VOX AC15 Heritage... TB Channel bleeds into the EF86 channel.

[SoulFetish]

This was something that I checked on day 1 or day 2... I'm going to say, no. It's difficult to tell, really, as there is limited equalization on that channel. 1 x 3 way voicing select + 1 x bass shift switch.

I'm a bit nervous about grounding things, as I've never done this before. As long as a terminal or lead references to ground, is it okay to bypass these points directly to the chassis?

In this instance, it is perfectly fine to directly ground that grid. But, it some circuits, it isn't okay.
But, since we're troubleshooting noise issues here, we can AC ground different points of the circuit to help us locate where the source of the problem is and eliminate others. Here is how we can do that:
Take a fairly high value/high voltage film capacitor, lets say .47µF/630V (or as close as you can get to that), and connect one lead to ground via an alligator clip and VERY CAREFULLY use the other lead as a probe. You can use this to AC shunt different points of the circuit (ie grids, plates, pots, whatever) to ground. This won't upset any of the DC operating points for the most part and will not harm your amp. But keep in mind, please, that when probing high voltages, the cap will charge up and can shock the shit out of you. So, it's good practice to discharge it after each time you probe the a part of the live circuit.
I want to ensure you're comfortable and understand what I'm describing so you can do this safely, so let me know if you're unsure of anything.
The two most common ways that this problem can happen is that you have a compromised ground in the channel that is bleeding through. In this case, I'd be suspicious of the volume pot. The other common cause is a failing filter capacitor that is not decoupling as it should. This could be from the electrolyte degrading, or from one of the solder joints failing (probably on the grounded lead). But, there are always all of those really uncommon ways things can go wrong as well.
This will help us look for the common ways first.

[Cooper1962]

Do you have a pic of the oem tube shielded in shrink wrap?

nosaj

Sure! Here's a OEM EF86 in a combo version. Notice the floating socket mount... The head version (which is what I have) doesn't have this.

[Cooper1962]

In this instance, it is perfectly fine to directly ground that grid. But, it some circuits, it isn't okay.
But, since we're troubleshooting noise issues here, we can AC ground different points of the circuit to help us locate where the source of the problem is and eliminate others. Here is how we can do that:
Take a fairly high value/high voltage film capacitor, lets say .47µF/630V (or as close as you can get to that), and connect one lead to ground via an alligator clip and VERY CAREFULLY use the other lead as a probe. You can use this to AC shunt different points of the circuit (ie grids, plates, pots, whatever) to ground. This won't upset any of the DC operating points for the most part and will not harm your amp. But keep in mind, please, that when probing high voltages, the cap will charge up and can shock the shit out of you. So, it's good practice to discharge it after each time you probe the a part of the live circuit.
I want to ensure you're comfortable and understand what I'm describing so you can do this safely, so let me know if you're unsure of anything.
The two most common ways that this problem can happen is that you have a compromised ground in the channel that is bleeding through. In this case, I'd be suspicious of the volume pot. The other common cause is a failing filter capacitor that is not decoupling as it should. This could be from the electrolyte degrading, or from one of the solder joints failing (probably on the grounded lead). But, there are always all of those really uncommon ways things can go wrong as well.
This will help us look for the common ways first.

Should I first attach one end of the cable to the middle terminal of the TB volume pot, and the other end to the chassis, then turn on the amp?

[Cooper1962]

I'm curious... What does this diode + capacitor + resistor circuit assembly do?

[Jazz P Bass]

I have seen that on Marshall amps.

Some kind of reverse protection chassis reference.

[SoulFetish]

I'm curious... What does this diode + capacitor + resistor circuit assembly do?

[ATTACH=CONFIG]54195[/ATTACH]

To me it looks like it presents an impedance to ground currents, so they don’t loop through the chassis. The cap I would imagine is so any HF that ends up on the power supply has a low impedance path to the chassis.. the diodes are usually there to pass fault currents in the event of an electrical failure.

[Cooper1962]

I have that on Marshall amps.

Some kind of reverse protection chassis reference.

Thank you.

[Cooper1962]

To me it looks like it presents an impedance to ground currents, so they don’t loop through the chassis. The cap I would imagine is so any HF that ends up on the power supply has a low impedance path to the chassis.. the diodes are usually there to pass fault currents in the event of an electrical failure.

Thank you. I didn't get a chance to work on the amp today.

[Cooper1962]

It's still bleeding over, so it's not an issue with the caps that I replaced.

[SoulFetish]

It's still bleeding over, so it's not an issue with the caps that I replaced.

I don't understand, did you shunt different points of the circuit with the capacitor "probe" like we talked about? What caps did you replace?