Dentron GLA-1000 Plate Voltage Drop - HV caps?

[KD0PCJ]

I acquired a nice GLA-1000 (orig. model) three months ago, was getting good 1100v plate current; replaced the tubes and enjoyed ~450W output with 100W drive.  A drop in plate voltage to ~800-850v during idle was noted recently and a drop to 300W output.  Spare tubes yielded no difference, so ruled tube issues out.  I'm thinking HV filter caps, but am hoping for an experienced second opinion.  Thanks for any advice offered. 
K0IVY

[W1QJ]

It is possible a cap could be leaky or something.  Could be something else, but since there are only 3 or 4 caps in the amp why not change them anyway?  They are old and can stand being changed for good measure.  I recommend to change them for good measure and see what happens.  Cheap and easy to do.  Retrofitting a nwere cap should be a cinch.  MAYBE a 10 minute job.

[K8AXW]

QJ is right.... but you need to check the meter dropping resistor(s)....

Did you check the HV with a meter and HV probe or are you just going by the panel meter?

[KD0PCJ]

I am going by the panel meter, and of course an external watt meter (Drake W-4) to note the drop in RF output. 

[K7KBN]

If you're using high voltage equipment, you need a high voltage-rated meter (something like the 630-NA by Triplett, with good test leads would be ideal).  You need to be able to find out exactly what the HV is compared to what the manufacturer says it's supposed to be.

[KA5N]

Trying to get a sweep tube amp up and working and THEN trying to keep it working
is a demanding task.  Tubes are nearing unobtainium status and if found will be
expensive.  Also since the plate voltage is low (so as to not exceed the ratings of the
sweep tubes) you can't very well change to real transmitter tubes.
If you get it working correctly my advice is to sell it quick and invest in something else.
Even the venerable SB-200 is a much better bet for a real working amp that you can
still get tubes for and is easy to keep working.
Of course my advise doesn't cost anything and you may think it is worth what it cost.
Good Luck
Allen KA5N

[WX7G]

A piece of evidence is missing otherwise we could easily troubleshoot such a simple circuit.

If leaky caps drew enough current to pull the HV down that far they would explode in seconds.

The fact that the RF output power dropped as the HV dropped rules out the HV metering circuit being out of tolerance. This indicates the HV metering circuit is correct.

A question: Did the RF output drop coincide exactly with the drop in measured HV?

[W1QJ]

A piece of evidence is missing otherwise we could easily troubleshoot such a simple circuit.

If leaky caps drew enough current to pull the HV down that far they would explode in seconds.

The fact that the RF output power dropped as the HV dropped rules out the HV metering circuit being out of tolerance. This indicates the HV metering circuit is correct.

A question: Did the RF output drop coincide exactly with the drop in measured HV?

You are right about the meter reading.  I'd say your meter is reading correctly since it was reading 1100 volts and the power output was good.  Then, at the same time, the HV dropped and the power dropped.  If the meter reading was off, the power would still be there.  You are wrong about the caps though.  I have had several Dentron amps that had bad caps, but the HV reading was correct.  Then when you wnet to put a load on the amp, the HV would drop in half and the power would go way down.  Changed the caps (all fo them) and the power came back.  I would change the caps first thing.  It is a 10min job.  Then go from there.

[WX7G]

That's good information. So it's not that the caps are loading down the transformer but that they are not peak filtering. The HV meter is reading the average voltage. Well it looks like new caps are in order.

[W1QJ]

That's good information. So it's not that the caps are loading down the transformer but that they are not peak filtering. The HV meter is reading the average voltage. Well it looks like new caps are in order.

GOOD....do let us now if all is OK after you change them.  if not then we can look elsewhere, but I've had 2 Dentrons that had weak caps and no sonner a load was put on the PS the HV sang like a stone and the power went way down.  No bangs, no pops, no smoke, just no power.

[G3RZP]

Is it a voltage multiplier HV supply? If so, then it could well be high ESR caps without them going bang.

[KD0PCJ]

Many thanks again for the wisdom, gents.  Am on-line searching for new caps as we speak, and will replace R-4, 5 & 6 while I'm at it.  Reference dumping this amp...probably not at this point. I have cheaply accumulated eight spare 6LQ6/6JE6 and a few 6MJ6's to play with over the months...shared with my Swan 270.  I knew going in to this that a sweep tube amp was less than ideal and would be a challenge to maintain (that's the fun of amateur radio, eh?). 

[W8JX]

If you limit output to 350 to 400 watts the sweep tubes should last a long time

[K4DPK]

Since the output power and the plate voltage dropped at the same time, it is illogical to suggest they're not related.  The capacitors' electrolyte has begun drying out, and even in a full-wave rectifier system, since it is capacitor-input, voltage is dependent on capacitance.

Change them.

Phil C. Sr.
k4dpk

[WX7G]

The 6KG6 sweep tube is still made by Svetlana and with a change in socket wiring might be a good replacement for the 6LQ6 in the Dentron GLA-1000. The 6KG6 is rated for 35 watts plate dissipation and the 6LQ6 is rated for 30 watts.

The 1 kW CW rating of this amp does seem a bit high. At 65% efficiency and running OOK CW (50% duty cycle) the plate dissipation is 175 watts yet the tubes are rated for 120 watts. The forced air cooling does increase the plate dissipation rating but by how much I don't know. Given the previously low cost of sweep tubes and the limited transmit time in amateur service this might have been a good deal when the Dentron was made.