JCM 900 squawk on power off

[Randall]

I've got an early 90's JCM900 Dual Reverb that makes an awful squawk at power off, regardless of the standby position. I don't know where to start with this. It was not the complaint it was brought in for, that has been resolved. The noise happens approx. 2 seconds after the power switch is turned off. It seems to be worse if the amp has been on for a bit, and less so if it is powered up and down relatively soon. Could this be something about the original filter caps?

Also, not to confuse things, but I notice while setting the bias on the new tubes, and watching it for a few minutes, every now and again the current would go a few mA high or low for just a few seconds, and then settle back to where it was, this is both tubes. Could this be related?

[Enzo]

Monitor your mains voltage with a second meter. See if that jumps around at the same time as your current.


Does it still squawk if you turn the reverb controls all down?

Perhaps you can share the model number? I don't know what a dual reverb is.

[Randall]

Sorry, it's a JCM900 model 4500.

I was monitoring the plate voltage, it was stable, wouldn't that be just as telling, or even more so?

It squawks with all the controls turned all down, sometimes louder than other times.

http://www.thetubestore.com/lib/thet...-Schematic.pdf

[Enzo]

Sure, voltage somewhere. if it is stable, then it is stable, and we look elsewherere.

Maybe plug something into teh FX rern to disconnect the preamp.

[Randall]

It makes no difference if there is a dummy plug into the FX Return jack, eliminating the preamp.

[Sowhat]

This is a good one, the amp burps 2 seconds after shutdown and your fundamental theory is some some sort delayed pulse caused by capacitive discharge from one of the 50/50/500,s on the chassis...like some sort of partial internal short thing? I could easily buy that one, I've had similar things happen on other equipment before... I think? I usually hunt around with a scope for weirdies like that and use my spidey sense. :-) really, I can usually see these things better on a scope than using my meters for transient things like this.

Sorry, I'm really no help on this, it's just one of those glitches you have to hunt down no matter how far down the rabbit hole you go. Don't you just love it?

[g1]

re: squawk
Have you had a look at how the +/- 15V rails are collapsing after turn off?
Does it still happen if something is plugged into FX return? (I wasn't clear if your response in post #5 was about the squawk or the idle current glitch)

[J M Fahey]

FWIW plugging into the FK looop return DOES NOT eliminate the preamp from the equation.

This is sensibly placed where signal levels match pedal levels and not the LAZY solution of ill fitting it between preamp and power amp (where signal is WAY too high)
Anybody looking at the schematic will see that we have 3 Op Amps and 2 triodes *after* Loop return and before power amp proper.

Ans I can confirm that the squeak/squeal is probably some Op Amp going crazy when voltage drops too much.
Hey *I* get dumb kooky when my sugar levels fall too low, so I am not surprised.

FWIW many amps do that, including famous Gallien Krueger and a couple SWR.

Typical solution is to add a power amp mute, activated at turn on and turn off, so it stops squeals and pops.

Not *stop* really, they are still there, just muted.

[Randall]

I am primarily asking about the squawk, and maybe is the sometimes idle current glitch related. The squawk happens almost every time, the idle current glitch only once in a while, and I have not seen it today.

The rails are collapsing without anything unusual that I see. The positive side gets to zero a little before the negative side, but not by much. Also I just noticed the squawk is louder and different in the clean channel than it is in the lead channel. The clean channel goes up in pitch, while the lead channel goes down in pitch. I've seen a youtube demo of another JCM900 doing this, and it was the less loud, down in pitch sound. I could live with it if it was just sound, but the louder clean channel up in pitch sound is alarming. It sounds like a sci-fi ray gun or old video game weapon.

[mozz]

Monitor the bias, tap on the filter caps with a wooden stick, if the bias jumps, change the caps.

[nickb]

....regardless of the standby position.

Are you saying that when turn it to standby then it squawks? If so that eliminates all the low voltage rail collapsing op-amp type stuff. Can you hear it on the Direct out?

The tube characteristics change as the HV rail collapses and it is likely you are just hitting a spot where gain and phase is just such that it goes unstable, assuming that otherwise the amp is 100% normal. That being the case you could wire the unused side of the standby switch to isolate the EL34 cathodes so giving you a mute and prolonging the life of the El34's to boot.

[Randall]

"Monitor the bias, tap on the filter caps with a wooden stick, if the bias jumps, change the caps."

I have already done this, makes no difference. I have not seen it do it today, it has run very stable actually. I have not been able to make it do it, so perhaps it was something to do with my bias checker socket, who knows?

As for the squawk, since no one has yet offered the filter caps as being the possible cause, I am going to let it go. It was not the complaint it was brought to me for. And I have discovered if the standby is switched off for 20 30 seconds, the squawk is much more subdued and almost barely noticeable.

[mozz]

This may work as a quick check to see if the filter caps are the problem. Lower the value of the bleeder resistor across the filters if there is one. if the squack is faster, the caps are bad. I have seen bad caps that tested good but when i hit them with a wooden stick i could see the ripple increase.

[Randall]

It squawks whether standby if off or on, if I switch off the power switch right away. If I wait until standby is switched off for 20-30 seconds it is much less of a squawk, or even none. It does not sound like a tube going unstable, it's a very quick and sharp sound, rising or lowering in pitch. The whole thing takes about one second. Doubt this is an unstable tube.

This was not asked to be repaired, and since I know the client won't want to spend for me to put extra work into it for something that I cannot give a definitive explanation for what needs to be "repaired", I not going to chase this any further.

[Enzo]

Oh.
I was going to suggest isolating the problem. Pull the phase inverter tube. Does it still do it? If that kills it, the issue is before that stage. We can ground the left end of C7 at the PI input. Kills it? Pull V2, kills it? We systematically go back the signal path, either interrupting it or grounding it (the same thing really)to determine the point the problem is before or after. If you are done with it, so be it, but that is how we do this. Systematically isolate the problem.

SOMETHING is unstable, that is what an oscillating amplifier is - unstable. I don't know why it can't be a tube. Instability isn't a defect in the tube, it is a defect in the CIRCUIT.