6G15 build ground loop hum question

[tubeswell]

On the 6G15 I recently built, I rewired the ground setup and have it virtually hum-free on the 5G9 I built a while back. But when I hook the 6G15 through the 5E3 I built more recently, it hums quite a bit. (This with both sets of amps plugged from the same main outlet via a multibox)

The riddle is that the 5E3 and the 5G9 are both as quiet as a mouse without the 6G15 plugged in front. I've been wracking what's left of my lizard brain trying to work it out and I was wondering whether the way the PT primary is hooked up may affect the potential for ground loop hum?

All amps are 3 wire mains systems. Would reversing the active and neutral primary leads on the 5E3 PT possibly make a difference? (or would it do any harm?)

[txstrat]

Have you tried shielded wire in the 5E3? It may be some picked up hum.

[tubeswell]

The 5E3 has shielded wires from inputs to both V1 grid pins and from Vol to the driver stage grid pin. It also has a choke filter and elevated heater. It is a quiet as a mouse when my geetar is plugged directly in - dead silent on the hum-bucking positions. The only similarity about the 5G9 and 6G15 that is dissimilar from the 5E3 is that the 5E3 has elevated heaters (tied to output tube cathodes), whereas the other two don't. (Could that cause a ground hum when the 6G15 is plugged into the 5E3?)

That's why I was wondering about whether the VAC phase of the PT primary makes a difference or not

[loudthud]

There are two kinds of hum. The 50Hz sine wave kind and the 100Hz (rectifier) buzz. Usually they occur in combination but they have different sources. Does the 5E3 hum with the volume controls turned all the way down when the 6G15 is connected?

[tubeswell]

There are two kinds of hum. The 50Hz sine wave kind and the 100Hz (rectifier) buzz. Usually they occur in combination but they have different sources. Does the 5E3 hum with the volume controls turned all the way down when the 6G15 is connected?

Nope - It only hums when the 5E3 Vol is up. As soon as you kill it, it is silent (but that would figure right? - no signal getting through to the speaker).

[txstrat]

Did you think about putting the heaters of the 5e3 off the cathodes directly to ground. That should at least tell IF it causes the hum.
After all the grounds of the 5e3 and the 6g15 are at slightly different potential.

[Enzo]

Your two amps sit there alone silent. Your reverb sounds quiet through one but not the other. Try this. Set the two amps side by side not touching, and plugged into the same outlet strip. No hum, right? Now plug a good guitar cord into the two inputs. We are connecting their inputs together. No we don;t expect anything to come out, but did a hum appear? If so, you have a classic ground loop.

[tubeswell]

Did you think about putting the heaters of the 5e3 off the cathodes directly to ground. That should at least tell IF it causes the hum.
After all the grounds of the 5e3 and the 6g15 are at slightly different potential.

I put the 5E3 heater CT back at ground potential and it didn't make any difference. Hum is still there (But the idea seemed to have merit - cheers ) However I'm putting the heaters back to the cathode because now there is a noticeable hum in the amp just with the guitar plugged in

[tubeswell]

Try this. Set the two amps side by side not touching, and plugged into the same outlet strip. No hum, right? Now plug a good guitar cord into the two inputs. We are connecting their inputs together. No we don;t expect anything to come out, but did a hum appear? If so, you have a classic ground loop.

Okay I bridged the two inputs of the 5E3 and 5G5 and the hum appeared (as expected). However why would the reverb unit hum more severely in the 5E3 but hardly at all with the 5G9?

[Enzo]

Because every piece of gear has its own ground potential. That ground wire in the mains cord has resistance, even if low. You can plug two things into the same mains strip, but that doesn;t guarantee their chassis ultimately sit at the same voltage. Whatever little leakage currents there may be on that chassis flow to ground, and along the way they drop a voltage across the wire resistance. That results in the chassis having its own potential.

Two pieces of gear with the same potential won;t hum together, but two that differ will. The reverb agrees with the quiet one, but not the 5E3. it only takes a small fraction of a volt.

Individually your gear is working fine. You already kow the amps are silent, and you know the reverb can be with the proper mate. (Something I keep telling my wife) SO there is no point in looking for defective filters or anything - those filters have no idea what amp you connect to or not. You have a slight difference in chassis potential between two flawless items. The two ends of the mains cord are not the same place.

This is a problem to solve, not a defect to fix. That is why audio gear is full of ground lift switches, balanced lines, ground loop busters, and so on.

Welcome to the world of audio.

[tubeswell]

Individually your gear is working fine. You already kow the amps are silent, and you know the reverb can be with the proper mate. (Something I keep telling my wife)

Okay thanks Enzo. I had wondered about the role of individual unique chassis potentials.

I guess I should tell my wife something like that too (but I wonder if she'll appreciate the potential difference? )

[dai h.]

in an old thread on this where R.G. and Steve C. commented, it sounded like you could get them to work (together without hum) conventionally just by having the circuit ground to frame ground connected towards the input end on both (all the high current stuff off the chassis and only the picoamp return end to chassis and thus frame then to safety ground through chassis). Not sure if that was said exactly but that was what I came away with. And also, (according to R.G.) that the ground lifting scheme with the low value R plus cap and diodes around it isn't the safest scheme though it can work to cure the audio hum problem.

[tubeswell]

in an old thread on this where R.G. and Steve C. commented, it sounded like you could get them to work (together without hum) conventionally just by having the circuit ground to frame ground connected towards the input end on both (all the high current stuff off the chassis and only the picoamp return end to chassis and thus frame then to safety ground through chassis). Not sure if that was said exactly but that was what I came away with. And also, (according to R.G.) that the ground lifting scheme with the low value R plus cap and diodes around it isn't the safest scheme though it can work to cure the audio hum problem.

Yep I recall R.G. saying that about the ground lifting scheme in a thread where I was asking about designing the 6G15 to eliminate ground loops. However, now I wonder whether I could work out the potential difference between the two chassis and see by trial and error if I can vary the length of the mains cable to achieve the smallest possible potential difference between them (if that would work?)?

Otherwise its building or buying a good loop buster.

What's that other idea you 'came away with'? Can you sketch a schematic (- I'm not quite following what you are saying)?

[dai h.]

well, (since you are reporting a problem interaction i.e. hum) I'm making the assumption you have higher current portions (returns) going through the chassis. Sort of thinking since for example on stereo equip., say a CD player and tape deck if you connect left and right IN/OUTs to ea. other it is a loop but typically I've never experienced a hum problem(I've tried chopping the ground/return on one side to break the redundant connection but don't recall experiencing any difference). So I drew out the reverb to amp ground connections on a piece of paper to try to make it more clear to myself and if you do have a safety ground on both there will be a loop but I'm thinking since (or IF) the only circuit ground to chassis connection is the picoamp return end, then perhaps there won't be (an audio hum problem--potentials so close the redundant one won't matter). Then perhaps if that doesn't work maybe lifting the signal ground resistively would work.

[tubeswell]

Signal ground lift resistor

What value resistor would be enough to overcome the loop? 47R? (I was thinking I could try it out in a shielded box between the unit and the amp)