Peavey JSX Head, Intermittent Crackle

[AtomicMassUnit]

This amp has an intermittent crackle. It happens mainly about 1 minute after turnon from cold. It doesn't do it every time, and after it warms up for a few minutes, I can't get it to happen again.

I know what happened initially. Someone forced in 3 out of the 4 EL34s incorrectly, breaking the pins off. I checked voltages and replaced the power tubes. It works, and the only problem is the intermittent crackling.

None of the controls affect it, and I pulled the preamp tubes 1 by 1, none of them affect the noise, not even the phase inverter (in fact, pulling that one seemed to make it louder, but it could be my imagination). I also pulled the power tubes in pairs, and one by one, this does not stop the noise, except with ALL of them pulled.
Wiggling the tubes doesn't affect it.

I measured every resistor and diode from the PI through to the transformer, all are in spec. I have also checked the coupling caps out of the PI. C101 and 102 are not giving me a reading on my DMM capacitor setting. I haven't checked out the filament resistors or C105, or the bias supply.

I'm thinking maybe the noise is coming in from the bias supply. Any ideas?

[Enzo]

OK, first we must be on guard not to assume it is related to the tube mishap. It may well be, but it also may not.

But start with it. They crammed tubes in hard enough to break them. SO resolder every pin on every power tube socket.

The power tubes are on their own board, and it is supported by two screws at each socket, if I am not mistaken, so I tend to think overall board flexing was minimal, but don;t rule it out. Three pin connector to OT at the end of the board - resolder. 10-pin ribbon - resolder. The four large screen resistors - resolder.

You wiggles the tubes, but did you wiggle the parts? Get a wooden chopstick or other insulated thing and poke each screen resistor. Can't get at them? Just poke the board - tap on the board with the end of the stick. Noise?

Maybe one circuit, maybe not. Take ONE power tube and stick it in the end socket. Noise? Now move it to the second socket. Noise? And then the other two. If we can narrow it down to just one socket, that is a powerful clue. If however it makes no difference, then your bias supply sounds more likely.

That would affect all sockets. So the bias adjust and selector switch are right on the board near one end socket. Grab the switch and wiggle it. Noise? Is there a shaft on the bias pot? If so, wiggle it. Noise? If screwdriver aqdjust, set a screwdriver in the slot. DOn't turn, just tap the end of the screwdriver handle to send little vibrations into the control. Noise?

And get out your scope. Is the noise visible on the bias supply?

[AtomicMassUnit]

I don't have a scope yet. I will be getting one soon!

I resoldered all those connections, and spent a long time poking and prodding components, tapping the board, etc. It doesn't seem to be mechanical. Nothing I've done brings it on, or affects it while it's happening.

I took a power tube (tried different ones just to make sure), and tried it in each socket one at a time. The noise is there in all of them. Wiggling the bias pot, the EL34/6L6 switch did nothing, nor did tapping, poking around here.

One interesting clue though is this... when the amp is on, the mechanical buzz, the same sound I've heard from Hot Rod Fenders, other Peavey amps, etc, just the normal sound of it being on, whenever the crackling starts, this noise gets louder.

[Enzo]

OK, connect the amp to a resistive load, so no possibility of speaker sound. Does that still happen? What I am trying to make sure of is whether that mechanical sound is augmented by some small gain increase in the tubes or some such.

Disconnect the output transformer, and use clip wires to connect some other transformer there. it doesn;t have to match very well, all we want it to do is pass signal so we can hear noise if it occurs. We don;t care about tone or fidelity. If that makes no noise, then your original transformer could be failing.

If you do that, don't connect it to the speaker jacks, just to the tube board. Then clip wires from its secondary directly to some speaker. Don;'t worry about the NFB and stuff.

[AtomicMassUnit]

Ok, will do. I have an OT from a deluxe amp I can clip in. I'll just pop two tubes in, and be close enough.

I will also try the resistive load tomorrow.

[Enzo]

DOn't play through it, just turn it on to see if it still makes noise.

[Gtr_tech]

Sounds like something's drawing more current than it should if the transformer noise becomes greater when this event happens.

[AtomicMassUnit]

Ok, this is weird. I removed the connector that goes to the OT, soldered in a Mercury transformer to the right pins, clipped it to my speaker, and get no sound at all. I plugged the OT back in, get sound.

I know the connections are right, the center tap is orange, with brown and green on either side, I even ohmed it out (600ish ohms across brown/green, 300ish from orange to either side). I know the transformer is good, what am I missing?

The only thing I do hear is that buzz sound getting louder at approximately the same time the noise would occur, but of course I'm not hearing it since I'm not hearing anything through this transformer.

[AtomicMassUnit]

I'm thinking this is a bias supply thing. Whenever the noise starts, if I read the bias at the test points on the back, the voltage starts decreasing as the crackle gets louder. I have a little clip lead setup with a capacitor connected to a jack so I can use a 2nd amp like a signal tracer, and I am hearing the noise on both sides of the bias pot.

[Enzo]

Fair enough, or perhaps one of the coupling caps from the PI is going leaky.

[AtomicMassUnit]

I just unsoldered the coupling caps from the pi just to make sure, it's not them.

I am getting the noise on one side of the EL34/6L6 switch, but not the other, the bias part of the ribbon cable has no noise on it, but after that switch there it is. I'm thinking either the switch itself, or R123 (which measures good). R124 measures good also. I'm going to pop some 6L6s in there and flip the switch to see if I get noise then.

[AtomicMassUnit]

I replaced R123 and it fooled me into thinking it was gone, didn't make the noise for a bit, then there it was again. I'm thinking I may remove the switch and put some jumpers through the traces to make sure it's not the switch.

So far with 6L6s the noise isn't happening. Of course it doesn't do it every time anyway...

[AtomicMassUnit]

I replaced the switch... same problem. Any ideas?

I did get a scope, and I'm learning how to use it. I see the noise on the bias supply at the EL34/6L6 switch, but not at the ribbon cable. When the noise happens (at the ribbon), the waveform jumps, but doesn't show the noise.

It's still really hard to get it to happen at all, it does it maybe once out of every 5 times I turn the amp on.