Elimination of rather nasty hum?

proxy.php


The above circuit is what I am using to convert a stereo signal into mono before going to XLR (the only really viable option just at the moment - not using two D.I. boxes etc).

The 3.5mm stereo DIN minijack whatever-you-call-it (somebody tell me :p) is connected to a laptop output, and the XLR goes up to our booth via multicore.

My question is this: how can I eliminate the nasty hum I am getting? What would be causing it?

The "converter box" with the resistors/XLR jack etc in it unfortunately happens to be close to our dimmers, if that is a factor. This cannot be changed - if I fade lights up, there is a fluctuation in the hum.

Help please!!! Thankyou all in advance. :(
 
Last edited:
If your sound system and lighting system are coming from the same power source, you'll get a hum. We had a hum problem recently coming from a computer (we used a minijack to rca to quarter inch through line matching transformer to XLR series of adapters), and we popped some ground lift cables (we have cables with ground lift built in), and the hum dissappeared. It worked better than the direct boxes we were using.
 
Firstly, I don't think there's anything wrong with your cable, but laptops do enjoy humming as much as is possible. This is caused by threethings in my experience:

1.) The laptop's sound card. Horrible, regardless of who made the machine.
Get something external and hi quality- M Audio's FastTrack USB runs on PC and Mac and is absolutely fantastic.

2.) Otherwise, Electricity. Depending on the venue I'm in, sometimes I need to run my laptop on battery because plugging it in to power causes and really annoying hum. To this day I don't know how to get rid of it, but I think it's probably due to the combination of gear used and the venue's usually dodgy power supply. Try unplugging the power cable and see what happens?

3.) Finally, given that the hum changes when you dim lights, maybe something is plugged into the wrong breaker somewhere, or crossing lines, maybe even somewhere past the DB, as in my venue until recently- headaches!

Good luck!
 
Last edited:
Have you tried lifting pin 1 on the XLR? I highly reccomend that you do NOT lift the AC ground, people are killed that way. But try lifting Pin 1 on the XLR, see if that helps any.

Wait a minute, just thought of something. You have a left/right source comming out of the laptop. That's summed into a mono signal, but by your drawing, that's split back out to pins 2 & 3. That would make pin 1 the return/neutral to the device. That doesn't make sense. Wouldn't you have the so-called "send" going to, say, pin 2, while the "return" would be on 3, leaving pin 1 as the shield? If my logic there is wrong, please correct me.
 
Last edited:
I'm also wondering if an addition of a diode to the wiring of the device might not eliminate so possible issues with feedback, < electronic not audio> back into the outputs of the sound card.
I agree that using some external, possibly USB, device might not also help, Getting a clean signal in the first place is an issue.

Quote "Eboy87I highly reccomend that you do NOT lift the AC ground, people are killed that way.

I've never heard of anyone being killed by lifting the ground on Audio equipment, but I guess it could happen. Ground lifting and Isolation is probably the most common remedy for eliminating issues ariseing from "ground Loop" issues. This particular problem sounds a lot like a ground loop issue, or possibly a revesered neutral somewhere in the room causeing a feedback loop to your mains power supply.
The fact that the converter box is located next to your dims is most likely the culprit though. Assuming two things here; "the box" is not sheilded and you built it yourself, I would suggest doing the following.
A. go back in the box make sure all solder connection are clean and good.
B. make sure wire leads are as short as possible.
c. Make sure leads on ends of resistors are as short as possible.
d. If the box is plastic;
I. Get foil duct tape and cover all surfaces inside the box making sure to ground the foil to the "case" of all your connectors. < make sure there is a connection between the outside of the xlr jack and the "outside" of the 1/8" mini jack.

II. Make a Faraday sheild around the out side of the box using screen wire < just wrap it in screen wire steel not plastic> make sure the screen is grounded to an outlet or conduit.

Most likely your hum is inductive, if your that close to the dims. Fist rule in sound, "Never get your cables close to power cables.". No, wait, that's the second rule of sound, first rule of sound is " Everyone in theatre knows two jobs thiers and sound."
 
Yes, lifting the AC ground can kill people. Think about it, the ground is there to provide the shortest path between a fault and the ground. Let's say you have a console that shorts out. THat's 120V, 20a that's now running through the chassis. If there isn't a ground connected (read, if you lift the ground with one of those 3-prong to 2-prong cheater), it has no quick path to earth. Now let's say the engineer leans on the console, he is now the shortest path between the board and ground, letting that current flow through him and electrocuting him.

Remember the pastor in Texas who was electrocuted? The cause was a faulty ground. Electricity can kill. If you don't beleve me, search over at ProSoundWeb. Never mess with electricity if you don't know what you're doing.

Now, you can lift the "ground" on Pin 1 of a balanced line run, as long as the rest of the equipment isn't properly grounded. That may be where we are confused. In the drawing, the shield is labeled "ground". My warning was referring to lifting the AC ground, not pin 1.
 
Last edited:
Chris I'll have to see if I can keep my legend title and help solve this problem.

First I see two distinct problems:
1) The current wiring
2) The hum

1) The wiring
I don't see why the wiring from ranes 13 didn't work. You didn't connect pin 1 to the ground at the splitter did you? I would double check the wiring. Pin 2 should be connected to the output to of the two resistors. Pin 3 should be the ground from the circuit but Pin 1 should only be connected to the screen of the mic cable at the mixer end. At the other end there should be no connection between the circuit and pin 1. Make sure the shield connector on the XLR has not been wired to Pin 1.

But if you have tried all this and it didn't work ignore it.

You said a DI box wasn't an option but if you look in the May 2006 isssue of Silicon Chip magazine it shows a simple passive DI Box. It even lets you connect in two resistors so you could use it as the combiner and and DI box in one package. The transformer they are using from Altronics only costs $17 then you need a switch,XLR socket , a couple of jack plugs, and a metal case. As a minimum you would just need a jack plug, xlr socket, transformer, wire, metal case. You can buy the article from Silicon Chip website although you can probably find it in a library.

2) The hum
The other posters have made good points about the hum.
The first thing I would want to is establish the source of the hum.
I would turn the packs completly off and see if I get any hum, in case the laptop is producing some. If it is, try, as someone suggested, running on battery to see if this improves the problem.

Once that is sorted try the dimmers, which are probaly the main source, set all the dimmers to about 30% that should have a fairly high level of interference generated. Then move the laptop and cables around in the space to see if you can minimise the interference. Check the leads aren't running parallel to the dimmers or power cables. If they have to cross power cables do it at right angles. You may need to add an extra lead so you can move as far away from the dimmers as possible.

Hope this helps
 
Yes, lifting the AC ground can kill people. Think about it, the ground is there to provide the shortest path between a fault and the ground. Let's say you have a console that shorts out. THat's 120V, 20a that's now running through the chassis. If there isn't a ground connected (read, if you lift the ground with one of those 3-prong to 2-prong cheater), it has no quick path to earth. Now let's say the engineer leans on the console, he is now the shortest path between the board and ground, letting that current flow through him and electrocuting him.

Remember the pastor in Texas who was electrocuted? The cause was a faulty ground. Electricity can kill. If you don't beleve me, search over at ProSoundWeb. Never mess with electricity if you don't know what you're doing.

Now, you can lift the "ground" on Pin 1 of a balanced line run, as long as the rest of the equipment isn't properly grounded. That may be where we are confused. In the drawing, the shield is labeled "ground". My warning was referring to lifting the AC ground, not pin 1.


Since I lit my first show about seven years before you were born I think I know something about electricity.
Using a ground lift is at least a temporary solution for diagnosing a problem and at most an emergency solution for a fix on the fly.
I don't remember the story about a pastor in texas that was killed by a microphone perhaps you could provide a link to the story ? Since the "power" side of an audio console is almost completely isolated from the "mix" side of the console I find it hard to beleive but hey people have been killed talking on the phone during a lightning storm.

Ok editing now I'll buffer it a bit, Eboy, I don't want to start some kind of flame war, but please, make sure you know who your'e talking to before making a statement like that. I'm going to assume that maybe you are putting your statement in for the benefit of others that might make the assumption that they can do anything they want with electricity cause they saw a guy do it once on t.v.
 
Last edited:
If I remember correctly, it was not a faulty ground on the audio side, but rather a water heater for the pool he was standing in. There was an 80v short circuit in the tub, so when he grabbed the microphone it found ground through him and the microphone.
But that doesn't mean the audio system itself isn't susceptible to being energized. If you lift the power ground on a piece of audio equipment, you do open the potential for 120v to fly around on the "mix" or microphone side of your sound system.
This is the same sort of problem with ungrounded backline equipment and the perfect example of why it's never appropriate to lift a power ground. Even doing it temporarily is akin to climbing without a harness just for a short bit.
If a ground lift is needed, there's a larger problem at hand which isn't being addressed. As van said, it can be a temporary solution, but I don't think it should be. Try using iso-transformers or running power extensions from FOH to stop ground loops.
 
Last edited:
Van, I didn't mean to sound condescending (sp?), I'm sure you have way more experience than I do, and I'm not debating that fact. I just wanted to point out the dangers to others, since most of the people on here are high schoolers. I've had friends who've gotten a good zap from gound lifts, and I don't want others to go that same path.

Yes, and as much as I hate to admit it, it can be a quick, temporary solution, but there are better solutions out there, especially in the audio world, as jbeutt pointed out. The last I heard about the pastor was the heater wasn't properly grounded, and the current found it's way through him instead of the ground. It's not the best example, but it still illustrates the point.

Like I said, I just want to point out to the younger crowd that electricity isn't something to be trifled with, it is deadly. I'm sure you and the other older heads know that better than anyone, but there might be those on here who don't know, or have the "it won't happen to me" attitude. Cheating on a test may work sometimes, but cheating with power has deadly consequences.

Sir, yes, I do know who you are, it says so in your sig, and yes, I do know something about electricity as well, and how to wire things the correct way. I may not be working in the field as of yet, but I'm no moron either.

Please excuse me, it's been a very, very, very bad day for me so far, so I don't mean to sound like the county ***hole.
 
Last edited:
Couple of points to consider

The major problem with ground lift on the ac side, is that most three wire devices do not have any double insulation setup like most Consumer electronics gear with simply a two wire setup. SO what you can have is in case of a problem a ac signal on the neutral could be passed directly to the performer. The main problem is that mics are typically using the case as a ground, so you could very well have serious voltage present on the mic case if there was a problem. So lifting a ground on the ac side can be quite dangerous.

The other problem that I have found is that just because you have a three prong you do not automatically have a good ground all the way back to the building ground, outlets might not have a continuous ground wire, might be using the conduit as the ground and a loose screw in the conduit connection can have a faulty ground.

This is why in a typical pro setup, you use a power distro wired directly into the panel so that you KNOW what is in the system. I have seen situations where someone say, this outlet is fine, it is the only thing on the breaker only to find out that that is wrong, and that the coffee maker is on the line and you have all sorts of problems, the least of which could be tripping the breaker.

IN general you will save your self endless amounts of problems if you ALWAY use a transformer isolated connection to connection between a balanced signal path and an unbalanced signal path. Believe me when I say that you will be driven crazy trying to chase down these problems.

SO it does not have to be an active DI box, but a Line isolated transformer is IMHO essential. So there are plenty of devices around at various price points that convert back and forth between balanced and unbalanced. Personally I would take the left and right signals, add the resistors like shown, and then put the signal into a transformer that decouples it from the xlr connection.

These solutions are down in the under 50 dollar range

Sharyn
 
Eboy Truce ! I think we had the same day ! :oops: Teched a show tonight, haven't seen the props person in 3 days and they showed up tonight and still don't have all the real pieces, had my floor for Inspecting Carol get stuck last night and had to crawl under it and spend and hour and a half today in an 18 inch crawl space Arggghhhh !
Ok just for discussion, looking at the wiring diagram, I am intrigued about the lack of a "backflow" system. tieing both channels < the left and right outputs> without isolating them via a diode would seem to me to be an invitation to blowing out the op amp on the sound card of the laptop. I s my thinking here wrong ? I would think that in the event of one channels signal overiding the other < say you panned everything right for example> that the resultant lack of signal on the other channel could cause it the left channels driver circuit to over heat and blow. I'm not an electrical engineer, is somebody else here ?
 
This topic reminded me of a RaneNote I read a couple of years ago, so I went ahead and dug it up. It describes a similar, but slightly different circuit that might be useful. Check it out (scroll down to summing):

http://www.rane.com/note109.html

I'm not entirely sure why it works, and it's too late at night for me to work the circuit out, but it might at least get you started.
 
You can rule out the sound card on the laptop straight away, I have tested that.

Unplugging the laptop probably won't help as I still had the hum when using a battery-operated CD player.

I am sure it works fine without dimmers on, but will test.

Should I completely cut the shield cables on both ends of circuit? That's pin 1 on XLR and the exterior of the TRS. I mistakenly put "ground" on the diagram.

Eboy87 said:
Wait a minute, just thought of something. You have a left/right source comming out of the laptop. That's summed into a mono signal, but by your drawing, that's split back out to pins 2 & 3. That would make pin 1 the return/neutral to the device. That doesn't make sense. Wouldn't you have the so-called "send" going to, say, pin 2, while the "return" would be on 3, leaving pin 1 as the shield? If my logic there is wrong, please correct me.

I don't understand why that would be so.

Where would I put a diode and why?

@ Cutlunch: My original ciruit was done exactly according to your post http://www.controlbooth.com/forums/showpost.php?p=41147&postcount=8
However it would NOT work without bridging pins 2+3 of the XLR.
Is 10 000 Ohms resistance too much?

cutlunch said:
1) The wiring
I don't see why the wiring from ranes 13 didn't work. You didn't connect pin 1 to the ground at the splitter did you? I would double check the wiring. Pin 2 should be connected to the output to of the two resistors. Pin 3 should be the ground from the circuit but Pin 1 should only be connected to the screen of the mic cable at the mixer end. At the other end there should be no connection between the circuit and pin 1. Make sure the shield connector on the XLR has not been wired to Pin 1.

So there should be nothing connected to Pin 1 on the XLR socket on the input side - just plug an XLR lead into it and run to mixer without anything on pin 1?

Yes I did originally connect pin 1 to shield at the XLR socket, however there was no shield at the TRS jack end (as per Rane guidelines). This didn't work. When I bridged pins 2 + 3 I got a signal. I left it like that and it worked fine...until I turned on dimmers (I think). Then it was humming like hell. Finally, I reconnected the shield on the TRS end and it made no difference.

What is the difference between "ground" and "shield". The audio lead I'm using has four wires and a shield braid + foil.

As soon as I understand what I should be doing here I shall put up a different circuit for confirmation.
 
I looked at the circuit I gave you to convert the stereo to mono and I can see where some confusion could come. The designer talked about shields where we would talk about signal ground. Sorry I didn't pick up on this earlier but with a background in electronics I automatically made the adjustment.

I also looked at the rane note and see where you might have got confused. You are thinking about a stereo jack when no. 13 applies to mono. You do bring the signal from the laptop as a stereo feed. But after you join left and right channels together by using the two resistors it becomes a mono signal. And that is how you must think of it when wiring it.

Think of the output of the two resitors being connected to the red line on diagram no. 13. The black line is the sleeve from the jackplug.

If you wired it so that the sleeve of the stereo jack plug ( from the laptop ) is connected to pin 1 of the XLR and the output of the resistors to pin 2 & 3 you could be setting up a path for hum. To wire it the original way take the output from the resistors and only wire it to pin 2 of the XLR. Take the ground which is on the sleeve of the jackplug (ground) and connect it only to xlr pin 3. In the convertor connect nothing to Xlr pin 1. This should work and maybe lower the hum level.

If not we could try wiring it in another way.
In the convertor box at the laptop wire the XLR socket so that the sleeve from the jack plug ( ground ) is connected to both pins 1 and 3. Connect the output from the resistors to pin 2 of the xlr. This should work but won't necessarily get rid of the hum.

If the hum is really bad you will need some form of transformer isolation.

Ok the difference between ground and shield is:
The ground, in this case called signal ground and on unbalanced line it the path the signal returns on. Sought of like a battery circuit with a bulb current flows from one battery terminal to the bulb out the bulb back to the other battery terminal. That is an over simplification but it gives you the idea of a circuit flow.

The shield is meant to do what it says and sheild the audio signal from interfence. It does it by being wrapped around the outside of the signal wires . So when interference comes along it goes down the shield to a chassis ground. This ground is outside the signal path so should help minimise the interference. On the cable you are using you only need the shield and any two of the other conductors. Just make sure that at the convertor end the shield is not connected to anything. If the leads are premade XLR don't worry about disconnecting the shiel from pin 1 Just make sure the XLR socket on the convertor has nothing wired to pin 1. This only applies if you use the first wiring circuit ( not the one when when pins 1 & 3 are connected).

Let us know how you get on.
 
Take the ground which is on the sleeve of the jackplug (ground) and connect it only to xlr pin 3. In the convertor connect nothing to Xlr pin 1. This should work and maybe lower the hum level.

Just make sure that at the convertor end the shield is not connected to anything.

:neutral: :?:

Thanks cutlunch and everyone else for all the fabulous help.

Didn't you just say to connect the shield to pin 3 XLR at the converter end with nothing on 1?

The shield is currently connected to the sleeve of the jackplug (long bit) and is on pin 1 XLR. You're saying connect the shield/ground to pin 3 instead of 1 and put the resistor output to pin 2.
 
You guys are going to go crazy trying to get this to work, and IMO if it does eliminate the hum it will be only until something else in the system or environment introduces hum again.

Most of the techniques you are talking about work to convert Balanced to unbalanced, say going from xlr out on a mixer via a short length of cable to an rca connection on an amp. Not the best but done all the time and usually works IF you make sure you plug the amp and the console (or the two devices) into the same power/ground source.


IF on the other hand you try to create a balanced output from an unbalanced input such as you are trying to do, it will IMO not work without a transformer.
The because of how a balanced signal works based on differential reference to the signal pair and NOT the shield, when you try to "create this" without isolating the shield you have all sorts of hum problems. I have seen people spend hours pulling their hair out trying to find these problems, and have just pluged in a simple isolator and the problem goes away instantly.

An inexpensive two channel solution is the ART unit
http://www.musiciansfriend.com/product/ART-CleanBOX-Two-Way-Stereo-Converter?sku=180613&src=3SOSWXXA

You would have two signals conbined like the original diagram, using resistors that approximate the input inpedance, and then use the art unit to do the conversion to balanced.

The resistor network with the third resistor that attemps to creat a floating ground in my experience is unreliable, and still doe not give you any isolation.

Sharyn
 
Chris I'll have to see if I can keep my legend title and help solve this problem.
First I see two distinct problems:
1) The current wiring
2) The hum
1) The wiring
I don't see why the wiring from ranes 13 didn't work. You didn't connect pin 1 to the ground at the splitter did you? I would double check the wiring. Pin 2 should be connected to the output to of the two resistors. Pin 3 should be the ground from the circuit but Pin 1 should only be connected to the screen of the mic cable at the mixer end. At the other end there should be no connection between the circuit and pin 1. Make sure the shield connector on the XLR has not been wired to Pin 1.
But if you have tried all this and it didn't work ignore it.
You said a DI box wasn't an option but if you look in the May 2006 isssue of Silicon Chip magazine it shows a simple passive DI Box. It even lets you connect in two resistors so you could use it as the combiner and and DI box in one package. The transformer they are using from Altronics only costs $17 then you need a switch,XLR socket , a couple of jack plugs, and a metal case. As a minimum you would just need a jack plug, xlr socket, transformer, wire, metal case. You can buy the article from Silicon Chip website although you can probably find it in a library.
2) The hum
The other posters have made good points about the hum.
The first thing I would want to is establish the source of the hum.
I would turn the packs completly off and see if I get any hum, in case the laptop is producing some. If it is, try, as someone suggested, running on battery to see if this improves the problem.
Once that is sorted try the dimmers, which are probaly the main source, set all the dimmers to about 30% that should have a fairly high level of interference generated. Then move the laptop and cables around in the space to see if you can minimise the interference. Check the leads aren't running parallel to the dimmers or power cables. If they have to cross power cables do it at right angles. You may need to add an extra lead so you can move as far away from the dimmers as possible.
Hope this helps

The Chris in question me?
Now I will admit I have not looked at the circuits in question, but I would have done the simple combiner with a pair of resistors or whatever and feed the output of that to XLR pin 2, and then connect XLR pins 1 & 3 together and to the sleeve connection. Connecting this to the shielded box would also be a good thing I think. Now abolutely DO NOT do this if you have phantom on the channel.
I would have a copy of that issue of Silicon Chip around if you need it.

Ok just for discussion, looking at the wiring diagram, I am intrigued about the lack of a "backflow" system. tieing both channels < the left and right outputs> without isolating them via a diode would seem to me to be an invitation to blowing out the op amp on the sound card of the laptop. I s my thinking here wrong ?
...
I'm not an electrical engineer, is somebody else here ?

No electrical engineer here, but I think I can comment, though as always, anyone is welcome to correct me if I am wrong. Sorry to put it bluntly, but that idea has a sum total of zero chance of working. Simple reason. Audio is an AC signal. A diode will only pass DC in on direction. So, you won't achieve the desired backflow prevention for what of a better term, and it will seriously affect the audio signal. Not sure how exactly, but I am reasonable sure it would not be pretty at all as you are only getting half of the waveform.


The audio lead I'm using has four wires and a shield braid + foil.
As soon as I understand what I should be doing here I shall put up a different circuit for confirmation.

This audio lead for the XLR end? If so it is star quad (made by or copied from Canare). The theory is that the blue wires get joined each end as do the white (or whatever other colour it might be). Some complex reason why this reduces the amount of interference you get in the cable.

I am tending to think a transformer might help to solve some of your problems. As I said before, I have the article somewhere if you want it (PM me). Now you are using a shielded box right? And keeping the unbalanced cable to an absolute minimum (if you have the unbal cable close to the dimmers then it will pick up the interference as much as or more so than the converter box.)

Hope some of that makes a degree of sense.

Edit: Someone was asking about the pastor who was electicuted. The thread on that is here: http://www.controlbooth.com/forums/showthread.php?t=2989
 
Last edited:
:neutral: :?:
Thanks cutlunch and everyone else for all the fabulous help.
Didn't you just say to connect the shield to pin 3 XLR at the converter end with nothing on 1?
The shield is currently connected to the sleeve of the jackplug (long bit) and is on pin 1 XLR. You're saying connect the shield/ground to pin 3 instead of 1 and put the resistor output to pin 2.

What I am saying is that the sheild is different from the ground. On the cable the shield is the braid/ foil this does not get connected at the converter end.

Get the diagram for the convertor and cross out where the designer has written shield replace this with the word ground.

The sleeve (ground ) of jackplug ( convertor end) is connected to the internal conductor of the cable which at the mixing desk xlr end is connected to pin 3. You have got the resistor to pin 2 right.

The shield is not used as part of the signal path. If as you have it currently wired have the shield connected at both ends this makes a loop hum current can flow in. It is often recommended that when two pieces of balanced sound gear are connected using XLR leads only one end has the shield connected. This is beacuse the shield connector is usually connected to the chassis earth. Connect the shields together and you can have a path through them and the electrical earth causing a loop.

When you think of the balanced input into a mixer think of it like a transformer, which a lot of them may have. The transformer only has two legs so one leg gets the output from the resitor. The other leg gets connect to the signal ground which is the sleeve on the jackplug. This creates a circuit loop for the current to flow in. Think of the shield as a wire connected to the metal case of the transformer it is meant to drain away any inteference so it can't get induced onto the audio signal. If you connect the shield to to any part of the audio signal path then any fault current flowing in the shield now has a path into the audio signal.

So to reiterate again, the sleeve (ground) of the jackplug is to be connected to pin 3 of the XLR.. The resistor output is to be connected to pin 2 of the XLR. At the convertor end there is nothing to be connected to pin 1 of the xlr.
 
The problem with this last approach is that on the on balanced pc side, the ground and shield are connected, but on the balanced side, the mixer thinks it has the differential signal on pins 2/3 and that the shield on pin 1, SO the actual signal processing is different, and if there is any noise on the ground side of the unbalanced side it is going to cause you a problem.

In addition remember that the lines level levels are different unbalanced -10 dBV source and balanced +4 dBu. This is about a 12 db difference remember that every 3 db reduces the level by half. In addition the resistor network further reduces the level of the signal, so you are needing more gain in the pre amp, and of course if there is noise you are just going to amplify it.
On top of all this the insides of a PC or notebook from an electrical noise
standpoint is very noisy, so your are adding to the problem.

anyway just a suggestion
Sharyn
 

Users who are viewing this thread

Back