Please help me understand a ground loop phenomenom

[firby]

Hi.

In my studio everything is running off of one outlet that is branched out through a series of quality cabling that goes into a furman power conditioner. It is quite a bit of stuff. A 64 channel Trident 24 desk, some pres, a computer, an aux rack with delays and verbs and a compressor rack which is mainly connected via inserts on the groups. There are lots of line level signals from synths and samplers.

That is all good. However, there is one place where there is another path to ground. That place is a mackie 1202 that is connected by the drum set to provide a headphone mix. That mackie is connected to two aux outs on the Trident that carries the headphone mix. Then that mackie 1202 is connected to a separate power outlet by the drumset. The mackie doesn't do anything other than amplify the headphone mix coming from the Trident.

The other day I had the studio gear on and I plugged a vacuum cleaner into the outlet where the mackie was connected and I heard the vacuum cleaner come out the bigs on the Trident. Pretty loud but not damaging. So, I unplugged the mackie mixer and the sound went away when I turned the vacuum cleaner back on.

I get that a studio should usually have one path to ground and I have a grounded extension cord that reaches over to the mackie from the furman that is generally set up. When I do drum tracks the headphones always sound clean no matter wether the mackie is plugged into the furman or the remote outlet.

Why did the vacuum cleaner make the monitors howl ?

Can somebody please explain the mechanism how this works ?

Sorry for the long post. It's challenging to explain all of it concisely.

Regards.

[Les]

I don't think your case is that much of a ground loop issue. I do realize that getting a perfect star ground is often difficult or impossible.

A vacuum cleaner typically has a large universal brush type motor.
The sparking of the brushes generate massive amouts of radio frequencies.
Any place where balanced line common mode rejection is compromised (like at an xlr or other connector) is a potential antenna for RFI.

You could probably track down and get rid of RFI in the monitors whith filters, but really does it matter? I don't think you'll be running the vacuum cleaner much when recording or mixing!

[Matt Syson]

Hi
I think Les's comments are about right, don't use your vac while recording.
All this stuff about 'balanced' and RF rejection is only true to a point in PRACTICAL terms, meaning that although the theory is correct, once you get to frequencies much above 'audio range' the actual degree of 'balance' and the capabilities of the gear to actually reject any imbalance is VERY suspect. Balancing of cable is simply presenting 'noise' equally to an input stage. This stage then has to be able to seperate the 'noise' from your wanted signal. This process is actually very tricky to do well and consistently, especially at higher frequencies.
As powerline noise is most prevalent, the majority of gear is at least moderately good at rejecting it's presence, and it is relatively easy to do so. Beyond say 30 KHz getting an accurate 'balance' from one end of a signal run to another is far more tricky. This includes the output stage (sending) and the input stage (receiving). Companies used to be proud of their in / out stages and would produce graphs of what could be expected. Nowadays it is hard enough to get them to say if it is balanced or not and if so by which 'technology' as it has SERIOUS implications in some situations.
I doubt you have a 'ground loop' issue but it is 'simply' your vac being a wideband RF noise generator, which is being picked up somewhere in your system.
Matt S

[nosebleedaudio]

DC motors (most vacuum cleaners use DC motors) also can induce a boat load of trash on the AC line, if you looked at it with a scope you would no doubt see some serious trash and hash, spikes...

[JohnRoberts]

A ground loop is an often misused catch all term to describe ground noise problems. Ground "loop" suggests an actual physical loop, like a one turn transformer winding, that converts the magnetic field existing across that loop into a voltage.

Professional sound interfaces generally pass the audio data on two signal conductors that are independent of actual ground voltages, which will be connected via the third conductor and/or shield. There will often be ground potential differences between disparate chassis imposed on these audio signals common mode (the same voltage added to both), so they cancel out in differential receivers that subtract - from the +.

Since you are not receiving audio back from your external different grounded satellite mixer, the source of noise appears to be from a common ground connection in the signals being sent to the satellite.

This suggests a design flaw or weakness in that ground noise should be ignored everywhere. That said large consoles contain hundreds of local audio paths internally that are unbalanced and not treated differentially for short distances, so the ground noise corruption is getting into one or more of those paths.

The single robust ground approach to keep a studio quiet, is actually a concession to this generally insufficient design treatment of signals wrt grounds (google pin one problem for a discussion of ground corruption with respect to I/O)..

The simple answer for "it makes noise when you do that", is don't do that.

JR