"Noam" wrote in message
oups.com...
I checked the big transistor that was on the heat sink and it looks
fine. I took it out to test it, and also I checked in-circuit and
there are 130V across B-C and E-C while running (so I guess it's an
npn). There is also a capacitor with 20V on it and a capacitor with
130V on it.
I don't quite understand the stuff about the missing variable voltage.
How would I check for this?


Arfa Daily wrote:
"Noam" wrote in message
oups.com...
Hi,

I have an old 0-100V (0.2A) HP power supply (HP 6116A), which is
broken. The output is always 0.6V (when on). Nothing is obviously
burned or exploded inside.

The problem started when one of the people in my lab connected a
capacitor backwards across the supply's output, giving it a large
negative voltage spike.

Any ideas on what could be the problem? The supply has a transformer
and several largeish capacitors. However, the inner circuit is pretty
complicated. My guess is something near the output is broken, but what
could it be?

Any advice would be greatly appreciated.


Best guess would have to be the series pass transistor ( probably fairly
large and on a heatsink ).

Confirm with a voltmeter that there is a highish voltage on its input -
probably 120v or so and may be on the collector or emitter depending on
whether it's an NPN or a PNP type. If input volts are present, check for
a
voltage on the base connection, which varies up to around 100v with
rotation
of the set voltage pot.

If there is no input voltage, problem is back in the raw stage ie
rectifier
/ smoothing / maybe surge limiter resistor.

If the variable voltage is missing, disconnect, base lead and check
again.
If it's still missing, problem is back in control loop. If adjustable
voltage is present, and input voltage is present, check to make sure that
there are no shorts across the output of the transistor ( decoupling cap,
protection diode, overvolt zener ) then replace transistor.

Obviously, this is a fairly simplistic approach, and assumes that it is a
'
conventional ' design linear supply, but should catch most common
problems
for this type of fault. If problem does prove to be in the control loop,
then things will get rather more complicated, but suspicion should fall
first on any semiconductor devices ( transistors, control chip, diodes,
zeners ).

Arfa

Assuming then that it is NPN, if you have 130v on the collector, you would
expect the emitter to be the output. To check the control voltage, you need
to measure volts between ground and the transistor base, whilst turning the
set volts pot up and down. You should see a variable voltage between 0 and
100v, and the emitter should follow the base, less around 0.7v.

But before doing this, switch your meter to ohms, and just check the
resistance between the emitter and ground. There just might be a short
across the output from a diode, as I, and another poster suggested. If you
don't see a short, go ahead and check the base volts, as detailed, and let
us know.

BTW, try not to top-post, as it makes the thread difficult to follow for
people ;-)

Arfa