[Milsurplus] 1MC system and B+ problem

[WA5CAB at cs.com]

Good deal on the working unit.  On the non-working one, if you've only 
checked for leakage with an ohmmeter, you should run a megger test.  If you 
don't have a megger, and if the rectifier circuit is just the standard full wave 
with HV secondary center tap grounded, you could run the test using the HV 
transformer as the source.  Pull the 866's and disconnect everything from 
the filament windings.  Connect one of the HV leads through something like a 
10 Megohm 1/2 watt resistor to the filament winding.  If that blows the fuse 
or the 1/2 watt resistor, the filament winding is pretty definitely the 
culprit.  If you have or can borrow a set of solid state replacement rectifier 
stacks, with everything disconnected from the filament winding, connect them 
in the circuit and run a test.  If that works, then again the problem is 
probably the filament winding.

In a message dated 4/3/2010 5:20:57 AM Central Daylight Time, 
robandpj at earthlink.net writes: 
> I had the thrill yesterday of sounding the general alarm on the 
> WWII-vintage 1MC system aboard MASSACHUSETTS.  500W from 4 push-pull, parallel 838s. 
>  "This is just a test, DANG DANG DANG....."
> 
> Rather annoying problem with a second PA unit that we were trying to get 
> running again.  After isolating a leaky filter cap and everything downstream 
> from the rectifier, we were still blowing fuses when we put the B+ on.  We 
> could run the HV transformer alone, disconnected from the 866A rectifiers 
> or even with either side of the transformer connected, but with both sides 
> of the transformer connected to the rectifiers we got a lot of current as 
> indicated by bright purple flash in the rectifiers before the HV primary 
> fuse blew.
> 
> We suspect an arc in the rectifier filament transformer, which shows no 
> sign of leakage from the rectifier winding to the case or any of the other 
> windings under test with an ohmmmeter.  
> 
> The next step is to try replacing the filament transformer, which also 
> runs the 838s, with a seperate winding for each 838 filament.  Weighing in at 
> something like 50lb and located just above the deck, this is what we call 
> fun.
> 
> Anyone have any other suggestions?
> 

Robert  Downs - Houston
wa5cab dot com (Web Store)
MVPA 9480