Ground Hum in Church

[StradivariusBone]

So we have a worship center that is one of those large, gymnasium size rooms with multipurpose in mind. Sound booth is at the far end, opposite the stage and the amps are kept in a large storage room to SR. The amps are fed power from a junction box with 6 20A circuits, installed sometime after the original sound system was replaced for the current one. We use a tri-amp setup with two additional amps for four channels of monitors and two bridged mono amps for each side of the low end.

On stage there are several amps (bass, 2-3 guitars), two keyboards, etc. The bass amp is DI, the guitars have 57's in front of them. Keyboards are DI, no amps. Power wise- I assume these all are fed from the same transformer which probably feeds the HVAC as well as a giant walk-in freezer/fridge.

Now, I don't profess to understand the intricacies of building electrical design, nor fully understand how to tackle grounding issues (I doubt I'd be posting if I did), but we do have a bit of a ground hum throughout the system, primarily through the mains. It is not noticeable if we drive the preamps up on the individual channels and keep the master in the -10, -20 db range.

One very strange thing I've noticed is that we have a floor mounted TR terminal for one of our monitor outputs. There happens to be a 90 degree plug that runs from that output to the speaker. One day the side of the plug was contacting the faceplate on the terminal and lo and behold there was 60hz through that channel of monitors. Nothing was powered on at the moment.

The main reason I'm posting now is that a few parishioners who are in the band have been making noise about buying power conditioners for the equipment on stage. Every other piece of gear is behind a Furman on a sequencer except for the stage stuff. We've recently added some LED lighting and the power distribution is a mess currently, so I'm all for a couple of conditioners to at least divide what's going to lights vs. audio in some way. What I don't think this will do is solve our ground hum problem. I've heard a lot about what conditioners do, and Furman's own explanation leaves a bit to be desired. What is "noise" when referring to an AC line? Is it random oscillations in the sine wave? Does the conditioner smooth out the sine wave? If that's the case then it wouldn't do anything to affect the ground loop (not that I want to eliminate a ground from a power cable- because I don't).

What are your tricks and tips for eliminating mains hum?

 

[Senorfish]

Noise in a sound system is a difference between the 'reference to ground' amongst the various parts of the system. Whatever the actual ground voltage is (hopefully not much!!) the equipment treats that as "zero". But if different pieces of equipment have slightly different "zero" reference points because of little bits of stray current floating around... then you get hum.

Study up on ground loop issues. Also check for easy things like flourescent lights, lights on wall plate dimmers, does it happen when Air Conditioning is on or off, etc.

Do not use those cheapo little ground lift electrical plug adaptors to 'lift' the ground on the guitar amp, bass, anything really power cable. You could potentially electricute a musician or singer when they grab their instrument.

Best solution is to have a dedicated panel for just the sound system in the building. It should not share the nuetral or ground circuit with any other panels in the building. But this requires extensive work. Another option is to bring in a isolation transformer and make your own independent neutral circuit. But its a big piece of equipment and is generally used when you're running power off a 'service disconnect' usually so you can seperate sound power from lighting power.

The easier immediate solution besides looking for flourescent lights and such, is try using passive DI boxes on the worst offending audio cables. Also inspect the internal wiring of beat up mic cables and patch cables to look for frayed or backwards wiring. Oh last thing and this is important... your whole sound rig needs to be using balanced XLR and TRS connections. long runs of like a guitar cable with only 2 conductors as opposed to 3 will act like an attenna and pick up hum. Use DI's to convert unbalanced lines (keyboard, guitar, etc) into a balanced cable (XLR) for the long distance run to the console.

 

[JD]

Most sound system hums are ground-loop related. First step is to identify which path the ripple current is taking. This could be between the board and power amps, between the board and effects racks, between the sound system and a direct connect on stage (like keyboards, or guitar direct box, etc.) Start by working backwards from the power amps. If you pull the input jacks on the power amps does it go away? What about between the board and active crossover? Divide and conquer. The fact that you can crank a channel and decrease the gain on the amps and minimize the hum sounds like it is somewhere after the board or between the board and what is next in the chain. Ideally, the board should be powered from the same source as the power amps, and share a SINGLE POINT ground at the location of the amp rack. In practice, usually the board is plugged in an outlet far away and therefor any small currents found on the ground path end up superimposed on the snake that loops back to the stage.

 

[AlexDonkle]

Most sound system hums are ground-loop related. First step is to identify which path the ripple current is taking. This could be between the board and power amps, between the board and effects racks, between the sound system and a direct connect on stage (like keyboards, or guitar direct box, etc.) Start by working backwards from the power amps. If you pull the input jacks on the power amps does it go away? What about between the board and active crossover? Divide and conquer. The fact that you can crank a channel and decrease the gain on the amps and minimize the hum sounds like it is somewhere after the board or between the board and what is next in the chain. Ideally, the board should be powered from the same source as the power amps, and share a SINGLE POINT ground at the location of the amp rack. In practice, usually the board is plugged in an outlet far away and therefor any small currents found on the ground path end up superimposed on the snake that loops back to the stage.

Agreed, ground loop is the most likely cause and the location needs to be tracked down. Ground loops are really just 60Hz AC current traveling along the outer shield of an audio cable, from one device to another. If the audio cable connecting these 2 devices isn't also carrying a DC voltage (i.e. 48V phantom power for microphones, or intercom / IFB cables), then you can disconnect the shield at one end of the cable, typically disconnecting the shield at the output device is best. This should stop a ground loop, and doesn't involve lifting the power-ground pin which, as mentioned, is very dangerous.

 

[themuzicman]

So first and foremost -- don't waste your money on power conditioners. They are a giant waste of money and don't do anything. In practice, they are just rack-mounted power strips with lights on the front.

The ideal would be a service that has a separate ground run just for audio, that has an isolation transformer between the switch and the PD.

As stated by others, you need to find where the grounding issues are coming from -- you mention that the HVAC and Walk-In share the service. This will complicate your search, as the offenders could be coming on only once in a while. Don't discount that it could be within your system and your system alone however. A ground loop hum occurs when there is difference in potential between the ground of two different devices, and your only way to fix it is to either a. isolate the offending ground or b. get everything on the same ground with the grand goal of having everything on an isolated single ground!

That being said, my guess is the same as JD's and it is in your amps.

Find a day when you can have the building alone, and you can flip breakers for other devices, should it come to that. Start with everything on that is normally on during a service both in the building-at-large and in your room, including bass cabinets and guitar amplifiers. You don't need to care about microphones though. Get every point that is its own circuit powered on in the system -- which I'm assuming is a. FOH b. Ampland and c. Stage. Do something simple at stage like turn everything on, and plug a CD Player in on stage and output it into your snake. Go to the console gain it up appropriately. Can you make the ground hum louder in the system via this channel? If yes, the problem is your console or before it. If no, the problem is after the console. Now keep that input rolling still, go to the amps and click them up a few clicks. Did the ground hum get louder? The problem is in your amps.

--------------
Now, a whole other thing to consider is if your gain structure is set up correctly. Without having ears on the issue, I want to suspect that you may just be hearing noise floor hiss as you have said you are keeping your Stereo Bus at -20 to -30dB and gaining up the channels to hide your hiss. This sounds a ton like you are making up for a portion of your system that is gained up entirely too much and injecting noise floor noises into the speakers. Go dial your amplifiers back all the way. Now go throw the Master Fader/Stereo Bus on your console to 0dB Unity Gain. Plug an iPod into the desk, or have someone speak on a vocal mic -- throw the input channel to 0dB, make sure your input is set to Line Level (if an iPod), keep the gain down, and now click up your Amplifiers one click at a time until you have enough gain out of the system. Now throw a vocal mic into a channel, throw the fader to -10dB and talk into the mic -- is it loud enough in the room for conversational level talking audio? If yes, you're good -- and you even have headroom on the fader to throw a dynamic mix. If it isn't, click the amps up another click or two. Hopefully this will get your hiss under control.

 

[Senorfish]

We used to do concerts in a 300 seat venue running everything off of a single 20amp circuit. Bi-amped PA with subs, stage, 6 self powered monitors, and FOH. Since everything was plugged into the same outlet, we didnt get any ground loop hum (unless someone rented in some cheap pole mounted NSI or Leprecon 4 channel dimmers). Yes we were pushing the limits of the 20 amp circuit... but it would run as long as the bass and kickdrum level into the subs was kept to a moderate low level. During louder bass notes though you could see the voltage meter taking a bit of a dive to 105 volts or less.

In another venue we'd run everything off a 30amp circuit and that provided better amperage headroom for heavier bass. Outdoor concerts we'd have a 60amp feed for the whole rig.... and would only pull about 30amps. This was with modern efficient QSC amps though. Old antiquated amps will have a much higher need for juice.

The setup would usually be drop the amps next to the stage, run musicians off extension cords to the same power conditioner, and then run a heavy 12 gauge 100' power cable from stage to FOH, all off the same source.

 

[FMEng]

I agree with Mr. Donkle that the easiest way to solve ground loop isssues is to:
1. Make sure that all interconnects over any distance are truely balanced, and
2. Drop the shield at one end of the line-level cable.

If that doesn't solve it, more severe ground loops can be resolved by inserting a high quality audio transformer in the audio line. Note the word "quality." A transformer can work miracles.

NEVER try to solve a ground loop by cutting ground connections on an AC plug. To do so is a potentially deadly safety hazard.

The only thing that any "power conditioner" might do is reduce the severity of spikes and surges from something like a nearby lightning strike. They do absolutely nothing for the typical malady of large audio systems, which is a ground loop. Power conditioners, like Furman's, are more marketing hype than anything.

As for special, isolated power for audio systems, fine if you can do it, but there is no necessity in doing it. This has become legend because it can prevent problems, but in most cases those problems could have been resolved by much simpler and cheaper methods. I have built many large radio studio facilities without anything fancy for power and I never have ground loops. Why? Because we follow 1 and 2 above.

 

[StradivariusBone]

I should have clarified, I understand well-enough that it's a ground looping issue. I was more or less trying to get an answer on what a power conditioner actually does beyond voltage suppression and cable management.

I'm pretty confident that the loop that we are hearing (since it's persistent through the main out and doesn't change when channels are muted or not) is as you all guessed between the board and the amps. The board outputs to a DBX DriveRack PA, which I don't think it has a ground lift switch on it, nor do I think the the outputs on the mixer have that option. Are you suggesting I detach pin 1 on one side of the cable? If I do that, I'm guessing I'd want to label it for future reference.

Occasionally we'll get noise from the bass amp/keyboards, but those are all lifted at their DI's. That's always been transient (and probably cable issues from my guess). We're in Florida and the humidity does have an effect on grounding when it's really wet or really dry. The other question I had revolved around the ground hum from the cable grounding on the conduit without anything turned on. Does that imply a potential between building ground and the amp ground driving that monitor?

I'll try the gain structure @themuzicman suggested. I dial in gain on most things so that it's averaging about unity with an occasional peak above on the VU meter. Once it's there I usually find the faders at unity to start. Low and Mid amps are wide open, highs are notched at 2/3 input gain. The main faders then dial back the overall mix to a level that's not insane. I'm interested to see how it sounds running the amps a bit lower and the board hotter (at least from the main out). Not getting any hiss from the system unless the main fader goes above unity.

Lastly, I never advocate pulling a ground pin. Anytime I find a cable without one I introduce it to my good friend, Mr. Diagonal Side Cutter. I was zapped once playing bass due to an internal grounding failure in the amp head. Thankfully the ground pin was still in the power cord.

 

[JD]

Is your active crossover at the amp rack or board end? If it's at the amp rack end, then I would look for a good one-to-one, line level transformer to isolate the signal. Never liked disconnecting the shield line on active input (transformerless) equipment as there is always the possibility of spiking the circuits. That being said, most ground loops are measured in millivolts, not volts, so there is not much potential to get rid of. If the crossover is back by the board, then you have a few more transformers to buy! Also, if the monitor feed is from the board, and the amps for the monitors are in the same rack as the mains, you will need them on those feeds. Best bet, get one and disconnect everything except one line on the mid feed or low feed. (Whatever one you can notice it best on.) Try putting the transformer on that line. If the hum goes away, then you know it's worth the effort to isolate the rest. One other thing to look at... It there any frame connection between the stage box end of the snake and the rack frame? (Seen a couple of rack mounted stage boxes in my day.) If so, mount that puppy in a different rack, or isolate it from the frame. Speaking of remounts, many systems have the crossover at the board end and it never gets touched. If this is the case, consider moving it to the amp rack as this will radically cut down on the number of transformers you need to buy.
One last thing to check- Are the lines from the board to the amp rack truly running balanced? Sometimes a shortcut is taken here or there and they look balanced but are actually running unbalanced. A quick check with a meter should solve that question.

 

[StradivariusBone]

Crossover is at the rack so thankfully only two lines of main running in that direction. Four channels of aux-fed monitors though might mean a bit more investment. I might try dropping the ground on one run to verify that it makes a difference, then push for an iso transformer. From what I've read, they seem to function similar to the optisplitters for DMX, just in an analog format. A perusal of B&H, Full Compass, et al seems to indicate that like all things audio it appears there are some favorites and then some not-so-favorites. Any preferences as we dive in?

The snake(s) are two 12x4 fantails with a stage box and then the older building wiring that arrives under the stage in a screw terminal backboard. 3-4 of those older runs have shorted and we normally survive off the two snakes for most things, but I wouldn't be surprised if it was touching building ground or conduit- thinking about that now, that might even be where our monitor runs come through. I'll have to investigate that too, but probably not much isolation possible on that end, save from just eliminating those runs from the board altogether- which might not hurt anything. I think we only use one of those for a mic occasionally, the building was originally designed for traditional style worship and has since been retrofitted (on occasion) for the past 15 or so years.

I'm about 98.2% positive that we are balanced front to back, but I leave that 1.8% open since having worked audio in a church setting (amongst other places) for the better part of 20 years I have learned that through God all sorts of wonderful and sometimes stupefying things are possible- unbalanced LR runs, wireless crosstalk from guitarists' unannounced Christmas presents, vocalists kicking a mic input causing a short to ground loudly, digital player piano coming to life in the middle of a prayer...

Thanks for all the input!

 

[rwhealey]

Are you suggesting I detach pin 1 on one side of the cable? If I do that, I'm guessing I'd want to label it for future reference.

http://www.sweetwater.com/store/det...e=&network=g&gclid=CLTgktWqlMMCFRE1aQodl2gAlQ

This will do the trick and will keep all your cables intact.

While the comments here have been pretty good, as evidenced by your musicians running out looking at "power conditioners" there are a lot of power myths out there in the sound world. If you really want to learn about sound system grounding pick up a copy of this book: http://www.amazon.com/dp/0240802861/?tag=controlbooth-20

 

[Robert]

I suggest you take your power for all audio on the stage from the same outlets you use for your amps. The company I work with takes their power for audio from one source and distribute it to each location. Front of house (mixer, processing, etc.), stage gear (amps, instruments, monitors), and their power amps all have the same power source. This is usually a pretty easy thing to accomplish with a few extension cords.
Check to see what else in powered by that same panel. Look for odd loads on the panel, this could be dimmers or anything with a compressor (refrigeration). I did an instal that worked fine in the day, but in the evening went to crap. Turns out the daiquiri machines were on the same buss and puked all sorts of crap into the system.
After that you get to some real difficult troubleshooting. There could be a component going bad or some bad wire. You can try taking each wire and piece of equipment out of the system one at a time. Might work, may not.
Cutting the ground pin in your cables is not the best direction to head in.

 

[Bubby4j]

Another thing you can do (potentially without buying anything) is using 2 passive DI boxes to lift the ground. XLR > DI Box (ground lifted) > TS Cable > DI Box > XLR.
May result in quality degration, especially since you'll have a short length of unbalanced, however you can try it. It's worked for me in the past. Dunno if this is advisable though, someone correct me if I'm wrong.

 

[derekleffew]

A Pin1Lift seems much simpler:

, or build your own AND LABEL THE HECK OUT OF IT!

See also
http://www.rane.com/note110.html

The best end to disconnect [the shield] is the receiving end.

and
http://peavey.com/support/technotes/soundsystems/humandpin.cfm

In this case, pin#1 should be lifted at the mixer position.

So should the Pin1Lift go (or cable with shield at one end disconnected be) at the console or amplifier? Since the experts disagree, I say, whichever solves the issue.

 

[Amiers]

I had issues with this as well. But Derek is right a ISO is your first step i popped two in on my sound console and it was right as rain. Start it simple.

 

[FMEng]

One last thing to check- Are the lines from the board to the amp rack truly running balanced? Sometimes a shortcut is taken here or there and they look balanced but are actually running unbalanced. A quick check with a meter should solve that question.

It isn't just the wiring. The cable and connectors between the console and the rack could be wired balanced, but if the output stage on the console is un-balanced, or if the input stage at the amp rack is un-balanced, you have an interconnection that cannot reject noise. Unfortunately console aux sends are notoriously un-balanced, and manufacturers love to bury that fact deep in the manual. Often, the only way to tell is to have a look at the block diagram.

 

[FMEng]

Another thing you can do (potentially without buying anything) is using 2 passive DI boxes to lift the ground. XLR > DI Box (ground lifted) > TS Cable > DI Box > XLR.
May result in quality degration, especially since you'll have a short length of unbalanced, however you can try it. It's worked for me in the past. Dunno if this is advisable though, someone correct me if I'm wrong.

That can work, especially when desperate, but there are some caveats. First, the DI box has to be passive. It won't work with a FET design like a Countryman. Secondly, you are using transformers designed for low level signals, not line level. When a transformer saturates from too much signal, there will be added distortion and lost bass response. A handy trick in a pinch, though, because the hum is usually worse.

 

[JD]

That can work, especially when desperate, but there are some caveats. First, the DI box has to be passive. It won't work with a FET design like a Countryman. Secondly, you are using transformers designed for low level signals, not line level. When a transformer saturates from too much signal, there will be added distortion and lost bass response. A handy trick in a pinch, though, because the hum is usually worse.

Good diagnostic tool! Set the pads at "0" and give it a try. I want to reinforce that part about signal level- Although DI's can take a line level input, the pad is before the transformer. The transformer itself is usually not capable of passing a hot signal. Everything may sound fine during the check, but when you get to the performance levels it is cringe time!

I want to add something about the "pin one lift." I am opposed to it, but it will work. The reason I am opposed is that as a servicer, I have had to repair the differential input circuit on many a power amp. As long as there is not much voltage at play in the ground loop, no damage will be done. But, if there is a significant voltage there (think Hum Switch on an old Fender amp) it will pop the input circuit. In this case; where the board is grounded, the rack is grounded, and there is a minor loop, the pin one lift should work fine.

 

[StradivariusBone]

It isn't just the wiring. The cable and connectors between the console and the rack could be wired balanced, but if the output stage on the console is un-balanced, or if the input stage at the amp rack is un-balanced, you have an interconnection that cannot reject noise. Unfortunately console aux sends are notoriously un-balanced, and manufacturers love to bury that fact deep in the manual. Often, the only way to tell is to have a look at the block diagram.

The board in question is a GL2800 A&H, the manual does indicate that aux sends are balanced. Something I just thought about is the possibility that there is a TR cable in that chain, each run goes to a Behringer graphic EQ and then to the long run to the amps. On the LR run there is a "sonic maximizer" in-line (please don't judge, it was out of my hands) that mysteriously has remained in the bypass mode for some time. I think it might be worth it to remove it altogether just to test the theory.

So gleaning the information from this thread-

-Check all interconnects to ensure balanced runs from top to bottom.
-Isolate LR and pull pin 1 (try both sides) and observe results.
-Isolate each monitor run, pull shield (on both sides) and observe results.
-If pin lift works, create short piece of xlr/trs to lift shield on whichever end is necessary
-If pin lift works, but sound quality suffers explore line level isolation transformers in-line (ultimately this seems to be best solution, provided that the initial tests prove that it's the issue).

It might even make sense to use a DI after our bass amp's internal out. That thing always has hum issues and as I said before it has had internal grounding problems.

One last question- I know phantom power requires a common to function, but does lifting the shield have any effect on the common mode rejection of the pair?

Derek- those are great links, btw. Definitely bookmarked!

 

[JD]

Phantom power does present a problem on ground lift as the phantom power is the potential between the signal line (2&3) and the shield on pin 1.

Most times, noise on a phantom is caused by leakage from one of the signal lines to ground/shield due to moisture. Common mode rejection rules do apply otherwise.