KMD XV100SD Low Output and Tubes Red Plating

[SteveJ]

Hi all,

Hoping I could get some help on this one. The amp initially had a blown mains fuse. I replaced a diode in the rectifier and upon further testing found that one of the power tubes had shorted (V6 or V9, can't remember). I replaced all tubes with JJ 6550's. I set the initial bias at 35 ma. Problems are very low output (10% maybe) and the tubes start to red plate, V6 and V9 for sure, I think V7 and V8 are as well but I shut it down before I could be sure. They are fine at idle, but when I run a signal or guitar into it and turn up the volume it starts. The bias current is around 105 ma on V8 and V9 when it is turned up, grid voltage stays around -55v and plate voltage is @ 480V.

I cut loose the protection diodes (D9, D10) and caps. Screen and grid resistors check and look good. I replace the bias cap and check the circuit earlier during troubleshooting. The PI circuit seems okay, swapped tube, 100k and 82k resistors as well as the others in the PI test okay. Voltages in the PI are inline with the schematic. Large filter caps checked okay for value and esr. I spot checked some of the smaller caps best I could without pulling the preamp board. I did pull the power board and checked all components.

I retentioned the power tube sockets and re-flowed the solder joints. Ran it with the two outside tubes pulled and also with another pair of tubes to see if that corrected the low volume issue, no change. I don't think its the O.T., could be wrong though. Can't figure out what I'm missing. I appreciate any help

Thanks,

Steve

[g1]

Make sure there are no shorts at speaker jacks, plugs, cables or speaker.
Maybe try wiring a known good speaker straight to the OT, with the stock wiring disconnected.

[SteveJ]

Good suggestion, I'll check it out. -Thanks

[SteveJ]

Okay, cut loose all of the OT wires and hooked direct to a speaker, no change in volume output. V6 starts to red plate faster than the others but all tubes are red plating.

[g1]

You said the bias voltage was holding steady when the red-plating was happening?
Try a pair of power tubes excluding V6.
Maybe try the neon bulb tester from geofex
http://www.geofex.com/FX_images/xform_test.gif

[SteveJ]

Correct, the BIAS voltage stays steady at -55v. I have an extra pair of tubes that I swapped out trying to eliminate a tube issue, I pulled the quad of JJ's and installed them in the middle (V7,V8). I may have got impatient here, it did nothing for the loss of volume, but I did not watch for red plating. I'll check that. The shorted winding test looks worthy, I work on that as well. Thanks, g1! On another note, why is it when you try to fix and old friends amp, it always turns out to be a dog?

[nickb]

Hi all,
...
The bias current is around 105 ma on V8 and V9 when it is turned up, grid voltage stays around -55v and plate voltage is @ 480V.
...

Really? The current goes to 105mA on one side with no signal? All you did was turn up the gain? That ain't right.

[SteveJ]

Really? The current goes to 105mA on one side with no signal? All you did was turn up the gain? That ain't right.

Yeah, I really wasn't sure about that. I'm using a Bias King and that is the reading when a signal is injected and the volume is turned up high. I'm running 1k in from a signal generator.

[nickb]

Yeah, I really wasn't sure about that. I'm using a Bias King and that is the reading when a signal is injected and the volume is turned up high. I'm running 1k in from a signal generator.

OK. So this was with a signal. Bias must be measured with no signal, by definition.

What you measured was the average current with a signal. So what is odd is that V6 and V7 aren't doing the same.

Anyway, I didn't mean to butt in, it's just that that jumped out at me.

[SteveJ]

Thanks for the input Nick. The bias is set at 35ma at idle. The bias voltage stays steady at -55v under all conditions. I know that is pretty cold, but I was starting there when I turned it on and noticed the output was low.

[Enzo]

Bias is like idle speed on a car engine. Bias is by definition an idle measurement. As soon as you run signal through the amp, you can not measure bias, you are now measuring tube current. Just as you cannot measure idle speed on your car tach while driving down the road.

[g1]

I'm a little confused here about the red-plating. Does it happen when idling, or only when signal is applied.
If it is only with signal, and there is low power output, are you monitoring with a scope? Can you see any oscillation?
Otherwise, I still think you should try the neon tester.

[SteveJ]

It only red plates when a signal is applied and the volume is turned up. That and very low output are the symptoms. I will try the neon tester, I just got in from work and will try to get back on it this evening. I also put the scope on it.

[SteveJ]

Bias is like idle speed on a car engine. Bias is by definition an idle measurement. As soon as you run signal through the amp, you can not measure bias, you are now measuring tube current. Just as you cannot measure idle speed on your car tach while driving down the road.

Thanks, Enzo. I do only measure bias at idle. When I quoted the measurement in the first post, I was just trying to note the current draw of the tubes when the volume was turned up. Didn't know whether or not that information would be useful. Apologies to all if it was confusing.

[SteveJ]

Finally wrapped this one up tonight. Turns out the output transformer was bad. I dropped in a Classic Tone 100W Marshall replacement. Had to enlarge the mounting holes just a little with the Dremel. New tubes, bias cap, resistor and a diode in the rectifier. Thank you all for the help.