Power tube arcing

[ChubeRules]

I just received a 64ish Vibroverb for repair for blowing fuses. First thing, I pulled the power tubes and found one of them was arcing between the plate and heater pins, leaving burns on both the tube and the socket, and only on one of the tubes.  The tube was also shorted interally.
 
I removed the chassis for further inspection and found the 100 ohm heater resistors fried, which I replaced. I also noted the 1w 470 ohm screen resistor browning on the tube that wasn't shorted and no brown of the arcing tube.  I replaced the screen resistors though they still measured true.
With the power tubes removed I checked voltages all around and found nothing unusual.  I then replaced the power tube sockets.  And rechecked voltages.  Also checked the Spague Atom filter caps that look and test as new.


I replaced the bad tube and powered up with a dim bulb limiter.  No signs of arcing.  So I appied full power, still no signs of overheating, checking with an infrared thermometer.
 
So I plugged in and the amp sounded good.  I rechecked voltages, all good.  I left the amp on for hours and rechecked, and still sounded good, although I never drove the amp very hard. So I returned the amp to the owner. He said he played the amp for about 20 minutes before the fuse popped again.  He brought the amp back and I repeated all of the above.


I found the same (now new) power tube and socket had arced again.  But this time the tube didn't short interally.  I replaced the 100 ohm resistors that had fried again, and tubes, and power up gradually again.
All was well.  So we tested the amp again and drove it hard this time.  All smiles.


I just got the amp back again.  He played it shortly and popped the fuse and fried the resistors as well as shorting the tube internally.
Since the problem reoccurs at the same tube I'm ready to replace the output trans.  I checked the OT resistance and got an overall reading of 162ish ohms and 75 ohms off one leg and 85 ohms off the other.


Any thoughts or suggestions much appreciated before I buy a new OT

[shooter]

tube and socket had arced

did you clean/replace socket?  with tube out, power off, is there ohms from socket plate pin to anywhere else?  If so, fix that.
If not, with power tubes out, what is the bias voltage?

[ChubeRules]

Yes, I replaced both power tube sockets with new ones.  With tubes out, no continuity from plate socket pin to anywhere except the OT center tap. The bias voltage is -51.7 vdc

[sluckey]

Since this only happens to your customer you may want to question him about his playing style and how hard he drives his amp. Ask specifically if he usually connects to an external speaker cab.

Try this simple transformer test but replacing the OT is probably the next step. And if it turns out that the OT is not the problem, well, if you're running a repair shop you need one of these on hand anyway.

[davidwpack]

A few years ago, I fried tubes in an amp and realized that the processor I plugged in was a preamp/processor.

[ChubeRules]

I was called out of town for a week and just got back to try Stuckey's trans test for shorts using a neon bulb and a battery.  Bad news, it tested good.  No shorts.
 
I did vet the user for playing style etc. and he's says he never runs the amp past 3 1/2 on the volume knob.


So, next, I'll be swapping in a new OT, anyway.  Hmmm.

[shooter]

I worked on an Ampeg that did the same thing, after tubes n OT, I found the grid caps were the problem, causing big DC spikes n weirdness during play, idle was fine.