Proper wiring to avoid ground loops

[JohnRoberts]

There's no reason for you to be aggressive with me, Abbey. Just because I expressed my opinion.
The Dr was not very polite, but maybe part of the reason for that was that English might not be his native language, which is not the case for many members on the forum. My opinion is that everyone's opinion should be listened to and should be questioned too. There is no reason to reject ideas in advance without critical analysis, which is constantly being done here.

I implemented balanced power in several serious mastering studios, in several smaller production studios that had rather poor mains voltage quality, etc. At the time when I was more involved in recording club live jazz concerts, my entire setup was powered using an isolation transformer in a balanced connection.
The finding that a balanced power supply is unnecessary if the studio is correctly wired with balanced cables is often an oxymoron, how many mixers have you worked on that had balanced inserts...Furthermore, every more complex studio consists of at least one computer, and quite a number of devices with SMPS power supplies. In this sense, the quoted article from ePanorama is quite out of date.

I am not hiding from anything, but I am surprised that such experienced and old engineers on this forum have such a closed mind.

And finally, given that this discussion will probably end like the others, with your and JR's word being the last without the possibility of someone answering you, by abruptly closing the thread, I wish you all all the best in the New Year and a little more understanding and relaxation.

Once again my reputation exceeds me.
===

Almost back on topic, back in the 80s when I was over Peavey's recording products division (AMR-Audio Media Research), I investigated making a plug-in balanced power distro for small studio use. Through Peavey I had access to low cost transformer manufacturing (we purchased truckloads of magnetics). Our in-house agency approval guy could help me negotiate UL approval. My brief investigation revealed that there was not much there there. Not enough demonstrable benefit for me to base a capital intensive new product category on.

Around the same time I had a wire company make us our own private label quad mic cable. I never pursued that product either, despite the fact that it actually worked. Coincidentally a recording guy working in Peavey's purchasing department at the time, stole my several hundred foot sample reel of quad cable.

===
Happy new year to you too....

JR

 

[abbey road d enfer]

There's no reason for you to be aggressive with me, Abbey. Just because I expressed my opinion.

Do I read "Now that you have so effectively driven away one member of the forum" as just an opinion?
And your last line was not very nice either.

The Dr was not very polite, but maybe part of the reason for that was that English might not be his native language, which is not the case for many members on the forum.

The "Doctor" has perfect control of the english language, except for politeness. He's been recently banned from GS for teh same reasons. Note: we didn't ban him, he chose to leave the room after having made it stink with his farts.

My opinion is that everyone's opinion should be listened to and should be questioned too. There is no reason to reject ideas in advance without critical analysis, which is constantly being done here.

I don't think ideas have been rejected in advance. Speaking for myself I'm quite familiar with balanced power and I know that it's not the answer to the OP's problem. Actually I doubt rewiring his set-up with balanced power would provide a significant improvement in his situation.

I implemented balanced power in several serious mastering studios, in several smaller production studios that had rather poor mains voltage quality, etc.

How can balanced power improve poor mains voltage?

At the time when I was more involved in recording club live jazz concerts, my entire setup was powered using an isolation transformer in a balanced connection.

I would be tempted to say that you had not properly analyze the nature of the problem. The legendary sledgehammer to kill a fly.

The finding that a balanced power supply is unnecessary if the studio is correctly wired with balanced cables is often an oxymoron, how many mixers have you worked on that had balanced inserts...Furthermore, every more complex studio consists of at least one computer, and quite a number of devices with SMPS power supplies. In this sense, the quoted article from ePanorama is quite out of date.

How would balanced power solve conducted emission problems?

I am not hiding from anything, but I am surprised that such experienced and old engineers on this forum have such a closed mind.

What you consider closed mind is the experience and knowledge that allows some of us to prioritize solutions.
You may (?) have noted that there is a number of subjects taht I don't deal with, because I know I'm not competent enough, but in this case, I know what I'm talking about.

And finally, given that this discussion will probably end like the others, with your and JR's word being the last without the possibility of someone answering you, by abruptly closing the thread,

Well, as moderators, we are some kind of police, so not many love us, we're used to it.

I wish you all all the best in the New Year and a little more understanding and relaxation.

Thank you, but I think you have a wrong appreciation of my state of mind. I do listen a lot, and have sometimes shown that I accept that I'm wrong and can apologize. I sleep very well.

 

[moamps]

Nowhere did I say that using a balanced power supply is the cure for all problems caused by ground loops. From my experience, it is the last possibility to improve the characteristics of the system when it is no longer possible with the usual methods. Or in another case when it is necessary to temporarily ensure the best working conditions, for which I gave the example of recording jazz concerts in clubs. Of course, you can declare it inappropriate use of a hammer, but that system gave me excellent and repeatable results in limited time and space conditions.

Poor quality mains voltage means more problems. First, the mains voltage is significantly distorted, which is the result of overloading the local power distribution transformer, as a rule odd harmonics are dominant in the amount of more than 10%. Additionally, due to the unsymmetrical load, a DC component may also appear. Second, overloading, saturation of the transformer also means that its output impedance will increase and that turning on and off different loads will cause larger short-term drops or rises in the supply voltage. This will also cause the appearance of short-term increased impulses in the ground wires.

A balanced power supply can help here by preventing those higher order leakage currents from circulating in the audio cables. And the value of the leakage current also depends on how the power supplies in the devices used are designed.

A balanced power supply also means that the impedance of both live and neutral leads (maybe we shouldn't call them that now) is the same, which is not the case with a conventional power supply, which can also help to reduce the influence of magnetic fields on adjacent cables. So we get a balanced power transfer, doesn't that sound interesting?

In fact, I have not seen any serious work on the analysis of the leakage current in audio cables, from where it originated and what its form and harmonic structure are. I counted at least three sources, but I didn't manage to deal with it in more detail. This is where my criticism of the solution to the "pin 1 problem" lies, where it does not deal at all with the analysis of the leakage current and any procedures to prevent it from flowing through the shields of the audio cable.
After all, I'm a "transformer" guy so I'll manage somehow.

 

[abbey road d enfer]

Poor quality mains voltage means more problems. First, the mains voltage is significantly distorted, which is the result of overloading the local power distribution transformer, as a rule odd harmonics are dominant in the amount of more than 10%.

No isue with that. Anyway the secondary voltage in most PSU's is heavily distorted by the rectifiers. Unless using half-wave rectification, 3rd harmonic is dominant.

Additionally, due to the unsymmetrical load, a DC component may also appear.

This, I don't get. What "unsymmetrical load"? Do you mean "unbalanced"? Unbalanced load cannot result in current asymmetry.

Second, overloading, saturation of the transformer also means that its output impedance will increase and that turning on and off different loads will cause larger short-term drops or rises in the supply voltage. This will also cause the appearance of short-term increased impulses in the ground wires.

A balanced power supply can help here by preventing those higher order leakage currents from circulating in the audio cables. And the value of the leakage current also depends on how the power supplies in the devices used are designed.

Here the balanced power xfmr assumes the role of isolation xfmr. Just like a balanced connection does not need to be referenced to ground to perform.

A balanced power supply also means that the impedance of both live and neutral leads (maybe we shouldn't call them that now) is the same, which is not the case with a conventional power supply, which can also help to reduce the influence of magnetic fields on adjacent cables. So we get a balanced power transfer, doesn't that sound interesting?

I take the point. I'm not sure about teh practical results of it.

In fact, I have not seen any serious work on the analysis of the leakage current in audio cables, from where it originated and what its form and harmonic structure are. I counted at least three sources, but I didn't manage to deal with it in more detail. This is where my criticism of the solution to the "pin 1 problem" lies, where it does not deal at all with the analysis of the leakage current and any procedures to prevent it from flowing through the shields of the audio cable.

Well these sources are probably the same as mine. I don't take them as gospel, but just a starting point for reflection.
The analysis is good but not exhaustive, so one has to do some homework, and the proposed solutions are like any receipe, they need the cook's touch to rech excellence.

After all, I'm a "transformer" guy so I'll manage somehow.

No doubt about it.

BTW, I've been faced once with such a distorted mains that a rack of Crown amps kept blowing fuses. There was about 10Vdc. I wasn't allowed to investigate the cause (Disney's advisors don't want to be challenged), but I suspect some dimmers had gone haywire.

 

[moamps]

No isue with that. Anyway the secondary voltage in most PSU's is heavily distorted by the rectifiers. Unless using half-wave rectification, 3rd harmonic is dominant.

This, I don't get. What "unsymmetrical load"? Do you mean "unbalanced"? Unbalanced load cannot result in current asymmetry.

I meant loads that contain the implementation of diodes for power regulation (ovens, etc.) where one half-period is loaded and the other is not.

Here the balanced power xfmr assumes the role of isolation xfmr. Just like a balanced connection does not need to be referenced to ground to perform.

Yes, but the balanced power supply gives a little more quality.

BTW, I've been faced once with such a distorted mains that a rack of Crown amps kept blowing fuses. There was about 10Vdc. I wasn't allowed to investigate the cause (Disney's advisors don't want to be challenged), but I suspect some dimmers had gone haywire.

Yes, the old dimmer types with classic bulbs were really terrible sources of interference and distortion. Their regulation was simple, sometimes without a circuit for detecting the passage through zero voltage.


Edit:
My experience with older Crown amplifiers as well as many dbx devices is that the power transformers were not well designed in terms of insufficient magnetization reserve (if any at all) and that they had a lot of problems with the transition from 60 to 50Hz due to saturation.

 

[Voyager10]

The 'centre tap connected to ground' part doesn't seem to be bringing much to the party, if the main issue here is current flowing through L-E and N-E capacitors in the powered device. These currents will only be equal and opposite to the extent that L and N are equal and opposite, and also the capacitances are equal. Given typical tolerances of +/-10% or 20%, that's only in the region of 20dB reduction.

A pure isolation transformer (no centre tap) would do better here because the L-E and N-E capacitor currents would have to balance - there's nowhere else for them to go. (L and N would float w.r.t. E, and the actual L-E and N-E voltages would end up being in inverse ratio to the capacitances).

I'm also wondering whether putting a big lump of metal in the way simply attenuates a lot of the HF noise, and that's where most of the benefit comes from?

 

[abbey road d enfer]

I meant loads that contain the implementation of diodes for power regulation (ovens, etc.) where one half-period is loaded and the other is not.

OK, I was not sure that's what you meant.
I believe (correct me if I'm wrong) that this type of implementation is now restricted to very small power appliances, precisely because of the risks of generating DC..

 

[abbey road d enfer]

I'm also wondering whether putting a big lump of metal in the way simply attenuates a lot of the HF noise, and that's where most of the benefit comes from?

It probably does, however one must not neglect the magnetic field introduced by such a transformer.
In a studio I was working for, such a transformer was installed in a closet. There was a tie-line junction box in this closet. Since these tie-lines were seldom used, it took some time to figure out that these had become noisy. Moving the junction box a few feet up solved the problem when it became noticed..

 

[aaquilato]

I had a client in LA that was building an immersive studio, and was quoted $150K for a balanced power system. He called me in a panic and I explained to his current electrician how to wire the studio, and then ordered him a 10KVA 220V to 110V transformer for $2K, had the electrician jump the phases together in the panel and he has never had a micron of noise in his studio. Not balanced, just isolated. When I worked for Record Plant Remote, we had the same thing. Every outlet was on one phase. Never had an issue. Balanced power involves grounding the center of a transformer, which makes no sense to me, accept it eliminates grounding problems that should be resolved correctly.

90% of problems come from having outlets on opposite phases, Remember that in your existing panel, every other breaker is on the same phase. Sometimes having an electrician move all your outlets to one phase and lamps, lights, microwaves and fridges to another gets rid of all the problems...

Not all electricians are as smart as you Paul. And not having any real EE background myself I couldn’t explain properly the problems I was having other than there was voltage where there shouldn’t be. in some places where there is “no code” like here in TN, builders can do whatever they want to cut corners and make things easier for them, not the home owner.
The wiring scheme in my home in TN is so bad I couldn’t afford to have an electrician come to rewire my house.
Is was more affordable to add small balanced power units to clean up all the electrical “noise” in my studio space. Now those outlets are free from any ground problems.

Nah, balanced power makes very little difference.

Small balanced power transformers eliminated all the noise and ground issues in my small studio space.
The wiring was so bad with different voltage between N-G in different rooms it was the easy fix.

 

[radardoug]

Thats not balanced power. Thats using the inherent isolation of a transformer to get rid of ground loops.

 

[fazer]

The wiring was so bad with different voltage between N-G in different rooms it was the easy fix.

Not sure if this is your problem but over time the mains panel breakers become loose on the lockdown screws. The compression of copper wire flattens out and the connection becomes resistive. The breaker screws and the wire is not as snug and also happens on the neutral and ground.

My brother had a greenhouse business and I would go in once a year and replace fan motors if need be but many times the first thing would be to tighten all the connector and crips and wire nuts. I’d finish and things would behave better. I have done the same thing to studios when servicing problems. I’ve found breakers loose as well as ground wire and neutral strips. This includes checking ac outlets that are more of a pain to get to. If an electrician used the push in terminals and daisy chained outlets then I guarantee there are terminals with wires that have become resistive. In my own home studio I always run new clean power cabling with new added breakers from the panels in the last three houses I’ve owned. It’s a pain but solves numerous problems and provides a clean earth path for my audio gear.

 

[JohnRoberts]

my house is the poster boy for lousy wiring not to mention 2 wire outlets.

I have since wired in a handful of GFCI outlets near water/plumbing.

JR

 

[terry setter]

Thank you, Moamps, you're a true gentleman.

Happy New Year to all!

 

[CalavoBob]

Back to the subject of this thread.....
Getting rid of those pesky noise floor sounds. A friend had an apartment in a hi- rise next to another hi-rize with an FM radio station tower on top. His stereo, expecially the phono picked up rf. We tried everything: putting the gear into a steel rack, adding ferrite beads on every i/o connection, grounding the rack to earth, a power line conditioner, even shielding the speaker wires, and all of it helped some. Granted this was unbalanced home stereo stuff. There was still rf bleed. Finally I looked at the 3 foot 3-wire AC cable coming into the rack and got the idea to slip a woven copper mesh shield over the 3 feet of power cable, neatly heatshrink it at both ends and bond it to the conduit at the wall outlet. The rf went away.

 

[Cqwet Dbdfte]

These days you can buy shielded AC cords (cheap) on Amnz, foil and drain wire bonded at each end to ground, maybe just as good as the above

 

[Brian Roth]

RFI can be a Booger, and not just when next to a broadcast transmitter. Cell phones can phart into audio paths.

AC mains cords with shielding CAN help sometimes.....always worth a try!

Bri

 

[drtechno]

Hi, have 60hz hum in my recording setup, and I am suspecting it is due to ground loops. I have a preamp and compressor that have their chassis grounded to earth, and also both have both input and output xlr pin 1 grounded to the chassis. If I remove the grounds on xlr pin 1 from both outputs, is that the proper way to break the ground loops in my setup?

Yes it is.
The rest of the guys here apparently like debate everything. You can also ask a different pro the same thing on gearspace,com or realgearonline.com if you want real professional opinions apparently.

 

[Cqwet Dbdfte]

"Proper way" meaning what? The litterature is full of examples where shields are bonded to case at only one end.
Looks like the hum problem was a groundloop and the cause was found.
Seems proper to use the fix. No laws broken.
Hum can have a few sources, so a simple answer is not always possible.
Henry W. Ott's book is some 300 pages.
"Noise Reduction Techniques in Electronic Systems". Pages 43, 44, 45 gives experimental values for noise reduction in 11 grounding/shielding examples that seems relevant here.

 

[abbey road d enfer]

Hi, have 60hz hum in my recording setup, and I am suspecting it is due to ground loops. I have a preamp and compressor that have their chassis grounded to earth, and also both have both input and output xlr pin 1 grounded to the chassis. If I remove the grounds on xlr pin 1 from both outputs, is that the proper way to break the ground loops in my setup? Also somewhat related, my microphone does not have xlr pin 1 connected to the mic body, should I make that connection? Thanks.

Why don't you try it, instead of asking us? Apparently, according to some superior mind, we are just a bunch of idiots lacking experience...

 

[JohnRoberts]

I am surely repeating myself but multiple ground noise contamination issues are lumped together and all called "ground loops", despite few actual loops but that is a starting point. This is a mature topic and well inspected here over the years.

"Pin1" problem is dominant in lots of legacy (older) gear, while most modern gear handles shield grounds properly.

"Proper way" meaning what? The litterature is full of examples where shields are bonded to case at only one end.
Looks like the hum problem was a groundloop and the cause was found.
Seems proper to use the fix. No laws broken.
Hum can have a few sources, so a simple answer is not always possible.
Henry W. Ott's book is some 300 pages.
"Noise Reduction Techniques in Electronic Systems". Pages 43, 44, 45 gives experimental values for noise reduction in 11 grounding/shielding examples that seems relevant here.

my copy of Ott is C. 1976. on page 80 he describes "combination or hybrid" shield grounding where one end is grounded through a HF capacitor.

JR