Ground Loops

[Robh3606]

Most of us have speakers with fairly high sensitivity. I am sure all if not most of us have been plagued by this at one time or another. I just saw a recent post about a problem and just spent over 2 hours trying to ring the last bit of hum out of my system the best I could. As usual just stumbling blind I found the right combination of grounds between my unbalanced, balanced, and cable TV feed. I don't want to tell how much time I have spent chassing the bugger down in the last 6 months. Why don't we pool our experience and try to put together a reference on the best aproaches that have worked for us.

Rob

[Mr. Widget]

This subject gives me a head ache just thinking about it.

Rane has an audio glossary on their web site with some very useful information.

http://www.rane.com/note151.html

[Ken Pachkowsky]

Originally posted by Robh3606
Most of us have speakers with fairly high sensitivity. I am sure all if not most of us have been plagued by this at one time or another. I just saw a recent post about a problem and just spent over 2 hours trying to ring the last bit of hum out of my system the best I could. As usual just stumbling blind I found the right combination of grounds between my unbalanced, balanced, and cable TV feed. I don't want to tell how much time I have spent chassing the bugger down in the last 6 months. Why don't we pool our experience and try to put together a reference on the best aproaches that have worked for us.

Rob

Thats a great idea. I should make a flow chart of my current setup. Its a mixture of different connections.

Ken

[Doctor_Electron]

Hi Robb! I would also have recommended the Rane tech note. But...sometimes noise problems can defy all the technical theories, standards, and practices. For example, after a project at the local Air force Base, which involved moving 40 instrumentation recorders, their racks, patchbays, and associated power systems to another room, the installation would not meet the applicable minimum noise standards. All cabinetry was insulated from the "computer tile" floor with special phenolic sheeting, all power was fed by a state-of-the-art EPE "power cart", with isolated ground circuits to each rack. Each rack had a huge copper isolated ground buss, with the usual "uniground"/"star ground" configuration. The room had three seperate grounds, two being isolated...utility, technical, and crypto.
After everything (this was a five month job for one engineer and four techs...all the gear plus over 4,000 input signal lines, which if not patched to a recorder channel was "normalled" thru its patch bay jack as an output, i.e. 8,000+ signal lines) was rechecked for errors (none), engineering sent in the grounding/power quality gurus with their Dranetz analyzers.
All said & done, at the time no definitive cause(s) for the anomaly were located. But here's the funny part...the only way the noise spec could be met was by connecting all the isolated grounds together in the room, [ the exact antithesis of standard practice ] !
A deciscion was made to leave the utility & tech grounds connected, as long as no problems were introduced in other areas {none]. The crypto ground issue was addressed through other means, which I am not at liberty to discuss.
The moral of the story is...if the theory & dogma won't fix such problems, don't rule out other approaches which normally make things worse.
BTW, years later, in the course of an unrelated project, it was found that somehow a low level of AC voltage was being induced in the rebar of the structure, and combining the iso grounds was merely cancelling out part of that AC being induced in that area. Of course, the levels detectable by the recorders is much lower than anything likely to be encountereed outside such an installation.
Go Figure!

[Robh3606]

Here are my latest experiences with this problem. I had an intermitent problem which is a real B^&%& to find. I came from no where went on for about six months. Late a night I would get a hum. It was like clockwork and I never figured it out. My system is on a dedicated line so this was really driving me nuts. Then my service was upgraded from 100 to 200 amp service so everything was replaced. Got a constant hum from that! Looked around the house and found my outside cable ground disconnected from the grounding spike. Hooked it up and silence!!!! Yes!!! Then the intermitent problem came back!!!! This time it would happen whenever no constant time like the last one. It ended up being one stinking connector where the shield was loose. Still had a problem though. When the cable line was disconnected from the VCR hum city! So I went hunting again and finnaly ended up using seperate grounds for each. I had always had them tied together before. Sorry for the long winded story.

So my two grounds wires in the sytem are from the turntable ground on my reciever to on of my Urie EQ's. Then another from the active crossover rack to one of the amps in the amp rack.

The cable is just grounded to my metal outlet box and now I can pull the cable line and no hum!!!

So maybe I won the battle for a little while anyway.

I will post a system schematic later too so this makes more sense,

Rob

[Steve]

Aloha
Really great and helpful thread.

My son has a computer driven audio and video studio set up in his room. From day one we fought ground loops in the studio and my audio set up in the living room.
I had run new dedicated 20amp 120V lines to each room of the house, for the computers, my audio set up and his studio. I had to ground lift everything, even used the "cheater" plugs to get rid of the ground prong on the plugs. I don't like doing that. Now, I don't use the plugs and ground lift is off on most of the electronics. . My Altec 9440A sub amp loved to reproduce the hum. I even had the electrical company out to check the neutral on the pole in our back yard to be sure it was good and solid and no corrusion. Electrical company said the hum wasn't there doing. I even installed a new ground rod to be sure.
Called the cable company, they said it was the electrical in the house and not their fault. They came out and installed all new cable runs ( that later had to be redone as they put too many staple holes in them and too tight of bends) They did connect up the cable ground that wasn't installed or connected before to the water pipe. That caused horrible rolling bars and moving lines on the TV. A sure sign of ground loop problems. Cable company came out again. Disconnected the ground and said I have 50+ volts coming out of the house on the cable. Must be a bad tv or ground. Found the tv that was doing it.... trashed it. Install a new tv and the ground loop noise was reduced in volume by about 3/4. To do any recording or turning on the sub amp, still, we had to unscrew the cable at the house connection.
I researched and ordered from Parts Express, Isolation transformers #180-075 for each cable connection in the house and the main cable feed. These have up to 1ghz frequency operation so as to not mess too much with my cable modem, hopefully. That solved all the ground loop problems except for his keyboards. One, a Korg Triton, the other a computer keyboard connected to his Akai MPC toy via USB to the computer. The Triton was because it used another outlet fpr power, so we use an optical cable now and solved that. The TV cable to the house still doesn't have a ground. When I run a ground myself, the ground loop comes back, as well as the rolling bars on the tv's. I don't measure any voltage coming from the house. I still have some hum in the portable telephones that have a wall wort. Yes, I did unplug those to check.... didn't help. I am in the process of changing out all the ceiling fan speed controls at the wall and returning them to their original pull chain operation. Later this month I hope to change out all the light dimmers to a normal switch. One thing that does happen, when I change the speed on my ceiling fan in my room, the sub in the studio pops. That is the first wall control to get rid of. The sub is on a dedicated line.....different than the fan. Oh the fun......just keeps on going.....

Some of the pages and sites that I have found on ground loop problems. Some technical, some in laymans terms that I better understand.

http://sound.westhost.com/earthing.htm
http://siber-sonic.com/broadcast/GLoopwhatis.html
http://ethanwiner.com/dimmers.html
http://dplay.com/tutorial/cablehum.html#atten
http://elect-spec.com/faqgrdl.htm
http://documents.epanorama.net/documents/index.html
http://gbaudio.co.uk/data/ground.htm
The rane site has a great pdf file

I hope some of the above links aren't repeats.
hope this helps some. I have a lot more work to do....
At least for now we can record without any ground loop noises.....
I am looking forward to the time when most all cabling can be optical.....


Steve

[Doctor_Electron]

I'm gonna keep it real short this time...
http://industryclick.com/magazine.as...&magazineid=31

[boputnam]

Hey, Rob...

We had a go at some of this on the Forum a bit ago - with some helpful input from those similarly afflicted. Here's the Link: http://www.audioheritage.org/vbullet...=&threadid=171 At least this will all be woven together!

I just returned from my annual 4th of July gigs in New Mexico where I spent considerable time with my engineer making cables to improve my home rack(s). I have a very slight GL that appears to originate between the consumer preamp and the pro EQ - a most common local.

I'm not going to rewire stuff tonight - too much SW airlines bronco body - but will in due course and will post the (hopeful) success!

[boputnam]

OK, tonight I fired-up the Pace solder iron (git yerself one - they are tops!), and began the forensics...

I elected to work backwards - from amps to preamp. This seems a smart idea. Also, wherever possible, plug any 1/4" in first (either unbalanced or TRS) and the XLR last. 1/4" plugs short the circuit during connect/disconnect - XLR's do not.

Turn everything on.
Unplug everything, starting from the preamp forward, and note the change in ground loop (GL).


Start with the power-amp to crossover connection.

The JBL 5234A active crossover is an odd beast - it has balanced inputs, unbalanced outs. Doh! With trial and error, I found I needed to drop the shield (Pin 1), and only connect Pin 2 (signal positive) and Pin 3 (signal negative) on the XLR end (connection to QSC amps). On the crossover end, I used 1/4" unbalanced, and left the shield wire unconnected.

"Dixey Bell"!! - no ground loop (GL) This removed nearly 90% of my GL noise.

Crossover to EQ

Here, I have a moderate GL. It is insensitive to using a "normal" balanced cable (both the crossover ins and the EQ outs are balanced), or a balanced cable with Pin 1 dropped - this should have worked, but did not.

More, tomorrow...