Some of the regular readers may recall I had an RS35M
(bought new in 2002) blow its diodes in July 2006.
These were two 35A, 50V bridges made by Diotec. Astron
only uses the positive half and wires them in
parallel. I replaced these with Vishay GBPC35
rectifiers, tested the supply, and put it aside.

A friend loaned me his old RS35M which got the
repeater back on the air in 90 minutes.

I bought a brand new RS35M supply in July 2006 and
finally got around to installing it in October 2006.

The other day the same thing happened. The repeater
had been quiet all morning, someone called me on it,
and mid-way through the CW ID (after transmitting for
about 10 seconds), it just went off the air. I grabbed
the repaired RS35M, went to the site, and put it into
the repeater, pulling the other one back down to my
shack. The repeater was back on the air after 3 hours
(hey, it was cold outside and I didn't feel like
driving up there right away).

Back on the bench, I tried a new 8A fuse; it blew
immediately. I measured the resistance across the
diodes; I read 0 ohms, but without disconnecting them
from the transformer, this is not an accurate reading.
I'd surmise that one or both bridges have a shorted
diode in them. Seems like the same failure as the 2002
supply had. This one, however, is still under
warranty. I'm not sure it will be economically
feasible to ship it to Astron where they'll put in
exactly the same diodes, in the same configuration,
where they'll just blow again. Until I hear back from
them, I'm not going in there to disconnect the diodes
to actually measure them.

I've purchased some new 50A 1000V bridges which I will
put into these supplies from now on. No more "diodes
in parallel".

Incase you ask, the load on the supply is 25 amps at
14.0 volts, it has plenty of forced air cooling on it,
the environment is 70F, and the repeater is low usage:
a couple of hours per day with a sustained usage at
drive time of about 45 minutes on a busy day.

I was taught that running semiconductors (diodes and
transistors) in parallel without some kind of load
balancing components is a bad thing. I'd rather have a
single pair of diodes in these supplies instead of the
pairs of bridges. Anyone else care to comment pro or
con?

Needless-to-say, my next power supply will be a
different brand.

Bob M.


 
____________________________________________________________________________________
Do you Yahoo!?
Everyone is raving about the all-new Yahoo! Mail beta.
http://new.mail.yahoo.com

Reply via email to