ground loops aplenty?

[niftydog]

You're right in that an inductor does resist changes in current flow.. but...

people often forget that mains hum (50 or 60Hz normaly) is an AC SIGNAL that lies whithin the audible range.  Any attempt to stop it from circulating in the audio path is going to also affect the bottom end of your sound.

I don't know where this guff about Boss power supplies stopping earth loops via a diode in the ground line came from, but all it would do is rectify the hum.  Thus you get not a 50/60Hz hum, but a 25/30Hz RECTIFIED hum... which is obviosly easier to filter.  Nevertheless, it's still there and might still cause you problems.

In a simple sence, a ground loop hum is the symptom of too many ground connections.  Ensuring that each device only connects to ground at ONE point is the best way to stop ground loop hum.

[jubjub]

Ok, I now have all this info which is brilliant. Thanks guys. But how do people do it then. All these cats that use a number of amps on stage. Do they just put up with the buzz?  It seems to be such a common thing to do yet when I try and do it there's no end of bloody complications. Is it the amps I'm using? Just an old Marshall jmp head and an old Fender pro reverb. On their own they seem to be ok. Would the potential difference be less between to amps of a more similar nature? Is there some box that costs thousands of pounds called a de-looper that these guys are using? This rash is getting out of control. 'Oh nurse' !

[jsleep]

I don't know if we've hit on this yet:

http://www.generalguitargadgets.com/v2/index.php?option=displaypage&Itemid=102&op=page&SubMenu=

as you can see in the schematic, this is an example of a transformer isolated splitter, A/B or comination.  Uses inexpensive mouser audio transformers.  Since we aren't usually "hiFi" in the guitar amp world, it does a good job.

JD

[downweverything]

i think paul initially had the right idea, maybe simpler you could find a 1 to 1 transformer with dual secondarys, dont make a balanced output just throw the chasis connection away on the output and just use the transformer secondaries (dont connect either output negative sides to the chasis).  ive done this before and seemed to work fine.

[Paul Marossy]

JD-

I have seen that A/B Splitter Box before. It appears that the signal only goes to either A or B, though. I think jubjub wants two simultaneous outputs...

downweverything-

I think you can see where I was headed. I also thought of a transformer with two secondaries, but I have never actually seen one like that made specifically for audio. Someone somewhere probably makes one of those, though.

niftydog-

I hear what you are saying, but the open low E on a guitar is 82.41Hz, so if you cut the 60Hz band, theoritically, it shouldn't matter. If it were bass on the other hand, well then it may be a problem.

[smoguzbenjamin]

paul there's a switch for both Look closely

[Paul Marossy]

Oh, duh. My mistake. I see that on the schematic now that I opened my eyes. I'm just not doing well on this thread, am I?  :oops:

That's what happens when I am doing three things at the same time at work and posting here, too.

[smoguzbenjamin]

Take a deep breath, calm down and relax. Works for me. Oh yeah and try to ignore any idiots that come across your path

[downweverything]

i wasnt correcting you either, i was just adding to your idea, all is cool...

[Paul Marossy]

No worries, mates.  :wink:

[smoguzbenjamin]

I might need to build one of these. Dang, so many projects, so little cash.

[niftydog]

I hear what you are saying, but the open low E on a guitar is 82.41Hz, so if you cut the 60Hz band, theoritically, it shouldn't matter. If it were bass on the other hand, well then it may be a problem.

But it's difficult and expensive to design a filter that attenuates 60Hz sufficiently, whilst not affecting 82Hz;  Phase shift becomes a problem.  Can anyone say "4th order Chebyshev high pass filter with limited band pass ripple"?


Folks, we're all forgetting that we're talking ground LOOPS!  A ground loop occurs when current flows in a circular fashion from ground point to ground point.  ie; a ground loop indicates a superfluous ground connection.  Find the unnecessary connection and get rid of it!

If you ensure your devices are only connected to ground at one point then there is no potential for a loop.

Picture this.  Amplifier 1 and 2 are both connected to the mains earth via the power point.  Then, you connect a signal cable between the two, suddenly there is a circular path for ground currents; from amp 1, through amp 2, via the signal cable back to amp 1.

Jubjub, try this for me.  Connect up to each amplifier in turn, one at a time to ensure that the amps themselves aren't the source of the hum.  Now, connect up your system so that the hum is evident.  Then, unplug your guitar;  The hum should remain.  Now, lift the shield on the cable feeding the second amp; I'm betting the hum should vanish.

Just make sure your amps are ONLY connected to ground at ONE point.  Trust me, this should work.

PS; a bunch of balancing transformers may help, but it requires a differential input stage and a bunch of stuffing around.  In the end you could STILL end up with ground loops.

Think of a decent DI box.  They have a "ground lift" switch even though they are using balanced signals!

A 1:1 transformer might work, but it's a lot of hassle compared to just lifting that shield!

Lift that ground, just to entertain me, and let us know how you go!

[Paul Marossy]

niftydog-

Points well taken. Ground loops are a pain in the neck, and sometimes you still get them even with a Direct Box. I never said that what I proposed would be a sure fire way to eliminate it.

What I don't understand is this: What happens when you remove the ground of an AC signal which swings below and above ground? I have had many instances where things will not work (no signal, just hum) with the ground not being connected. With this in mind, if you lifted the ground from one of the amps, then what does the signal to that amp do? Does it find a way to reference itself to ground through the other amp's chassis?

[niftydog]

Does it find a way to reference itself to ground through the other amp's chassis?

er, in a word, yup.  Yes you can run into the problems you're describing, but it's worth trying all the same!  As long as the circuit gets a ground from somewhere, it should work 9/10 times.  It all depends on the exact circuit in question.

Again, it goes back to what I was saying.  You're not removing the ground from the amp entirely.  You just want ONE ground connection, and most amps get that via the power supply, which is ultimately connected to the mains.  In Jubjubs case, his guitar is grounded via the first amp, there's no need to connect it to the second.

Also remember that ground is meerly a reference point.  An AC signal doesn't have to pass "through" ground, it can float above or below it.

If you have NO ground connection at all, you'll either get a reduced volume or no signal.

[Paul Marossy]

OK, that's what I thought. I have noticed that if the signal does still pass thru whatever device with the ground removed, the volume does drop some.

[Gearbuilder]

Hi there,
I use four amps in my main  guitar rig and i haven't got any ground loop,the solution is very simple ,i only cut the brain shield at one side of each cords runnig beetween my Fx rack, the cords are dedicated to this use (Red point or other mark). Only one normal cord Must be used(shielded on the two jacks) to avoid the ground return of your effects loops  The ground  return is done by this one . All your amps must be grounded with a three pinout prong.Like that the neg signal of the amps are separate in the audio path .It works very well.
Bruno

[Paul Marossy]

I wonder why someone hasn't come up with "ground lift patch cords"...

[jubjub]

IT's ALIVE!!!

I've done what you suggested nifty dog. And the hum went crazy, but after trying it on a few different combinations I found one that works.
I put 100k reisitor on the shield of one lead. Plug that into the Fender and all is well.
I've got them both cranked and there is no discernable extranious hum.
As to why no ones made a ground lift patch lead, the health and saftey people would never let it on the shelf. It couldn't get a CE mark either.
It's to general an item with a potentialy lethal aplication.
There has been some realy interesting ideas here. A veritable ground loop brains trust. Thank's everybody fro contributing. Now for some hum free plank spankin!!!

[Peter Snowberg]

When this thread is "done", somebody please remind me to add it to the FAQ. Thanks!

Congratulations on your success jubjub!

Take care,
-Peter

[Paul Marossy]

Cool, I'm glad that is working for you.

Oh yeah, I forgot about UL listings and all that...