Output transformer problem, or attenuator problem?

[TASBrendan]

Hey all,

I have had this Bandmaster RI come back to me after an initial repair, but can't see anything wrong. The initial repair was when the owner had accidentally plugged in his attenuator wrong and had the amp on with no load. One of the power tubes arced in the socked and the 100 ohm heater resistors had caught fire. New tubes, new socket, new resistors and everything is working as it should. I did a quick ohm meter check of the output transformer for shorts and read primary CT to each half at 130 ohm and 103 ohm, total ~234 ohms across primary, and no shorts to secondary. Did a quick low voltage test and measured ~4K:3 ohm, which all checks out. Speaker load is 2.6 ohms.

The amp came back because it blew a Mains fuse after about 10 min of playing with the volume at 7 (out of 12), and did it twice, but was fine below 7. One of the new tubes had a short, but everything else looked ok, so I sent it back out with a known good set of tubes. Now the story is that with the attenuator in place, the amp ran at 7 for about 15 min and then the volume dropped. He said the amp smelled hot, so he shut it off and let it cool off, then tried again with the same results.

Here's what I've tested now- and I suspect the attenuator - Ironman II Mini
The output RMS measured ~25Watts RMS, amp states 26W RMS
The output transformer still measures the same, with the ohm meter and running low AC through it without any shorts
I ran a tone through the amp into a 2.3 ohm dummy load on 7 and above for quite awhile without anything going out or burning. The output transformer got a little warm. The output voltage was steady at ~9VAC (RMS) at the load.
I ran a tone through the amp into the attenuator and into the 2.3 ohm load with the same ~9VAC at the speaker output. The attenuator started to get warm and the output transformer was a little warmer that before, but still no burning or power tube problems.
After some time of doing this, I did notice the output level jump up to ~15VAC and back down to 9VAC. I tried tapping the output transformer with no affect, but tapping the attenuator caused some voltage jumping so I shut it all down.

Am I missing something? I'm really reluctant to say it's the output transformer. The attenuator has some weird things about it, but I've never used one or had one around to measure. It states 8 ohms on the Amp input and Speaker output, but the Amp input reads different ohms depending on the setting- with the 2.3 ohm load plugged into the speaker output, the input reads 4 ohms at the lowest attenuation and goes up to 8 ohms at the next setting, 12 ohms, then 16 ohms, then 25 ohms on the highest attenuation.

So am I wrong in suspecting that the attenuator is varying the load on the amp and causing too much stress on the transformer and tubes? The attenuator has a reactive load, so is the amp is actually "seeing" what it needs?

The amp seems to get warm in a normal way under normal conditions and there is a residual smell from the resistors that burned.

The owner has said that the attenuator works fine with his other amp - a champ.

Thanks,

Brendan

[The Dude]

Have you checked bias or monitored output tube current? Cranking the amp would only exacerbate problems if tubes were already overbiased. I've had a couple of these amps with leaky PI plate to output tube grid coupling caps, but checking bias would tell you if you have that problem.

[Helmholtz]

The attenuator has some weird things about it, but I've never used one or had one around to measure. It states 8 ohms on the Amp input and Speaker output, but the Amp input reads different ohms depending on the setting- with the 2.3 ohm load plugged into the speaker output, the input reads 4 ohms at the lowest attenuation and goes up to 8 ohms at the next setting, 12 ohms, then 16 ohms, then 25 ohms on the highest attenuation.

So am I wrong in suspecting that the attenuator is varying the load on the amp and causing too much stress on the transformer and tubes? The attenuator has a reactive load, so is the amp is actually "seeing" what it needs?

Agree that this 8 Ohm attenuator is not suitable.
If the amp is designed for a 2.6 Ohm load anything higher than twice that impedance can damage the tubes (due to excessive screen dissipation).
Even worse, high load impedance may give rise to oscillation, resulting in very high (kV) peak voltages at the plates and the OT primary.

[TASBrendan]

Thanks to both of you for the responses!

First, the bias is good. The coupling caps had been replaced by a previous tech. I had biased the amp to 40mA before checking RMS output as per the schematic, and it held steady.

Second, good to know about the possible oscillation! The weird thing about these amps is that the screens and the CT are fed from the same HT after the choke, with no screen resistors, so the screens run about 10VDC higher than the plates. Fender changed it to their usual setup in the later versions, but this Tweed RI sticks to that early schematic.

I think the owner has had this amp and used the attenuator for awhile, but I'm not sure how long.

Thanks,

Brendan