How to blow an output transistor

Status
Not open for further replies.
Back to his question, and also reported by a buddy (who is a local amp repair guy) is that OP devices mysteriously blow upon turn on. He has seen it too many times to count.

I do not believe the chances of a device dying unpowered is worth discussing, however there is a definite correlation between turn-ons and SODS. (Sudden OP device death syndrome). :apathic:
 
sam9 said:
And since the "sudden death" is so sudden it could be almost impossible to tell the difference between already-dead and dead-due to-turn-on unless you suspect the possability in advance and hook-up some sort of instrumentation to record it.


So by process of elimination, you assume an elf did it while powered down? 😉
 
Barring immersion in acid or some radical physical impact, solid state devices will not die while dormant. It will be the power-on current surge that does the trick.
Tubes will very, very slowly deteriorate over time, even sitting on the shelf; the vacuum slowly degrades. However, the time frame for this--assuming a well made tube--is on the order of decades to centuries, so it's not the kind of thing you would normally worry about unless you were talking NOS tubes from the forties or fifties. Since easily 90% of the so-called "NOS" tubes out there are actually used, you're actually faced with an entirely different set of wear mechanisms if you're buying old tubes. The solution is easy: buy new ones.

Grey
 
The momentary short of the tip of the speaker cord plug as it exits the jack will not blow up the amp. If you were not playing at the moment, there would be no voltage on the tip, so no current to draw. If you were pplaying, there would be the instantaneous currrent draw, but the limiting circuits would prevent harm.
 
I've repaired a few hundred amps. time after time I've heard that it failed on startup. While repairing an amp I "softstart" it with a light bulb and variac. The last thing I do is turn it on under worst case conditions: While it's still hot, at high input voltage (worst case is presumed to be 132vac), with a signal and a load present.

I presume that the outputs fail due to a startup race condition that would not normally be seen during operation. Since most amplifiers are some form of a complimentary ouput stage, I think that in the few microseconds when the circuits are stabilizing, one of the outputs may get biased on and present twice the normal operating voltage across the other half of the of the output stage. The safe operating area of that stage is exceded and results in a failure of the weakest device. This in turn usually takes out the weakest device in the the output stage that first turned on. It is possible that the devices are slowly degraded by this turnon process until punch through occurs.

I think that it is also possible that a hot amplifier could fail during shutdown/discharge of the power supply caps when circuit stability is lost, but that is probably less likely.

One of my responsibilities at my day job is teaching troubleshooting classes to our uninterruptible power supply clients. I expend great effort to drive home the fact that, with few exceptions, power semiconductors are shorted or good.
 
just to add another wonderful example of peavey mistery, i have a peavey cs400x poweramp, wonderful amp. i did a 12hour club gig with it, it was running practically on clipping all night, and did a great job, just got a bit warm. still worked all night. i un plugged it, took it home, and a week later, when i needed it again, i'd lost a channel, just never came on again. i still to this day haven't fixed it, i still use the one working channel. I have learn't that complecated mathematics can be put in place to explain such occurences... its Sods Law.

(Don't know who sod was tho!)
 
led_zeppelin said:
just to add another wonderful example of peavey mistery, i have a peavey cs400x poweramp, wonderful amp. i did a 12hour club gig with it, it was running practically on clipping all night, and did a great job, just got a bit warm. still worked all night. i un plugged it, took it home, and a week later, when i needed it again, i'd lost a channel, just never came on again. i still to this day haven't fixed it, i still use the one working channel. I have learn't that complecated mathematics can be put in place to explain such occurences... its Sods Law.

(Don't know who sod was tho!)

Please the next time you mention SODS law, please mention the guy who put it together.... Ahem... yours truly. 😀
 
Status
Not open for further replies.