Theoretical ground loop question

If my understanding is correct, you would have a higher probability for problems.

Is the garage panel a sub-panel or is it supplied from its own meter and service entrance line from the street? I don't believe if it's a sub-panel it should have a seperate ground. In either case more grounds = more chances for ground loops.

Even if the sub panel is driven off the same mains feed, more ground wire distance equals greater risk of ground loop problems.

However, in a properly designed system, ground loops should not cause hum. If you get hum from ground loops, some component in your system has a design flaw (bug) that causes the hum. Best solution is to find the offending component and buy something better. A properly, fully groudned system will be less noisy than a system that has un-grounded parts because of a bad component.

Sub panels cannot have their own ground electrode. Only 1 ground electrode allowed at the main service entrance in the US.

When last I checked, if a subpanel is in a separate building (e.g., detached garage) it *may* have a separate ground. If it does, it must only be connected to the subpanel's ground bus, and not to the neutral bus. All subpanels, regardless of location, must have their neutral bus isolated from the ground bus. But in a remote location, where the subpanel ground must be carried a long distance from the main panel in the main building, the risk is much greater that the grounding connector will get damaged or disconnected, leaving the whole subpanel ungrounded, so you may ground the subpanel with a separate ground rod at the subpanel location.

This only applies to separate buildings, though. I doubt the original poster has two circuits in such close proximity if their breaker panels are in separate buildings ...

Only if the sub panel is electrically isolated from the main service panel. Otherwise, it is a no go at any distance.

I think Sands_at _Pier147 is referring to NEC 250.32 which has some specific instances for tying ground and neutral together at a remote panel in a separate building. As mentioned above, it doesn't sound like this is the situation with the OP.

..Doyle

Gotta disagree here. Few home theater components have any provisions to compensate for large ground potential differences between interconnected components. The only equipment naturally immune from ground loop trouble is that which is connected solely through a fiberoptic interface. Those large voltage differences don't occur in most installations thankfully but when they do they are usually a consequence of widely separated ground connections (loop area) and do not necessarily indicate faulty equipment.

Few home theater components even have three-prong cables. I can't believe they charge $1,500 for a piece of equipment that has a two-prong input...

And much more than that too.

Dunno that I would equate the presence or absence of a grounding power system with any particular degree of quality, reliability, or even safety (in the typical HT setting).

Whoaru99 is correct. On any equipment that third prong is a safety feature only. It never has been intended to prevent or reduce noise. Ground is not a noise sink. Unlike a washing machine, most HT equipment can be built to standards that don't require a safety ground connection. That's a good thing since the more ground connections there are in your system the more likely it is that you will experience some ground loop related noise.

While I agree ground is a safety, the problem is that with two prongs, you have to re-construct a signal ground, and there are different ways of doing that that may or may not be compatible. In the simplest case, you treat 0 as signal ground, when coming out of the rectifier, and call it good. In a more advanced case, you create a signal ground that's at the midpoint of phase and 0, and tie that to safety ground, creating a balanced power path. It is my understanding that it "shouldn't matter" if all signal cables have zero resistance, so that all the devices find common ground (uh... no pun intended :-). However, cables do not have zero resistance and hence, hum eliminators that work as ground isolaters actually often work, even though there "shouldn't" be any difference.
The conductor for safety ground is likely to be closer to zero resistance than your typical signal cable. It might also have all kinds of EMI on it. Which, again, is why I like balanced connectors :-)

FWIW, I went thru the tortures of the damned to eliminate grounding issues with my hi-fi, which is totally independent of the HT system. The hi-fi is supported by a separate subpanel and feeds three dedicated 20 amp lines with appropriate wire and receptacles. The difference in grounding potential between one of these lines and another is enough to cause hum on the system, which uses very sensitive (@103 db) horn speakers. The equipment itself is not shoddy, or in need of an upgrade- Lamm ML2 amps, Lamm Ref line stage, etc.
The grounding problem does go away by plugging everything into a single receptacle. In order to provide enough outlets I must use some form of power strip or conditioner (eg, Shunyata Hydra) but I really wanted nothing between the wall and the equipment.
I bought one of those Ground Zero thingies, and despite the skepticism with which others have greeted it, it works. All hi- fi gear is now plugged directly into the wall sockets, I can use the separate dedicated lines (eg, one for turntable, so motor controller does not interfere with sonics; one for front end electronics, eg, phonostage and line stage; and one, with two separate receptacles, to feed the powered subs at some distance from the first two AC line receptacles). No grounding noise whatsoever when I tie the Ground Zero device to the line stage and to the power amps on the woofers (which, as above, are fed by different dedicated lines).
Some folks float the ground, ie, use a 'cheater' to get around grounding problems but there are some risks of shock. (I also know one person who advocates lifting the ground on all components simply because it sounds better).

For my video rig, I had the electrician run a 220 line and use a step-down transformer which feeds some Richard Gray boxes. Despite the complexity of the video system, I have never had a problem with grounding noise there. (In fact, the step down also powers my tone arm aircompressor which otherwise makes a nasty 'zap' when it surges on; plugged into the transformer, that problem disappears).
Grounding issues are insidious, are not necessarily the result of defective or low grade equipment and take time to sort out. But, the first step is good clean power to the room.

Yes.

For the purposes of this discussion, a groundloop is defined as:

Any physical loop of conductor which is able to trap magnetic field within the loop, and generates voltage as a result of that magnetic field changing it's strength over time. This generated voltage is defined by Faraday's law of induction (for all the geeks).

Or

A loop which somehow taps into a change in voltage potential caused by a current.

First, the magnetic field thingy.

A generator works by changing the magnetic field within a loop or loops of conductors. They do this by rotating the loops or the magnets.. The crux is, the field is changing and it induces a voltage. If the loop is closed, like a ring, currents will be generated.

If you have two three prong devices, an amp and a source, and plug both into one wall outlet, you still form a loop.. Between the grounds. If you space the two cords as far away from each other as possible, the ground loop can pickup external field noise. This can be motors, lights, dimmers..or, indeed, the loop can pickup the hum caused by the poweramp current. This can be 60 cycle, or it can be haversines.

Bringing the two cords closer, that reduces the total loop and it's sensitivity.

In one of my apps, I have an amp type QSC RMX 1450 on stage, and a simple numark mixer in a balcony 100 feet away (as the cord flies). I feed this amp using single ended line level runs, not balanced.. To keep hum and noise down, I wrap the line level runs around a 100 foot extension cord. This cord powers my mixer/cd set. With tens of kilowatts of lights in the house on dimmer, the house A/C, a toro 700 leaf blower on stage, and a 250 kilovolt vandygraf on stage, there is no induced noise in my system. (this drives the audio guy here nuts, he uses rf's, DI's, balanced, and xfmrs..)

Ground loops of this nature will cause shield currents in the IC's. While this in itself is not very bad, most amps are not designed to ignore this, even differential inputs.


IR drops, what would you expect of a house?? In fact, chances are you have a voltage gradient on your property caused by the power company.

This discussion also pertains to 2 wire setups, but the path the currents take is not so obvious since neutral/not is the only thing involved.

Cheers, John