19 Oct 2003, 21:41, JIM HOLSONBACK (1:123/140), wrote to ALL:

Hi JIM.

 JH> Speaking of sparks and smoke - trying to microwave some "baked 
 JH> potatoes" this evening.  Saw sparks up on the wall of the thing, heard 
 JH> some accompanying "Poof" types of noises, and turned the thing off.

 JH> This is an old Sharp microwave- sticker on the back says made in 
 JH> SEP, 1980, so we've certainly gotten our moneysworth out of it.

 JH> Anyone know if it might be repairable, or is it on the way to 
 JH> joining Roy's  ATX power supply?

I like my old manual timer microwave with the temperature probe.  It heats
my coffee perfect every time!  When it dies [1981], I'll probably try
repairing it, as I haven't seen any like it since.

I did repair several over the years, not too much problem, most of the
older ones are pretty simple.  My Daughter's old one died this summer, but
having a rusted out base on the cabinet, I told her to just trash it.  She
got a cute little $49 one that suits her needs.

My brother-in-law gave her an old working Amana a year or two ago, and it
lasted two more weeks.  I looked in it, and there were circuit boards all
over the place, and parts I couldn't even recognize or identify, so we
junked that one also.

The old ones generally have a BIG XFmr, a BIG capacitor with a bleed
resistor across it, the magnetron, and a HV diode.  I don't think the XFmr
in those would ever die, wiring is monstrous!  [New machines use switching
PSs, no XFmr].  Never saw a capacitor go either.  One of my old Littons had
a dead diode, the other cooked a magnetron.  The diode [in 1974] cost me
$17, was a simple repair, the magnetron in the other cost me $400 [in
1975], and a LOT of conversation with the parts place, as I wasn't a
"licensed repairman" and he didn't want to sell to me.  Times
have changed though!

When the magnetron goes, it is most often a cathode short, the unit sucks
power like mad and dims the lights on older houses before blowing a fuse.  

Random showers of sparks flying around indicates an open diode, therefore
no DC and no oscillation.

As noted in Roy's reply, the diode is not 'testable' without a few tricks.
The easiest way is to simply connect a 9V battery in series with one of the
meter test leads on an old analog multimeter set to it's highest Ohms
scale.  The 9V is enough to overcome the reverse breakdown rating and get a
reading.  BE SURE to NOT try zeroing the meter while the 9V is connected! 
Do that first!   :-))

Goes without saying, don't trust that bleeder resistor to be functional,
check or discharge the capacitor first.  It can hold 2400V to 10KV, and
seems to have more 'punch' than a TV HV.  You NEED an HV probe to get
anywhere near the HV section, but I don't see any need to be concerned
about what the HV is.  It works, or it doesn't.

If you need a magnetron, "mat electronics" [believe they are in
PA], had a catalog of replacements for older units, ran from $24 to $95. 
Might be able to dig up an old catalog and see if they have any web
address.  

New units being so darned cheap and 'disposable', it hardly seems worth 
the effort, but it is still fun testing and trying to find the problem.


     Good luck...  M.

--- Msged/386 TE 06 (pre)