against this? Would clamping diodes on the input be sufficient? Because it's used by lots of different people who aren't all technically inclined it really needs to be bullet proof.