[Amps] Centurion amp problem

[Georgens, Tom]

I have come across an interesting problem with my Centurion amp ( 2 X
3-500Z).  The short version is that it seems that only one of the tubes
are operating, and I changed to known good tubes and the problem
persists.

 

The longer version is as follows.

 

The amp has been an integral part of my contesting station in Barbados
and has been a generally good performer, but it suffers in the hot,
humid and corrosive environment like all the rest of my gear.  The
screws in my new K3 have rusted in just 4 weeks.

 

I have been experimenting with computer controlled wattmeters as way to
provide more automation and protection when I found that new wattmeter
was giving dramatically lower power readings than the one I had (600 vs.
900).   I have a separate wattmeter for my other amp.  The new wattmeter
also seemed to be detecting more variability of the power measurement
with each sample.   Under continuous key down carrier, samples could
vary by 100 W.   The same wattmeter on an AL1200 had far less
volatility.

 

The wattmeter in the centurion, matched the old wattmeter, but the Plate
Current meter  barely crossed 400 mA.  I was never sure what meter to
believe but I had two meters supporting the belief in the higher number.
Until now.

 

Not sure what to do, I had another dual 3-500Z amp (TL922) as a spare.
I tried it in place of the Centurion and it measured what I would have
expected on the new wattmeter (about 1KW out) and much higher on the old
one.  Plate current was a bit over 600 mA in the SSB position.  I used
the TL922 during the CQWW this weekend and it worked great.

 

Reasoning that it could be the tubes being old, I decided to replace the
tubes in the Centurion with the ones in the TL922, but the problem on
the Centurion did not go away.  Testing the TL922 with the original
Centurion tubes would have been the best next step but I was out of
time.  Before I left to return home, I decided to put the tubes back in
the original amps.  The Centurion had not cooled down yet when I tried
to remove the tubes.  One of the tubes was still very hot at the anode
connection but the other was barely warm.  My conclusion was that the
amp has been running primarily, or completely, on one tube.  I certainly
looked like both tubes were getting filament voltage since they both
seemed to be lit up.  It is possible that one was illuminating the
other, but I doubt it.

 

It could be that one tube in each pair of is bad but, in the case of the
second set, one of the tubes would have to have failed in the 16 hours
between flawless completing a 48 hour contest, and being tested in the
Centurion.

 

I had to leave the amp behind but I am open to suggestions on spare
parts to bring down to attempt a repair.  I did open the chamber below
the tubes to look for obvious problems and nothing unusual was seen.
The amp has a lot of corrosion, but not in the area below the tubes and
the tubes sockets looked great.  All connections looked fine.   The
grids seemed solidly grounded to the chassis.  The plate voltage
connections looked fine as well.  There are .01 caps on the schematic
across the cathode pins on each tube that are not in the actual amp.
There are also .01 caps in the rf input circuit to each side of the
cathode that were still in place.  

 

There does not appear to be much that can go wrong.  The  only active
circuitry is the bias circuit but I am not sure how it can impact just
one tube.

 

I know for sure that the tubes from the TL922 were not matched and I
have no idea about the ones in the Centurion.

 

Not sure what other information would be helpful and I am  open to
suggestions

 

73, Tom W2SC 8P5A

 

Other than that,