Fails Open or Closed

Last night I did first ever real repair. I fixed a TS-9 Tube Screamer for a friend. It had a diode from +9V to ground that had failed closed. I thought diodes usually fail open.

Anyway, maybe some of the more experienced people here could tell me how parts typically fail (open/closed). I'm curious about resistors, caps, voltage regulators, diodes, transformers and ICs.

Diodes almost always short. this goes for many other diode based parts like SCRs and Triacs.

resistors almost always fail by burning open or changing their resistance.

Caps usually fail open/vented but sometimes will fail shorted if hit with a high tension voltage. this is usually followed by overheating and venting.

voltage regulators usually fail in two ways. low voltage/current failure usually shorts the input to the output. High voltage/current failure usually includes explosion.


Transformers can do many things. usually an open winding, sometimes a shorted turn.

ICs can fail in many many ways, not really a certain way of doing it either.

N-Mosfets usually short source to gate and then turn themselves on continuously until overheating and explosion. same for P channels.

This is not a certain list by any means because there will always be the exception to the rule. Just be careful and plan for both.

They always fail in the direction that does the most damage...

absolutely.

Thanks for the replies. They're good for a first guess in repairing something.

[quote author="dale116dot7"]They always fail in the direction that does the most damage...[/quote]

Yeah, the guitar amp builders say a $100 transformer will blow every time to protect a $0.99 cent fuse.

nah it's more like the trafo was poor quality(not related to price at all..) and died taking the fuse with it.

http://www.badcaps.net/forum/showthread.php?t=1219

http://www.avxcorp.com/docs/techinfo/failure.pdf

> I thought diodes usually fail open.

9 of 10 times, they go short.

BUT, diodes are often put in places where a short will pull BIG current. Like the front of a power supply. The power transformer groans a bit, the diode melts open (unless something else fails first).

In this case, I bet the wall-wart did not have enough muscle to blow the shorted diode open.

The heat gets the migration going, the N material becomes more P like, the P material more N like, so the semi conductor becomes less of a diode, and more like a selinium rectifier, which means more current, more heat, and eventually you get a short with a few ohms on it, depending on the junction and the current available to the damage zone. If the contaminated chunk is big enough for the current, it will stand. If not, see above.

Or, you can get the avalanche effect, with a little thermal runaway, that always works. This would be your dc failure mode, reverse breakdown, the whole nine.