JW wrote in message
...
Here's one I haven't seen before. Bought an EDC (now Krohn-hite) 521 DC
calibrator on the surplus market that was blowing fuses. There were two
power supplies that seemed to be causing the problem +-150V. Turns out
that there were two 22K 2W carbon comp resistors that are used as bleeders
across the two 470uF filter caps. Both had changed resistance - one was
about 2K, the other was 150 ohms! I've seen carbon comps drift, but never
saw ones that drifted *that* far. Also, in circuits like this where they
are dissipating over a watt, they usually drift high; at least in my
experience...

Well, after replacement it seems to be working well. May need a
calibration though.


If you still have them , try a neat 0.5mm thick Dremmelgrinding disc cut
across the middle and see if there is a gradation of resistance developed
across the material , and so along the length. Just under the surface
coating you may find the high conductivity path.
I keep a "black museum" of such oddities , don't know if anyone else does