[Amps] Amplifier frustrations

Tue Aug 1 17:49:42 EDT 2006

That is some story, Peter.  I think all good experimenting hams have had a
similar experience.  Thanks for sharing.

Reminds me of a particular antenna (center loaded vertical dipole) I was
trying to tune last year in preparation for DXpedition to 9H (Malta).  After
many hours of frustration going back and forth from shack to antenna to
locate the source of intermittent very high SWR, the ultimate find was one
of the loading coils was touching a bolt protruding through the mast.  Of
course as the wind blew, the mast would flex a bit and the coil would
alternately short and open.  A simple bend of the coil around the bolt cured
it.

As for the DVM, I am not surprised to hear of faulty readings on a low
battery.  It has happened to me more than once.  And no, there was no
indication of low battery other than bad readings.

73, Keith NM5G

-----Original Message-----
From: amps-bounces at contesting.com [mailto:amps-bounces at contesting.com] On
Behalf Of Peter Chadwick
Sent: Tuesday, August 01, 2006 3:27 PM
To: amps at contesting.com
Subject: [Amps] Amplifier frustrations

A tale that may amuse, and may even evoke sympathy........
Last year, I changed antennas. The 205BA at 62 feet and interlaced 4 ele
beams for 10 and 15 at 68 feet went, replaced by a 4 ele Steppir at 62 feet.
The tower, fed as folded monopole, moved its resonance from 2.6 to 3.72MHz.
No longer could it be fed with an L network with a variable C - it needed
variable L as well. Built a remote controlled tuner, using a vacuum variable
that I'd been given. Hi Pot? never needed one.........beautiful purple glow
in the variable. Anybody know where I can get a new vacuum for it? SWR went
through the roof, of course. Tried a smaller vac variable, and tried making
a padder capacitor out of sheet glass. More fireworks, and SWR again through
the roof. Tried some HV ceramic capcitors and they couldn't cut it either -
more fireworks. So I bought a cap from Max Gain systems, and eventually got
the mechanics sorted out for the motor drive.
I then found that the periods of arcing in the tuner appeared to have
screwed the amp. It would work on 3.5 MHz, but even into dummy load, went
mad at 3.7.  I changed the passive grid resistor to a Tin-Oxide, and still
it went mad. Plate current runs away, and I think the bias from the
regulated bias supply collapses, because the interlock drops out. OK on 20.
The RF choke was definitely small, and I played with that, with two chokes
in series. One of them was from the NCL2000, and being held off by about 85
microhenries from the full RF at the plate, I was surprised when that had
the smallest 'pi' start burning up on 40 - from the outside. That took the
series HV fuse. Stripped burnt wire off, intennding to have another go.
KH8SI appears, and I've no amp. This morning, he's actually gotten around to
working EU 30 minutes after the band opened on 20, but I can't cut it
without an amp - I did try.... Try firing up the amp after the work on teh
RC choke. Before the amp was put  into tx, there was a sparking noise. Open
everything up, and check. HV fuse OK, remember Tom Rauch's comments on
equalising resistors and Rich Measures comments on wirewounds. Measure them.
There's 8 off 22k and 2 off15k in series across the elctrolytic string, and
they measure about 900k ohm. Figure this needs looking into. Go to one
capacitor with 22k across it. Put meter across, and watch the resistance
climb as the cap charges. gets up to about 30 K. OK. Disconnect capacitor
from resistor, check resistor again. 22.1K. Add capacitor and resistance
goes UP!!
Repeat a few more times, wondering 'what the hell?' Why should 800 mFd
across a 22k resistor make the DVM indicate a higher than 22k resistance.
Get out the AVO - old fashioned VOM - and that gives sensible readings. Then
note DVM is intermittently flashing to say the battery is low. Change
battery, and sensible readings obtained. But that is a new one on me that
the DVM can be that wrong before it indicates low battery. Oh well, I got it
for nothing anyway...Probably something to do with spikes from the sampling
integrating.
Go back to chasing the sparks. Of course, this is where you need a Hi Pot.
Havn't got one. there's a HV xfmr under the ebnch supposed to be for a scop.
Check it, get 1700 volts. two K2AW rectifiers, a couple of caps, drive it
from a Variac and we have a hiPot going up to 4.4kV. Now I can use that 40kV
probe for the DVM.....apply to amp, sparking. Turns out that there's an open
on the RF choke that was sparking. Sort that out.
Try firing up on 160, efficiency poor, and suddenly the mains fuse for the
HV goes.. No spares.....drill ends of fuse and fit 30 gauge copper wire. NOT
best practice with the inside of the glass copper coated. Decide to try
20meters in case KH8SI is there again tomorrow morning -  they've probably
gone home, or conditions will be bad or something. Still, it seems to fire
up on 20 tonight. Leave it alone....
Conclusions. This nearly 50 year old amp, the Rf deck of which was designed
for the military as a lower power tx and uprated with bigger valves and more
volts by the OEM, has probably got some components (for which I don't have,
and can't readily get, spares) breaking down under RF load as a result of
the SWR when the arcing took place They are hard to pinpoint because when
they let go, all hell lets loose and the thing blows fuses, and at lower
power, theyr'e OK..
I've even got so depressed I've been looking up the prices of
Ameritrons......about the same number of pounds here as they are dollars in
the US. That is even more depressing....
The DVM thing is something totally new to me. Anyone else seen anything like
it? Hopefully, someone may learn soemthing from this tale of woe....maybe
that buying an Ameritron is a better idea than trying to build your own or
rework surplus!
73
Peter G3RZP
_______________________________________________
Amps mailing list
Amps at contesting.com
http://lists.contesting.com/mailman/listinfo/amps