[BC] Dud 4CX20,000 and the FM25K?

[Kyle Magrill]

If I were trying to keep an FM25k as a main, then definately, 
replacing that aweful IPA would be a top priority.  As an emergency 
or backup rig, I'm not sure I would replace the FM25K IPA amp.    
It is true that they are fragile, especially to lightning damage.  
On the other hand, it's not like they blow up every day.  I 
serviced three or four installations using the FM25k, one of 
which was combined to an FM50k.  We used to blow up a 
module every 3 weeks in summer, but we would often go 
all winter and spring without a problem.  At one site, we had 
the power company come in and drive 100' deep ground rods 
in and sink a well resulting in the failure rate going from 4 or 
5 modules per year to one or two.  

If it were my decision, I would weigh issues like: 

is the transmitter's IPA stage is still functional, 

can suitable replacement RF transistors can still be found
(if there are no available substitutes, then you really do 
have to replace the IPA),

What kind of power do you need from this rig,

What is the cost of replacing the IPA ($5k+ might be more than it's worth)

Over the years I noticed some things about the 25K.  
First, the IPAs are far more prone to failure when running at maximum 
output.  If you run them at 60%, they will run for years, if the rest of 
the transmitter is properly operating.  If there are other problems, 
especially in the PA, those IPAs will fail, so use that as an indicator of 
transmitter health. 

Second, the IPAs don't all blow up at once.  Usually only one will blow 
up resulting in a slight power reduction and it is usually possible to 
compensate with the others, especially if you don't need 27.5kW 
from the rig.  

Third, the PA efficiency should be around 80% at 20kW+ of output.  
If the efficiency is much off of 80%, look for a problem.  I noticed one 
slowely loosing efficiency in a combined pair.  The tubes were only a year 
old, so I started digging in.  After an all nighter, I finally found an inductor 
shown on the schematic that I couldn't locate inside the box.  I had 
checked everything else that made any sense so I wanted to find this 
inductor.  The notation on the schematic was P.O. L8.  I went nuts looking 
for it and, after several hours, gave up and finally called Harris at 4AM (which
I hated to do as a matter of professional pride).  Unfortunately, the on call guy 
at Harris that night was an AM engineer and he was of no help at first (though
it made be feel better that he couldn't find the darned L8 either), but then 
he had a revalation.  He said Harris uses the nomenclature P.O. to mean 
"Part of" so this must be a distributed inductor.  So, I thanked him 
and went back to looking.  Sure enough, soldered to the underside 
of the PA socket were little resistor leads every 1/8 revolution.  They 
were in the right spot on the schematic, so that had to be them. 
I had just taken them for DC grounds, which they were, but they also 
were sized to be rf chokes.  4 or 5 of them had disintegrated over time.  
Clipping the leads off of some resistors and soldering them in place instantly 
returned the PA to 81% efficiency, just in time for morning drive.  
I never would have imagined that 8 little resistor leads could be so 
important in a circuit carrying that kind of power, but they were. The next 
weekend I checked my other transmitters and sure enough there were some
missing wires under every one of the PA sockets.

Fourth,  the transmitter can be tempermental about its tubes.  I once replaced 
an aging 8990 tube with a rebuilt spare and the tuning and efficiency were way off.
We had a second rebuilt spare which I put in with the same result.  I put the 
old tube (also a rebuilt) back in and everything was normal.  I sent the two 
rebuilts back and got two more which both worked perfectly.  No cause was
ever mentioned.

Fifth, verify that the step-start ladder in the HV power supply is good.
At least one that I saw had blown the resistors in half so the full plate 
current was applied at startup every time.  Mostly this works, but it's 
healthier to step start these boxes. 

The transmitter is pretty solid and can be reliable, even with the old IPA.  
Assuming you get it up and running and put back to factory spec,
I would have no concerns about the unit coming online every time 
the plate button is pushed.  As a standby or emergency rig, I'd expect to 
get at least another 20 or 30 years out of the box, provided you exercise it 
every couple of weeks.  For the record, I had no control problems on any of 
them that I serviced, although the auto-power function can be frustratingly 
unresponsive at times.  Fortunately, there's a switch to turn it off.