Ground loop hum in an amp switcher

All frequent questions on switching: true or not true bypass, transistor-based or mechanical.
Post Reply
User avatar
Dr Tony Balls
Diode Debunker
Information
Posts: 843
Joined: 07 May 2009, 14:29
my favorite amplifier: Sunn Model T
Location: Brooklyn, NY
Has thanked: 4 times
Been thanked: 172 times

Post by Dr Tony Balls »

I'm building a switch box for a friend to answer a weird request he had. Basically he needs two ins for guitars and two outs for amps, and needs to flip between which guitar goes to which amp. There are the obvious ground loop issues to deal with when making any dual amp splitter. I tried to get around it on this one by making a ground lift on one output to sever the loop. Here's my schem:

Image

The two positions should produce the results of A->X,B->Y and A->Y,B->X. This works fine but i'm getting a decent amount of hum. The ground lift reduces the hum, but it doesnt go away. Can anyone explain why this doesnt work? I'm trying to figure out a solution. I was hoping to get this box to be passive so no transformer isolation or anything, but I dont know if thats possible.

User avatar
Lucifer
Cap Cooler
Information
Posts: 671
Joined: 06 Apr 2011, 12:59
my favorite amplifier: Vyse BBQ Blue
Location: Dorset, UK
Has thanked: 89 times
Been thanked: 72 times

Post by Lucifer »

Hi Tony,

As you've drawn it, both guitars go to amp X or both guitars go to amp Y - ie, the guitars are shorted together.

Maybe you also need a ground lift on amp X too, so that only one amp is grounded while the other lifted (and vice versa).
”Sex is great - but you can’t beat the real thing !” - The Wanker’s Handbook

User avatar
Dr Tony Balls
Diode Debunker
Information
Posts: 843
Joined: 07 May 2009, 14:29
my favorite amplifier: Sunn Model T
Location: Brooklyn, NY
Has thanked: 4 times
Been thanked: 172 times

Post by Dr Tony Balls »

Shit, you're right. Here's the corrected one:

Image


I could do the ground lift change that you suggest but would there be a point? Does it matter which output ground is lifted?

User avatar
Lucifer
Cap Cooler
Information
Posts: 671
Joined: 06 Apr 2011, 12:59
my favorite amplifier: Vyse BBQ Blue
Location: Dorset, UK
Has thanked: 89 times
Been thanked: 72 times

Post by Lucifer »

Possibly not.

I was just thinking that only one guitar will be played at a time, so maybe the unused one should be muted so as not to cause feedback (if the guitar and amp volume are still turned up).

You could do the muting by shorting the input to the unused amp - but that may give you a thump.

I've submitted a relay-based switcher that I made for a friend - but that just selects one of three bass guitars to one amp. You may be able to use some of the ideas and extend it for two amp. The topic heading is "A Switching Box" in the workbench section.
”Sex is great - but you can’t beat the real thing !” - The Wanker’s Handbook

User avatar
Dr Tony Balls
Diode Debunker
Information
Posts: 843
Joined: 07 May 2009, 14:29
my favorite amplifier: Sunn Model T
Location: Brooklyn, NY
Has thanked: 4 times
Been thanked: 172 times

Post by Dr Tony Balls »

Lucifer wrote:Possibly not.

I was just thinking that only one guitar will be played at a time,

Not correct, both will be played at the same time. Its basically two guitar players that need an amp-swapping box.

User avatar
Lucifer
Cap Cooler
Information
Posts: 671
Joined: 06 Apr 2011, 12:59
my favorite amplifier: Vyse BBQ Blue
Location: Dorset, UK
Has thanked: 89 times
Been thanked: 72 times

Post by Lucifer »

Wow, OK, that's a bit different.

How about if you switched not only the 'hot' side of the guitar signal, but also the 'earthy' side, avoiding any common grounding. As long as all the jacks are isolated from the box, and you use screened cables, you should avoid any hum.

It would call for a four pole changeover switch, but that shouldn't be a problem.
”Sex is great - but you can’t beat the real thing !” - The Wanker’s Handbook

User avatar
Dr Tony Balls
Diode Debunker
Information
Posts: 843
Joined: 07 May 2009, 14:29
my favorite amplifier: Sunn Model T
Location: Brooklyn, NY
Has thanked: 4 times
Been thanked: 172 times

Post by Dr Tony Balls »

Lucifer wrote:Wow, OK, that's a bit different.

How about if you switched not only the 'hot' side of the guitar signal, but also the 'earthy' side, avoiding any common grounding. As long as all the jacks are isolated from the box, and you use screened cables, you should avoid any hum.

It would call for a four pole changeover switch, but that shouldn't be a problem.

Yeah, thats what i'm doing. Got a 4PDT on order, but i'm curious as to WHY this is happening. What is the mechanism thats causing the hum and is there a simple passive way to fix it?

User avatar
Lucifer
Cap Cooler
Information
Posts: 671
Joined: 06 Apr 2011, 12:59
my favorite amplifier: Vyse BBQ Blue
Location: Dorset, UK
Has thanked: 89 times
Been thanked: 72 times

Post by Lucifer »

I'm as confounded as you are now, if you're switching like I suggested.

There should be no hum. Is it a lash-up outside of a box, picking up interference maybe ?

If it's in a box, maybe the case needs an earth from one of the amps to help with screening the internals.

If none of the above, then I'm stumped. Maybe one of our other experts can solve it.

Good luck :thumbsup

PS - in the switching box that I did for my mate, I deliberately kept the audio earth separate from the power supply Ov (used for the relays and LEDs) - and it hummed like buggery !
So I linked the two 'earths' together, and it all went quiet (no hum - but clean audio). That confounded me as well !
”Sex is great - but you can’t beat the real thing !” - The Wanker’s Handbook

User avatar
Dr Tony Balls
Diode Debunker
Information
Posts: 843
Joined: 07 May 2009, 14:29
my favorite amplifier: Sunn Model T
Location: Brooklyn, NY
Has thanked: 4 times
Been thanked: 172 times

Post by Dr Tony Balls »

Yeah, I tried completely closing the box to shield it and it didnt help much. The box is grounded to one amp through the X output jack, and the two inputs are grounded to the box, so that should cover all signal grounds.

User avatar
The G
Grease Monkey
Information
Posts: 558
Joined: 21 Jun 2009, 11:43
Location: Bucharest, Romania
Has thanked: 408 times
Been thanked: 284 times

Post by The G »

Do you still have hum if one or the other of the amps is not connected?

Anyway, here's a good read: .

User avatar
Lucifer
Cap Cooler
Information
Posts: 671
Joined: 06 Apr 2011, 12:59
my favorite amplifier: Vyse BBQ Blue
Location: Dorset, UK
Has thanked: 89 times
Been thanked: 72 times

Post by Lucifer »

I think the grounding of the two inputs is the problem.

Like I said in my earlier post, if you keep the inputs and outputs floating (ie, not grounded - except for one amp jack supplying the screening ground to the case - then you should not get hum.

The guitars will get their grounding through whichever amp they are connected to - they do not need to be grounded at the box too. If you have both input jacks grounded, you are effectively connecting the grounds of the two amps, through the guitar cables, together - thereby creating a ground loop and causing the hum.
”Sex is great - but you can’t beat the real thing !” - The Wanker’s Handbook

User avatar
Dr Tony Balls
Diode Debunker
Information
Posts: 843
Joined: 07 May 2009, 14:29
my favorite amplifier: Sunn Model T
Location: Brooklyn, NY
Has thanked: 4 times
Been thanked: 172 times

Post by Dr Tony Balls »

CRAP. Nevermind. It works fine.

I dont know what I did before that was causing hum but it works as expected now.

Post Reply