Gibson GA-15 RVT Noisy Reverb Recovery

[bobloblaws]

We cannot assume both pan jacks are grounded to the pan chassis walls. On many if not most, the input jack is not. It wwinds up that way because the cable plugged into it is grounded at the other end, but that can be involved.

But in my case I have continuity between pan and amp chassis with only either input or output connected. I just now went and confirmed both input and output sleeves are grounded to the pan.

Based on what Chuck said, we want only one connection, in or out, grounded to the pan? ( It's critical to isolate either the input or the output from the tank...)

I don't understand what he means by "keep either end grounded with their respective amplifier circuits".

[g1]

The pan jack ground scheme was why I asked whether it was the stock tank. Whether it is or not, in this case it can't hurt to experiment with different schemes to try to minimize the hum.
However, shorting the recovery tube grid to ground did not remove the hum, which I thought would take the tank out of the equation.

[bobloblaws]

The pan jack ground scheme was why I asked whether it was the stock tank. Whether it is or not, in this case it can't hurt to experiment with different schemes to try to minimize the hum.
However, shorting the recovery tube grid to ground did not remove the hum, which I thought would take the tank out of the equation.

I agree it can't hurt and I'm willing to give it a try but I'm not understanding what Chuck is suggesting. Can you help me with that, g?

[g1]

I think what Chuck H was referring to would mean breaking the physical connection between one of the tank jack sleeves and the tank shell (and insulating via washers etc.). Not an easy process, probably involves removing some rivets.

[bobloblaws]

I think what Chuck H was referring to would mean breaking the physical connection between one of the tank jack sleeves and the tank shell (and insulating via washers etc.). Not an easy process, probably involves removing some rivets.

And that's the extent of it? I didn't understand the last part of "It's critical to isolate either the input or the output from the tank and keep either end grounded with their respective amplifier circuits."

And could't the first part just as easily be tested by just running an alligator clip jumper wire from the amp over to the tank and leave the cable unplugged?

[Helmholtz]

And could't the first part just as easily be tested by just running an alligator clip jumper wire from the amp over to the tank and leave the cable unplugged?

No, you need to use shielded cable.
You might prepare a shielded cable having the shield connected/grounded only at one end/plug

[bobloblaws]

No, you need to use shielded cable.
You might prepare a shielded cable having the shield connected/grounded only at one end/plug

OK, but I thought we were investigating the possibility of a ground loop, as opposed to external interference, how does shielding come into play?

[Helmholtz]

OK, but I thought we were investigating the possibility of a ground loop, as opposed to external interference, how does shielding come into play?

Shielding is necessary to avoid additional hum.
Connecting the shield only at one end breaks the external ground loop between send and return grounds. So the reverb shell is grounded (i.e. connected to amp ground) only at one jack.

[bobloblaws]

Shielding is necessary to avoid additional hum.
Connecting the shield only at one end breaks the external ground loop between send and return grounds. So the reverb shell is grounded (i.e. connected to amp ground) only at one jack.

Well obviously I'm pretty green and I could be completely wrong, but for the purpose of simply confirming or ruling out a ground loop as the source of the hum surely I could tell the difference between what I'm hearing now and any additional noise that might be added by an unshielded wire, I would think. I'm skeptical that that is the issue anyhow as simply disconnecting the tank input cable would in effect do the same thing, no? And as it happens that has no effect on the hum I'm hearing. So combining that with g1 having been convinced that the tank was not a factor based on grounding of the grid making no difference I'm left wondering if I'm barking up the wrong tree here.

[Helmholtz]

Well obviously I'm pretty green and I could be completely wrong, but for the purpose of simply confirming or ruling out a ground loop as the source of the hum surely I could tell the difference between what I'm hearing now and any additional noise that might be added by an unshielded wire, I would thinkgrid making no difference I'm left wondering if I'm barking up the wrong tree here.

I simply doesn't make sense to test for hum reduction by breaking the outer ground loop while at the same introducing new hum of same frequency by using an unshielded wire.

[g1]

As the hum does not go away with the tank disconnected, or when grounding the recovery tube grid, I think it is safe to say it is not tank related.
DC heater and filter cap at node also did not help.
Perhaps something to do with the wiring from the plate of the recovery tube? Maybe picking up hum or something related to footswitch or it's jack?

[bobloblaws]

Perhaps something to do with the wiring from the plate of the recovery tube? Maybe picking up hum or something related to footswitch or it's jack?

Yep, I did check those possibilities. I have to admit I'm not 101% convinced it's not tank related as there is a difference with return cable connected or disconnected, but as I mentioned earlier disconnecting it adds hiss and its hard to tell if the nature of the hum in that state is exactly the same as it is when the cable is connected. In other words, the hiss masks the hum to some degree. But I guess the bottom line is whether it's mostly hum or mostly hiss, tank connected or disconnected, there is a noise problem when the reverb recovery signal is injected into the main signal path via R15, and that is the default when there is no footswitch attached.

In any case, at the moment another issue has cropped up so I can't even deal with the above until I get this sorted out. When I turned it on today I discovered there is a very loud hum that was not there previously, I've basically traced it to V2A. It's there even with V1 and V3 removed and V2B plate disconnected and gone if I disconnnect V2A plate . Earlier I had observed 315V on the low side of the preamp side of the transformer (I don't know if it's called primary or secondary) and 306V on the high side (pin 6), so a voltage drop of 9V. Now I still see 315V on the low side but 313V on the high side, voltage drop of only 2V. FWIW I measure 720 ohms across preamp side and 3k5 ohm across the power amp side with the high and low ends being roughly equal wrt the center tap. I disconnected the cathode bypass cap and verified the cathode resistor is OK. I swapped in another pair of power tubes and switched V1 and V2 and tried disconnecting the center tap of that transformer from the tremolo circuit. With all tubes back in there is guitar output at roughly normal volume but along with the nasty new hum. Ripple is about 2VAC, same as it was earlier. Do you think it could be that PI/trem transformer? I don't know what else to check.

[g1]

Do you think it could be that PI/trem transformer? I don't know what else to check.

Unfortunately it may well be. You had 9V across the primary before. Had the resistance been 720 ohms at that time, that would be approx. 12mA current. That is crazy high for this kind of tube.
I think it likely the primary resistance was significantly higher before, and there is now a shorted portion of it.

[Helmholtz]

Unfortunately it may well be. You had 9V across the primary before. Had the resistance been 720 ohms at that time, that would be approx. 12mA current. That is crazy high for this kind of tube.
I think it likely the primary resistance was significantly higher before, and there is now a shorted portion of it.

I don't think so.
The 2V drop would correspond to a plate current of around 3mA , which looks good. If the interstage transformer had shorted turns, there would be no output.
Plate current is essentially determined by bias, i.e. cathode voltage.
A plate current of 12mA would cause a cathode voltage of 12V, which makes no sense (would cause cut-off).

So if R16 is good, I vote for bad socket contacts.

[bobloblaws]

I don't think so.
The 2V drop would correspond to a plate current of around 3mA , which looks good. If the interstage transformer had shorted turns, there would be no output.
Plate current is essentially determined by bias, i.e. cathode voltage.
A plate current of 12mA would cause a cathode voltage of 12V, which makes no sense (would cause cut-off).

So if R16 is good, I vote for bad socket contacts.

OK, I'll look into that. My reading of 306V plate volts may have been incorrect or taken at a different time than I measured the bottom of the primary.