Can a bad cap kill an LM3875?

[SRMC]

I built my first Gainclone from Thorsten's inverted LM3875 schematic and one channel works perfectly. However, the first time I powered up the other channel the chip sizzled and fumes came out approx. 3mm above pins 3 and 4 leaving a nice little crater. I have checked to make sure that I have it wired correctly and didn't wire any of the polarized caps backward (been there done that). At this point I can't figure out what went wrong, but I want to learn from this. Could a bad cap on the V- cause this or did I do something wrong?

The cap in question is an old surplus 1000uf 25v cap (all I had left after blowing up my last 50v Panasonic FC) and I don't know how old it is. I have a 260va Plitron w/ dual 18v secondaries and dual bridge rectifier that also powers the other channel so I don't think that the power supply is the problem.

Thanks,
Steve

Hoping to soon have a Gainclone with both channels working

 

[carlosfm]

25v cap?

Hi Steve,

With that transformer you shoud at least use 35v caps.
The cap may blow up because of over-voltage.
Also, are you shure you mounted the cap with the correct polarity?
Remember, on the V- polarity the positive lead of the cap connects to the circuit ground.

 

[JOE DIRT®]

SRMC.....when anytime you are building an amplifier...first have the outputs of the power supply fused and not connected to the circuit.......apply a light load with a power resistor and then check the voltages.....let it run for awhile....keep a lower fuse value in and connect the amp if all checks out.........another thing is to use a signal injector and make sure the correct path is taken......one little short and you will blow the power amp...hope this Helps!



Cheers!!The DIRT®

 

[Peter Daniel]

It doesn't seem that cap was damaged, so it's probably OK and correctly connected. Maybe the insulating pad wasn't working correctly and you had negative rail shorted to ground?

 

[SRMC]

Thanks for the help. I have fuses after the rectifier on the power supply output (3a) but I'm sure they are way too high. I used the same mica insulators under both of the chips with a small amount of heat conducting compound that is supposed to be electrically nonconductive (http://www.arcticsilver.com/arctic_alumina.htm). My fear is that although I was using a stepped insulating washer that should have kept the screw centered in the mounting hole and prevented it from touching the chip, the screw attaching everything to the heatsink/case may still have shorted the negative rail to ground.

Thanks again.

Steve

 

[SRMC]

Problem solved!

It looks like the prize goes to Peter! It turns out the V- was shorting to ground. I found that I left a tiny burr when I drilled the hole to mount everything on the heatsink, and with the thin mica insulators I was using it still was enough to contact the back of the chip.

Thanks.