Exploding Tantalum capacitors

[fcwilt]

While browsing the Interwebs I ran into some info which I think is a fit for this thread. I'm quoting the info from http://jbwid.com/art82h.htm

 Causes: I can only speculate. Defective/marginal capacitor? Too much heat during soldering? 16-volt rated capacitor, without a sufficient safety margin for DCC? Tantalum capacitors have very low tolerance for voltages exceeding their rating, which quickly starts them down the path to thermal runaway.

This example makes me think that testing the caps, by applying a DC voltage (few volts lower than the rated voltage) to them for a minute before using them in the circuit will weed out any weak or defective ones. If someone does not have a variable DC power supply I think that even a 9V battery should do the trick.

It makes me wonder where in the circuit he connected the caps. If he connected it to the power from the rails going to the decoder it would surely do this. Keep-alive units have to be connected to the DC power rails inside the decoder.

Frederick

[peteski]

It makes me wonder where in the circuit he connected the caps. If he connected it to the power from the rails going to the decoder it would surely do this. Keep-alive units have to be connected to the DC power rails inside the decoder.

Frederick

True, hooking up directly to the DCC track voltage (which flips polarity many times a second) would certainly cause the caps to malfunction. However I think the fact that after he replaced that cap the circuit worked as expected (keeps the lights from flickering) seems to indicate that they were hooked up at the correct location, downstream from the bridge rectifier.

If you read the info on his web page (I provided the URL in my post) that is pretty clear.  I just quoted the part about the cap exploding, since that is what we are discussing in this thread.