Ground Loop / hum

[fjgaston]

Hi,
I have a problem with my mixer (Mitec EX31 - 48tracks) It generates too much hum. I think it comes from a ground loop.
I unplugged everything, muted every channels and just plugged my headphones and the hum is very low but still audible, it starts growing when I plug grounded gear like my audio interfaces (echo audiofire 12) my computer and monitors.
The hum is a also present (even more audible) in the monitors out when I plug my headphones in the aux output.
The problem was even worst with my previous computer, a cheap desktop PC, it seemed like I could hear the hard drive turning, but with my mac pro and macbook pro the hum is a lot lower but still too  much for a comfortable work place.
The mixer is a bit old (from the 80s) is it possible that the problem could come from bad electrolytic capacitors or bad PSU ?

I don't really know how to measure, I tried to measure it with a multimetre, I had 2 mV AC on the headphone output with the headphone at maximum volume.
Is there an easy way to find where the problem come from ?

 

[ruffrecords]

Do you get anything with nothing but the headphones plugged in and the headphones volume turned right down?

Cheers

Ian

 

[fjgaston]

No, with the headphone turned down, I have no noise in the headphone

 

[pucho812]

how old is the desk? Has it been recapped before?

 

[ruffrecords]

No, with the headphone turned down, I have no noise in the headphone

Are the headphones just connected to the aster bus or can the monitor other things?

Cheers

Ian

 

[midwayfair]

What frequency is the hum? Is it most prominent at 60Hz or 120Hz? Do harmonics appear?

Also, try using an audio probe to locate where in the circuit the hum appears.

A ground loop is usually a buzz more than a "hum". Grab an alligator clip or two, and locate the ground points in the circuit. Put one end of the clip to your chassis ground (wherever that is) and the ground point you're testing. Be systematic with each channel.  FWIW, I have found multiple times that I need more than one chassis ground even though typical star grounding only has one. In an amp, I used the input jack for one ground for the audio line of the PCB and a second chassis ground for the power section on the other side of the chassis. If I just ran a wire from the input jack to the chassis ground, I got a bit of noise. I ran into a similar problem with a preamp, and ended up grounding both input jacks to the chassis in addition to a star ground for everything else.

 

[fjgaston]

The console is more than 30 years old but I know the PSU has been serviced by the previous owner.
In the console some capacitors look weird :

Does it look like damaged capas to you (the ones on the top) ?

 

[JohnRoberts]

It is difficult to chase down noise sources in complex equipment (like a console). Looking at outputs there are multiple opportunities for hum to get in.

Back up and sniff closer to the inputs... Hi z cans can usually be patched into inserts or direct outputs of an input channel to gauge the noise contribution of the input stage.

Hum can come from common vectors like PS ripple, or poorly designed input circuits (google "pin 1 problem"). You could find multiple things to fix, or learn to live with.

JR

 

[fjgaston]

In fact there is some noise with nothing plugged in and every channel muted.
The buzz starts growing when I plug my direct output to my audio interface.
Maybe I should start by changing the electrolytics in the master channel.

 

[JohnRoberts]

never mind...   

JR

 

[radardoug]

The problem with people like you is you dont listen! John has given you good advice, have you taken it? Firstly, with nothing plugged in except your cans, you still have some hum. This would possibly indicate some power supply noise. Measure your power supply rails. The first thing I would look at is recapping a 30 year old power supply. Get a capacitance meter, $30 on Ebay. Remove one cap and measure it. If it is under its printed value, replace it. If its not, then chances are all other caps of this value are fine too.
People get hung up on changing caps, its not always the caps!!
John told you to read up on pin1. Do it!
Learn how the grounding is done in your console.

 

[fjgaston]

I listen (read) what you write and I try to understand (sorry if I misunderstood) and I'm giving the informations I have.
But I'm not sure what and how I should measure.

* I red about "pin 1" in google and understood problems appear when two grounded units have their ground/shield linked by their common pin 1 in this case the best solution would be to only use symmetric connections with no pin 1 ?

* How could I measure Power supply noise ? should I measure the +15v and -15v for AC voltage with my multimeter ?

I'll try this and the capacitance test.

Thanks for the advices.

 

[JohnRoberts]

I listen (read) what you write and I try to understand (sorry if I misunderstood) and I'm giving the informations I have.
But I'm not sure what and how I should measure.

* I red about "pin 1" in google and understood problems appear when two grounded units have their ground/shield linked by their common pin 1 in this case the best solution would be to only use symmetric connections with no pin 1 ?

You apparently don't understand or read a poor source of information.

Pin 1 problem is related to an input differential that fails to ignore noise on the shield (pin 1).  The correct solution is to correct the grounding and input differentials.  (pin one should be bonded to chassis ground, audio 0V not so much. Audio should be extracted from + wrt -, and do not lift shields).

* How could I measure Power supply noise ? should I measure the +15v and -15v for AC voltage with my multimeter ?

Using AC scale cap coupled you can measure AC ripple on the DC supplies.

I'll try this and the capacitance test.

Thanks for the advices.

This is still scatter shooting guesses...

I repeat divide and conquer... confirm that input is rejecting hum first, then you can work your way up the audio path to find out where it is getting corrupted (may be more than one source of hum).

Some mixer designs have multiple design problems, some have component deterioration too.

Was this mixer quiet before and suddenly got noisy? Was it ever quiet? This is an important question since it determines if we are looking for a design flaw, or component failure(s).

Looking for easy guess answers rarely work out. Do you feel unusually lucky?

JR 

 

[moamps]

* I red about "pin 1" in google and understood problems appear when two grounded units have their ground/shield linked by their common pin 1 in this case the best solution would be to only use symmetric connections with no pin 1 ?

Yes, you can try to cut the ground on one side of your jacks (just two channels for experiment) and see what you will get. Sometimes helps. I will not suggest you to rearrange your console grounding sheme.  But first, try to connect all your equipment to one correctly grounded power outlet using a power strip. This can help to minimize ground loops. Do you have a service manual of the console?

 

[JohnRoberts]

I do not advise lifting shield grounds, while there may be better places to connect them.

In a properly operating interface the shield is grounded on both ends.

JR

 

[moamps]

In a "properly"  designed small studio (all balanced outs and all differential inputs) the shield on the cables is not needed at all.

 

[fjgaston]

divide and conquer... confirm that input is rejecting hum first, then you can work your way up the audio path to find out where it is getting corrupted (may be more than one source of hum).

Some mixer designs have multiple design problems, some have component deterioration too.

Was this mixer quiet before and suddenly got noisy? Was it ever quiet? This is an important question since it determines if we are looking for a design flaw, or component failure(s).

Looking for easy guess answers rarely work out. Do you feel unusually lucky?

Since I have this console it has always been noisy, it was very noisy with my destop PC (could hear different strange noise), a lot less noisy with my mac (only a background buzz).
The buzz is still not very loud but it is noticeable and I guess it can be less noisy.

Some mixer designs have multiple design problems, some have component deterioration too.

The mixer is old and many electrolytic capacitors look bad but as you said "Looking for easy guess answers rarely work out"

divide and conquer... confirm that input is rejecting hum first, then you can work your way up the audio path to find out where it is getting corrupted (may be more than one source of hum).

Can I do this with no oscilloscope ? how ?

If it can help, one thing I did,  with my previous audio interface was I removed the grounding on the output (removed the connection to the shield and ground) then the noise in the console was gone

In a "properly"  designed small studio (all balanced outs and all differential inputs) the shield on the cables is not needed at all.

On this console tape in/out are wired like TRS inserts ,so no differential/balanced connection to my audio interface is possible if I want to use the tape in/out.

 

[fjgaston]

Well, I removed all the channels from the console and kept only the master channel to be able to isolate the problem.
Now when I listen to the headphone I have no noise, exept when I plug my interface to the aux return and solo the aux return.

One more thing, I just noticed the power cord has been damaged and poorly repaired. I red that it could be a source for ground loop.

A new power cord could only be a good thing 

I tested the PSU, the +15v is 14.8 and the -15v is -14.8, and I can't read any AC voltage but my multimeter can't read any AC voltage under ~2mV so I don't know if my test is worth something.

Using AC scale cap coupled you can measure AC ripple on the DC supplies

How do I do this ???

 

[DerEber]

Haha,

get a new Power cord and report back.

 

[JohnRoberts]

A new power cord could only be a good thing 

I tested the PSU, the +15v is 14.8 and the -15v is -14.8, and I can't read any AC voltage but my multimeter can't read any AC voltage under ~2mV so I don't know if my test is worth something.

Using AC scale cap coupled you can measure AC ripple on the DC supplies

How do I do this ???

Typical VOM can't measure small VAC in presence of much larger VDC. Using a decent sized electrolytic cap (say 10uF but could be larger or smaller than that) in series with VOM + lead can block the DC and allow AC measurement. When probing the + supply make sure the + lead of the cap is toward the + rail. For measuring AC on the - rail, reverse the polarity of the cap so + lead is toward the VOM.

Note: Power supply ripple could change depending on how many modules are plugged in.

JR