Customer stated that they inadvertently connected the speaker output to a mixer rather than the line output, and claimed it stopped working, no longer any output. Quite how this catastrophic failure occurred as a result of connecting to a high impedance input, not sure!
Clean some of the compound away to see the damage (obliteration and vaporisation of the ICs and surrounding tracks)!! This one is going to be a challenge:
Ordered replacement ICs (MPS 1019), whilst waiting to arrive, attempted removal and cleanup… Trickier than expected as the welding-action of the failure meant they were quite firmly stuck to the board. This is what remains of the poor ICs (the die fell out of one!):
Managed to start laying down some new copper traces to solder new chips to:
The other channel was not as bad:
After fitting new ICs
With a penny for scale:
Unfortunately there was further damage, the unit was stuck in ‘power-up-mute’, and turns out the signal from the microcontroller wasn’t getting out (the microcontroller I/O port itself was damaged so was not replaceable as had no means to program it), so a bodge was made to permanently leave the unit out of mute.
It meant a little bit of digital noise can be heard during startup if you listened really closely, but otherwise is undetectable during operation.
Calling that one a win! 🙂
Quite why connecting speaker outputs to high impedance input would result in such devastation is a bit strange though, and even if it was producing full output and load disconnected, the catch-diodes should have prevented any nasty voltage spikes from the LC filters?