Not all hums and buzzes are equally bad. Honeybees hum and buzz, but they’re super-helpful creatures that are generally interested in being left alone and making honey. Wasps, like the one pictured above, are aggressive jerks.

Of course, this site isn’t about insects. It’s about audio, where hum and buzz mean problems. Unwanted noise. Blech.

I recently got an email from a friend who wanted to know how to de-buzzify (I just made that word up) a powered mixer. When you mercilessly distill what I told him, you come up with a basic truth that covers all of troubleshooting:

The probability of an effective fix for a problem is directly proportional to your ability to isolate the problem.

Solitude

The importance of finding the exact location of a fault is something that I don’t believe I can overemphasize. It’s the key to all the problem-solving I’ve ever had to do. It doesn’t matter if the problem is related to audio signal flow, car trouble, or computer programming; if you can actually nail down the location of the problem, you’ve got a real shot at an effective (and elegant) fix.

The reverse is also true. The less able you are to pinpoint your conundrum’s place of residence, the more likely you are to end up doing surgery with a sledgehammer. If you can’t zero-in on a root cause, you end up “fixing” a certain amount of things that aren’t actually being troublesome. The good news is that you can usually take an iterative approach. All problems begin with “this system isn’t working as I expected,” which is a completely non-specific view – but they don’t have to end there. The key is to progressively determine whether each interrelated part of the system is contributing to the issue or not. There are lots of ways to do this, but all the possible methods are essentially an expression of one question:

“Is the output of this part of the system what I expect it to be?”

So…here’s a way to apply this to buzz and hum problems.

Desperately Seeking Silence

Talking in depth about the exact electrical whys and wherefores surrounding strange and unwanted noises is a little bit beyond my experience. At a general level, though, the terminology of “ground loop” provides a major clue. Voltage that should be taking a direct path to ground is instead taking a “looping” or “circuitous” path. A common cause of this is equipment receiving mains (“wall”) power from two different circuits, where each path to mains ground has a significantly different impedance. There is now a voltage potential between the two pieces of gear.

Bzzzzzzzz….

You can also have a situation where two device’s audio grounds are interconnected such that there is a potential between the two devices.

Hmmmmmmzzzzzzzz…

Anyway.

The first thing to do is to decide what piece of equipment you’re testing against. Maybe it’s a mixing console. Maybe it’s an amplifier. Whatever it is, you are asking the question from before:

“Is the output of this part of the system what I expect it to be?”

Or, more specifically…

“I expect this device’s output to be quiet, unless an audio signal is present. Is that the case?”

To answer that question, you need isolation.

---

WARNING: At NO point should you do anything to disconnect the mains-power/ safety grounds from your equipment. It’s there to prevent you from dying if the equipment chassis should become energized. In fact, as a start, try to verify that the mains-power sockets you are using actually DO provide a connection to “earth.” If they don’t, stop using them until they’re fixed. You may even find that your noise problem goes away.

---

To get isolation, start by disconnecting as much as you possibly can from the DUT (the Device Under Test). Of course, you’ve got to have some kind of way to monitor the output, so that might mean that you can’t disconnect everything. As much as possible, try to ensure that all mains-power grounds offer the same impedance – if it must stay connected, and it requires mains power, get all the power to connect to the same socket. A multi-outlet power tap can come in handy for this.

Is the output what you expect?

If yes, then something which was connected to your DUT’s input has a good chance of being the problem. At this point, if possible, treat each potential culprit as a secondary DUT in turn. If feasible, connect each suspect directly to your monitoring solution. If the ground loop manifests itself, and the suspect device requires mains power, try getting power from the same tap that the primary DUT is on. If the loop goes away, you’ve established that the two devices in play were likely having an “unequal impedance to ground” problem. If the loop stays in effect, you can jump back up to the beginning of this process and try again, but with the gear you had just plugged in as the new, primary DUT. You can keep doing this, “moving up the stack” of things to test until you finally isolate the piece of gear that’s being evil. (IMPORTANT: Any piece of the chain could be your problem source. This includes cables. You may need to pack a lunch if you have a lot of potential loop-causers to go through.)

If you can’t get the buzz to manifest when adding things back one at a time, then you might have a multi-device interaction. If possible, work through every possible combination of input connections until you get your noise to happen.

But what if the output on the original DUT was NOT what you expected, even with everything pulled off the output side?

At that point, you know that an input device isn’t the source of your trouble with this particular DUT. This is good – your problem is becoming isolated to a smaller and smaller pool of possibilities.

Try to find an alternate way to connect to your monitoring solution, like a different cable. If the problem goes away, that locates the cable as the menace. If you’re switching the connection, and the noise remains with no audio path, then the monitoring system has the problem and you need to restart with a new DUT. (If you’ve got a mixer connected to an amp and a speaker, and a ground loop stays audible when the mixer-to-amp connection is broken, then the amp is your noise source.)

If you’ve tried all that and you still have the buzz, it’s time to try a different circuit. Get as far away from the original mains-power socket as you can, and reproduce the minimal setup. If the ground-loop goes away, then you may have a site-wiring issue that’s local to the original socket(s). If the problem doesn’t go away, it’s time to take a field-trip to another building. It’s possible to have a site-wide electrical problem.

If the loop still won’t resolve, it’s very likely that your DUT has an internal fault that needs attention. Whether that means repair or replace is an exercise left to the reader.

Hopefully, you don’t get to that point – but you won’t figure out if you ARE at that point unless you can isolate your problem.