ORIGINAL: SteveD
Actually... active electronics in a guitar are a "lighting rod" for hum and signal interference.

I've got a bass player client with active pickups in his Spector bass. He has to stand very still in one place to avoid hum.

When he forgets and starts moving around... we start calling him Darth... because it sounds like he's playing bass on a light saber.

Steve, you are full of great points today. (Try not to stand too close... you may puncture someone!) It mystifies me even more why they wouldn't shield guitars with active pickups. I mean... it's a circuit with gain, just sitting there out in the open. It might as well be an AM radio front end.

ORIGINAL: losguy
... It might as well be an AM radio front end.

Too funny... sometimes it IS an AM radio front end. LOL

I put EMGs on bass and guitar and I can stand where ever I want and no noise. (Guitar - POD - Arrdvark)

I did not install the pre-amp that was included on E-bay auction because I thought this could be a problem with picking up noise - seemed a pretty flakey rig.

p.s - if you installed the EMGs yourself, I believe you are supposed to remove the shield lead that goes to the bridge. Maybe that is picking up the noise somehow ?

JR

SteveD - (or any of the other very helpful people on this long string)...

I have a similar issue with hum I wanted to ask you about. I am recording acoustic guitars with 2 pencil cond mics, into a MAudio Firewire 410 preamp. The mics record a hum - I don't hear it when I record, but only when I playback the recorded track (very frustrating, it's ruined multiple tracks!). My large cond mic works great (Rode NT2), no noise, so I think it's the pencil mics. They're cheap (Samson), but they sound great for what I'm doing... except the noise that is.

Will the hum eliminator work in this case as well, when mics are involved? I thought it was cheaper to buy a $60 Ebtech than spend several hundred on new mics. Appreciate your input.

Thanks, Greg

grey, just to make sure where the problem is, do you have access to another system that you can plug the pencil mics into? If the mics work there, then you can go down one level toward the problem. You're sure that you don't hear hum any other time?

ORIGINAL: losguy

grey, just to make sure where the problem is, do you have access to another system that you can plug the pencil mics into? If the mics work there, then you can go down one level toward the problem. You're sure that you don't hear hum any other time?

- great minds...

99% of the time when I have seen a ground loop it comes from the same thing. A microphone cable with the connector shell connected to pin 1 of the XLR. Here's the executive summary. Open your microphone cables XLR connectors. If you see a wire from the pin 1 connector to the grounding tab cut it.

Here's why this is a big deal. "Ground" is an imaginary reference point indicating a place on in the circuit where everything else is referenced to. Chassis ground or ac ground if you like is theoretically the 0volt reference point for ac power line. Unfortunately no two outlets will have "ground" pins at exactly the same potential because of different lengths and guages of wire and different load conditions for each circuit. For the sake of safety this is not a big deal. If the chassis of the refrigerator on one side of the kitchen is floating 5 Volts ac away from the toaster on the other side of the kitchen no problem. In some extreme cases it can be significantly higher and then it is a real danger but that usually indicates something drastic like a broken neutral.

OK that's ac ground. For your audio signal you also have a "ground" which is in a balanced signal the 0 Volt reference point for the audio signal. This is sometimes tied back to the chassis ground through a high impedance which keeps it from floating to a high level. So what happens when Pin 1 is tied to the shell of an XLR is the chassis ground is given a low resistance path to the audio ground. Oh oh! Now imagine that refrigerator and toaster are a power amp and mixer sitting on either side of a room. With proper isolation of the audio signal it doesn't matter what the two chassis are at because of the high resistance between them but with pin 1 shorted to chassis ground you've now connected the chassis of both units to each other through a 25' piece of 26 guage wire. The chassis' will try to equalize their voltages through the audio interface by forcing a current. Depending on the resistance of the wire you will get varying degrees of voltages impressed on your signal path.

That's all well and good for the guys with balanced interfaces but with an unbalanced interface there is a real problem. Most of the unbalanced equipment ties the "ground" side of the signal to the chassis of the unit. You only have a few options. Galvanically isolate the signals with a transformer being careful to make sure the ground isn't carried through, power everything from the same source to reduce the difference in chassis voltage and/or tie all the chassis together with a low impedance by putting everything in a rack.

For the original poster I find that if I'm forced to go from non-balanced to balanced I usually just ignore pin 1 in the XLR altogether and wire my unbalanced signal to pins 2 and 3.

I've done a lot of PA systems over the years and once I figured out this basic concept I never had to deal with a ground loop again. It's nice to idle thousands of watts and only hear gentle hiss. One other point, just because you bought a "professional" cable or snake doesn't mean it's wired correctly. I've seen plenty of big buck snakes wired with pin 1 to the connector shell. It's also why I make my own mic cables.

Dan

ORIGINAL: danwilms
For the original poster I find that if I'm forced to go from non-balanced to balanced I usually just ignore pin 1 in the XLR altogether and wire my unbalanced signal to pins 2 and 3.

Brute-force ground lifting on the audio cable is one way to do it, but forces all of the ground loop voltage to appear as a common-mode signal at the balanced input. Because the input is differential, it cancels the ground loop voltage, up to the CMRR of the input stage. But if the ground loop signal is big enough (like you say, it can be many volts) it can challenge and degrade the common-mode range of the input stage. I commented about this on page 2 of this thread.

It's probably OK for most live situations, especially for input stages that are transformer-coupled (they can usually stand upwards of 1000 volts or more common-mode). But I prefer to be more careful with sensitive recording preamp inputs. I have a four-step procedure outlined (again, page 2) that uses modern circuitry to allow safely breaking the loop at the power side, while keeping all signal grounds intact. (And it works, too!)

(edited typo)

losguy, this was also on page two:

The Ebtech solution you're onto is a good path, absolutely. I use the hum eliminator in a live bass rig and always bring one with me to live sound gigs, just in case. These are the signal ground isolation transformers. They work like a charm, but do notably roll off high frequencies as you run with lower operating levels.

Is this poster saying the humX rolls off the highs? Does the humX roll off the highs?

Considering they're cheap at 60-70 bucks apiece, that's awful expensive if they compromise the quality of thousands of dollars worth of gear.

ORIGINAL: TheFingers
losguy, this was also on page two:

The Ebtech solution you're onto is a good path, absolutely. I use the hum eliminator in a live bass rig and always bring one with me to live sound gigs, just in case. These are the signal ground isolation transformers. They work like a charm, but do notably roll off high frequencies as you run with lower operating levels.

Is this poster saying the humX rolls off the highs? Does the humX roll off the highs?

If I may answer, the answer is no. The Ebtech Hum-X is designed to go into the power path, not the signal path. So, unless you're putting your audio on the power line, I wouldn't expect it to change your audio by any amount worth considering at all. That's the beauty of this design.

Ebtech also makes a product called the Hum Eliminator, which is basically a balanced 1:1 direct box with a ground lift switch permanently set to "lift". It does go in the audio path and uses audio transformers, so yes, it could do something to your sound. What I have found with audio transformers, though, is that if you drive them actively with a low-impedance output, and not too hard (stick to line levels) they are remarkably transparent. Drive them with something flaky, and YMMV a lot.

I got a new Art Tube Pre and it induced 60 cycle hum.
Not nice.
Even OFF, it produced the hum.

Finally, I plugged it in to another 110v wall socket. No hum.

Appearantly some of my wall sockets are out of phase.
All are grounded, but some are grounded left, some right.

Makes me think the signal chain really starts at the wall, not the mic.

ORIGINAL: dmassey
Here's alink to some info some of you already know about:
http://www.rane.com/note110.html

Read the WHOLE page! If you don't, you'll miss the important part about how most manufacturers do not tie pin 1 to chassis. If you miss this section, you'll be mis-led into thinking that the common method is to leave the shields intact on both ends. This would be correct in a perfect world where the equipment did not tie the audio ground to pin 1 internally. Chances are the stuff you have ties pin 1 to audio ground, not chassis ground. Either way, there is a diagram of how to wire balanced cables with the ground removed at one end.
And the debate goes on...

Great link. But I differ somewhat from your conclusions. In a perfect world, all manufacturers would tie pin 1 to ground: but differently on inputs and outputs. Balanced outputs would tie pin 1 to chassis ground, and balanced inputs would tie pin 1 to both chassis and signal ground. It says so right on Figure 1 on the article, what they call "The Absolute Best Right Way To Do It" (for good reason). The problem comes in, as you say, when manufacturers deviate from this.

But the best solution for this imperfect equipment design is not to lift audio grounds, at least without providing a local ground reference for the input. That local ground reference is the jist of what they are talking about for the rest of the article. Floating grounds may work, but it's usually because the equipment can handle the CMR abuse, or there's a transformer somewhere restoring the local reference (see Figure 2, for instance).

In fact, lifting audio grounds (Figure 3) is what they call "The Last Best Right Way To Do It" (again, for good reason). It's still "Right", but it's the worst of the right ways.

Absolutely because you don't want the shell of the microphone connected to a different ac ground then any other thing you might touch. A proper mic shell is isolated from it's signal common by hopefully a few meg ohms. Think about it this way, connect the mic shell to the chassis ground of the mixer now plug a guitar amp into a circuit with a faulty neutral where the strings are floating to a dangerous level. Now smack those lips on the mic while holding that Dsus4. Bad things happen. The amplifier will try to force it's ground to the same reference point as the mixer by putting as much current through your kisser as it takes to equalize the mixer/amplifier grounds up to the compliance voltage of the circuit. If it's 25~50 volts you get a tingle. Higher then that and you can get a one way ticket to personally discuss your technique with Hendrix.

Granted mics can't cause a ground loop unless the shell is touching something. Fine on a stand with a plastic clip when no one touches it. But grounding the mic shell does not provide safety. It actually can present a danger by providing a current path through the musician. That's why guitarist's take the brunt of abuse when it comes to stage shocks.

I understand the idea of removing one side of the ground. That's a technique for providing a low impedance path for current on the shield. It's a way of mitigating capacitively coupled noise sources like radio stations. This works fine with shielded cables with a braid over the twisted pair and a seperate drain wire for the audio ground but usually isn't necessary because the capacitively coupled interference is so weak it can be drained by the Meg ohm resistance of the audio ground to chassis ground. It doesn't do squat for inductively coupled noise sources like lighting cables near your mic cables. Only balanced signals and increasing the distance between the transmitter and receiver can help with that. Did you ever wonder why mic cables have the conductors twisted? It's to try and make sure the interference signal from a magnetic field gets into both wires equally. It can't be stopped.

As for the danger to the equipment from vastly different chassis voltages well that's an example of a situation that shouldn't exist anyway. A simple plug-in three lamp tester available from an electrical supply store will tell you if a circuit is so bad you shouldn't plug into it. Yes I've seen bad outlets in clubs and have chewed out licensed electricians for connecting ground to the neutral buss bar at college gigs. When in doubt check it out. This stuff costs way too much money to blindly plug it in.

There's an old saying in circuit design. Follow the current. When you figure out where the current is flowing you have a much better idea of what's going on.


Dan

Yes we may be agreeing but not about the same thing. Pin 1 of the microphone must be connected to pin 1 of the mixer but neither pin 1 should be connected to that little ground tab that connects to the shell of the connector. If the manufacturer feels it's a good idea they will do it in the equipment. Don't do it in the cable.

Dan