Tripath TA103 blowup

[tiroth]

OK. I managed to blow up my TA103 within the first 15 minutes of testing. I removed the load (dummy resistor) while under signal, protect clicked on, and even after reset the amplifier never comes out of protect. Joy. The board is supposed to be stable into an infinite load but it looks like this is not so true.

Test info
I first checked for shorts in the output stage, that looks OK. Output diodes also measure like diodes still.

The Tripath driver seems to work normally, draws about 40mA when the 5V line is powered, which drops to zero once VSNEG and VSPOS are powered up. There is no excessive current draw on any line, but the VSNEG supply does have measurable draw whereas the VSPOS supply is essentially doing nothing.

First thing I noticed was the output of the National switcher powering the 12V line is only 2.8V. Is it likely that this part has failed? I couldn't really see how from the schematic, but the internals of the driver are hidden of course. Could it be working OK and the 12V line is just being pulled down due to a fault elsewhere?

Anyone have any tips on isolating the problem? Thanks very much for any input.

P.S. Yes, there is a lot of switching noise. Very noticeable on the scope as the bandwidth is limited from 1MHz to 30kHz.

 

[subwo1]

tiroth, how 'bout ohm meter the 12v input to the chip to its gnd and other pins to detect shorts. You might also disconnect the 12v switcher and substitute something like a 50 ohm dummy load to test it.

 

[theAnonymous1]

Those Tripath modules are more sensitive than a third degree burn.

I've probably killed 5 or 6 TA0104 just by looking at them wrong. My advise would be to not waste too much time or money and move on to something else.