Old Production Takes From an Old Guy

The topic for today’s post comes from a reader, Jonathan Mould. He wanted to know more about the electrical side of sound. This is an interesting topic because sound is both electrical and physical. Sound systems turn physical movement of air into electrical signals, then process and amplify those signals (sometimes changing them into streams of 1s and 0s and back), and finally turn those amplified signals back into physical air movement. It’s all kind of crazy when you think of it that way.

I started this post thinking I could tackle three topics and realized that even a cursory explanation of ground loops would take a whole post. So, you’re witnessing the beginning of a series here. First up, Ground Loops.

Ground Loops

To understand ground loops, one must first understand electricity. A full explanation is beyond the scope of this article, but here’s a brief description. Standard 120 volt circuits consist of three leads; a hot (current carrying) lead, a neutral (the return) and a ground. To vastly over-simplify, the electricity leaves the panel on the hot lead, travels to the appliance, does some work and returns to the panel on the neutral. The ground is properly called a safety ground and serves one basic function—to send any electricity right back to a safe place (the earth) if anything goes wrong inside the appliance. This is a preferred outcome (as opposed to sending said accidental electricity through your body to the earth).

Electrons go from the panel, to the appliance, do some work and then back to the panel on the neutral leg. This is vastly simplified, OK?

Electricity always wants to get back to ground, or earth, and will always take the path of least resistance. Should there be a short inside an appliance, the chassis of said appliance could become energized. If you touched it and happened to be providing a good path to ground, that current will flow through you. Since it only takes about 20 milliamps to stop your heart, it’s very possible that a short like that could kill you. This is why we never solve ground loops with “cheater plugs” that effectively lift the safety ground.

But I’m getting ahead of myself. In an ideal world, the ground wire in a circuit would have zero voltage potential. That is to say that if you put a voltage meter on the ground wire and tested for voltage to a “true” ground, you would get zero volts. In complex electrical systems like we might find in a church or theater, it’s possible to have a few volts of voltage potential on the ground bus (all the ground wires connect back to a ground bus, which is connected to a ground rod—at least they’re supposed to be).

In a large enough system, you could have multiple ground busses and grounding points, each with slightly different voltage potentials. Now, we’re only talking about a few volts here, and for most equipment attached to this electrical system, it’s no big deal. However, in the world of sound, it can cause havoc.

When we have voltage potential on one ground leg (in this case, the one of the right), it travels from one chassis (mixer) to the other (amp) via the balanced audio line, finally settling at the lower voltage potential of the earth, inducing a hum. Also simplified to illustrate the concept.

Remember that we send audio signals to and fro with balanced lines. A balanced line consists of a high lead, a low lead (the signal carrying pair) and a shield. The shield ultimately gets connected to the chassis of the audio equipment, say an amp or a mixer. Also connected to the chassis is the power ground connector.

When you connect to components with a balanced line, you are essentially connecting their chassis together. Now image that instead of both pieces of gear being tied to the same ground bus (like we always tell electricians to do, and sometimes they listen), they are connected to two separate ground busses. Imagine that one ground bus has a few volts of voltage potential and the other one does not. Plug that XLR cable in and you now have current flowing over that cable between the amp and mixer.

What happens when you send 60 cycle current next to an audio line? You hear a 60 cycle hum. What happens when the shield of your balanced audio cable is carrying a few volts of 60 cycle current? A really well defined 60 cycle hum.

Some think the easiest way to solve this problem is to decouple the two electricity ground busses by using a “cheater” plug on either the amp or the mixer. Cheater plugs disconnect the safety ground connection, and in some instances will solve a ground loop. However, it’s very dangerous and you should never be do this. Will do so kill someone every time you do it? No. But do you want to be the one who eliminated the safety ground when someone does get shocked or killed? Me neither. Don’t do it.

The right solution is to make sure all your audio equipment is grounded to the same ground bus. Electricians will argue about this sometimes if they don’t understand why, but don’t relent. As part of a proper system, all your audio equipment should be powered from the same panel (that is preferably powered by a ground isolated transformer). This is why doing the electrical work in a properly designed install can be a little more expensive than just running lights and outlets; all the conduit runs have to go back to the same spot rather than wherever it’s convenient. But I digress…

Now, if you can’t re-work your electrical grounding system, and you still have a ground loop, there is still a temporary solution. You can lift the ground on the balanced audio line on one end or the other. Most of the time, this will stop the ground. Doing this can theoretically leave you open to RF and EMI noise getting into the line as the shield is not shunted to ground on both ends, but in practice it’s usually a better alternative to the 60 cycle hum.

Sometimes ground loops will show up when connecting things like keyboards to a mixer using a DI. Better DIs will have a ground lift switch that disconnects pin 1 (shield) inside the DI and will usually break the ground loop. If your DI doesn’t have that feature you should get new DIs. If you can afford that, you can make up a short XLR cable that has Pin 1 open at the female end. Make sure you clearly identify this cable as a ground lift.

If you need to lift a ground, do it in the audio cable (or with a DI). This method is much safer.

There’s so much more I could say about ground loops, but this is already getting long. And to all those sticklers for detail, yes I know I glossed over some points and over-simplified things to make the concepts understandable. If the concepts in this post are new to you, I encourage you to spend some time researching them. There is a ton of information available on the web; start Googling ground loops and spend a few hours educating yourself. Your efforts will be rewarded.

#### Today’s post is brought to you buy Heil Sound. Established in 1966, Heil Sound Ltd. has developed many professional audio innovations over the years, and is currently a world leader in the design and manufacture of large diaphragm dynamic, professional grade microphones for live sound, broadcast and recording.

1. j-mould@hotmail.co.uk

Thanks for the post Mike. There's another church sound problem solved 🙂

2. j-mould@hotmail.co.uk

Thanks for the post Mike. There's another church sound problem solved 🙂

3. mike.katrina.allen@gmail.com

Thanks Mike, that was a really informative post! I knew a bit about ground loops but your post really made it click. I sometimes get ground loops at home from a coax cable running into a stereo. There must be some voltage difference between my house's ground and the cable company's.

4. mike.katrina.allen@gmail.com

Thanks Mike, that was a really informative post! I knew a bit about ground loops but your post really made it click. I sometimes get ground loops at home from a coax cable running into a stereo. There must be some voltage difference between my house's ground and the cable company's.

5. James B

Mike, as an electrical engineer and a technical driector your post is spot on. You must have some training in basic electrical theory.

MikeA, if you ground the cable splitter at the cable servcie entry point to the ground bar in your electrical panel it will make some of that noise go away. The cable system grounds are sometimes compromised at the street level.

6. Charles

This was a great post, and I cannot wait for the next ones! There are some really bad hums in my church's system and I am hoping to get them all on the same ground soon, even if we have to run new grounds to the panels. Anyways, great post. Thank you.

7. mbhsound@gmail.com

The majority of the power outlets on our stage run back to our A/V room, but on the back wall on the stage there are two outlets, one of which I decided to plug an amp into due to the stage setup, one problem… those two outlets are not tied to the A/V room…
I knew what ground loops were, but nothing like a real life example to help cognition.

I have to say though, this is probably the clearest explanation I've ever read/heard.
Great post!

8. mbhsound@gmail.com

The majority of the power outlets on our stage run back to our A/V room, but on the back wall on the stage there are two outlets, one of which I decided to plug an amp into due to the stage setup, one problem… those two outlets are not tied to the A/V room…
I knew what ground loops were, but nothing like a real life example to help cognition.

I have to say though, this is probably the clearest explanation I've ever read/heard.
Great post!

9. jliechty200@gmail.com

Those who wish to do further reading on this topic might want to include some material on the pin 1 problem. If you google "pin 1 problem," you'll find a wealth of information.

If manufacturers would actually connect the audio shield (pin 1) to the chassis, we'd have far fewer ground loop issues. This is the essence of the pin 1 problem, in that the cable shield is often connected to the audio ground in the circuitry of the equipment instead of the chassis. In this case, any voltage on the shield causes the audio ground to fluctuate, and thus adds a hum to the signal inside the equipment. This is why disabling the shield / pin 1 at one end of the cable will stop the ground loop in most cases.

The insidious nature of this problem is that it can occur even with well-designed power systems. Because the hot, neutral, and ground wires are randomly spaced and twisted within the conduit, small variable voltages can occur on the ground even in a system with a properly designed ground.

Ideally, you can buy equipment that grounds the pin 1 shield to the chassis, and use only that sort of equipment. Then, unless your power infrastructure is exceptionally horrible, you should have very few ground loop issues. In theory, there is no reason why a balanced connection shouldn't be able to reject a 60 Hz hum on the shield (or on the signal lines, for that matter!). In practice, well, uh…

10. jliechty200@gmail.com

Those who wish to do further reading on this topic might want to include some material on the pin 1 problem. If you google "pin 1 problem," you'll find a wealth of information.

If manufacturers would actually connect the audio shield (pin 1) to the chassis, we'd have far fewer ground loop issues. This is the essence of the pin 1 problem, in that the cable shield is often connected to the audio ground in the circuitry of the equipment instead of the chassis. In this case, any voltage on the shield causes the audio ground to fluctuate, and thus adds a hum to the signal inside the equipment. This is why disabling the shield / pin 1 at one end of the cable will stop the ground loop in most cases.

The insidious nature of this problem is that it can occur even with well-designed power systems. Because the hot, neutral, and ground wires are randomly spaced and twisted within the conduit, small variable voltages can occur on the ground even in a system with a properly designed ground.

Ideally, you can buy equipment that grounds the pin 1 shield to the chassis, and use only that sort of equipment. Then, unless your power infrastructure is exceptionally horrible, you should have very few ground loop issues. In theory, there is no reason why a balanced connection shouldn't be able to reject a 60 Hz hum on the shield (or on the signal lines, for that matter!). In practice, well, uh…

11. tonyschoborg@gmail.com

Thanks for the article Mike, great information.

I don't mean to be a stickler but your diagram of the ground lift XLR is incorrect.

Pin 1 – shield
Pin 2 – high (usually a white or red wire)
Pin 3 – low (usually a black wire)o

12. tonyschoborg@gmail.com

Thanks for the article Mike, great information.

I don't mean to be a stickler but your diagram of the ground lift XLR is incorrect.

Pin 1 – shield
Pin 2 – high (usually a white or red wire)
Pin 3 – low (usually a black wire)o

13. mike@churchtecharts.org

Tony,
Ooops! You are totally correct. Totally goofed on that one. Uploading a new, correct, version now.

Good catch!
mike

14. mike@churchtecharts.org

Tony,
Ooops! You are totally correct. Totally goofed on that one. Uploading a new, correct, version now.

Good catch!
mike