"Dave" wrote in message
...
Could a transistor going flaky and overdriving capacitors cause them to
swell? How about causing one of them to explode?

Just wondering what caused the systemic failure of the power supply PCB I
am looking at.

Thanks,

Dave



For some strange reason, switchmode power supply designers have a
predisposition towards putting any and all critical electrolytics, as close
as possible to any sources of heat that are also on the board.

This has the effect that from day one of the power supply's service life,
its electros are being baked by external heat radiation from these sources.
Now add to this that the caps that are fitted are often only 80 degree
types, and are not of the best quality low ESR type specially specced for
switchers, and you now have self-heating to add to the equation, when they
start trying to handle the large high frequency ripple currents that many of
them will be required to do, as part of their job. Now compound the
situation even more by putting the supply in a unit tht's designed for its
external appearance, rather than airflow around its inside, and finally put
it in a glass fronted cabinet with poor ventilation in your lounge, and you
can see why the electros will start to fail and eventually bulge.

Depending on where in the circuit the cap is, the failure may cause anything
from a simple supply shutdown, to a catastrophic cascade failure, involving
semiconductors and resistors on the supply itself and, sometimes, serious
failures of LSIs located elsewhere in the equipment, due to either the power
supply's regulation going out of the window, or huge high frequency ripple
voltages on the rails, that these devices just can't handle.

These days, this type of failure is the 'bread and butter' work of most
independent service engineers, so this mechanism of failure (which is not
confined just to power supplies either) is very well understood by us all.

Arfa