Wednesday 3rd Dec.1998
Hello all
In the last 20 years I've constructed 9 homebrew amps.-mostly
HF, some 6m,2m. I keep a record of everything that blows up,
and try to analyse why. You might be interested in some of
these experiences :-
Cooling - most commercial amps. seem to have poor cooling of
components - maybe the tube is cooled fairly well, but the
P.S.U. components are sadly neglected. In a recent amp. I
have 3 separate cooling fans in the P.S.U. - OK so the shack
sounds like the flight-deck of a B52 but reliability is
increased considerably ! Also, I discovered the hard way
that if the main blower/fan which cools the tube stops if
the main fuses blow/circuit breaker pops, then the core temp.
of a ceramic-metal tube can rise so much in a few seconds that
the tube is destroyed, even if the original fault is not with
the tube. All the blowers here operate on separate AC supplies
with their own fusing. (+air-flow sw.in amps)
Transformers - had several HV xfmrs go down, despite fusing
and cooling. No real pattern to this, just luck, maybe.
(One was marked 'Admiralty 1944' so it's hardly surprising!)
Had one or two HV xfmrs. with leakage (15-20Mohms) from sec.
to frame - this disappeared after several months of use, but
this is potentially a serious safety hazard.
Tubes (Valves) - 4CX1000A/4CX1500B series seem to fail C-G1
short - got several duds with this problem. Again no real
pattern/cause, although some flashovers suspected.
Feedthru caps. - watch the voltage rating - some so-called
500V feedthrus can't handle DC + RF and fail short-circuit.
Several time-consuming faults (during CQ WW of course)
because of feedthrus.
Connectors - I use 'C' type plugs/skts for HV up to 3.2kV
- only one failure so far - a surplus skt. which fried.
'N' type cable-mounted sockets do not like much RF.
Tube sockets - had a 4CX1000A socket fail - blamed the
manufacturer (!) - the silicone sealant on the screen
decoupling cap. had penetrated between the layers of
mica insulation, causing 'hot-spots' which eventually
went short-circuit.
Relays - watch the timing - burnt-out RF input relay
because of poor sequencing of T/R relays.
Roller-coaster inductors - removed from all amps.!
Problems with corrosion followed by arcing. RACAL make
some very nice roller inductors with 2 separate silver-plated
contacts, which press firmly on the coil - even these
proved unreliable in the long-run.
73
Chris GM3WOJ/GM8V
<chris.tran@zetnet.co.uk>
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|