Go Down

Topic: Can anyone explain a ground loop, and when tieing grounds together is needed. (Read 1 time) previous topic - next topic

carl1864

I've heard about the issues of ground loops many times.  There have been times I've heard it is necessary to tie the grounds of multiple devices together, and other times when it isn't.  However after much googling, I surprisingly have not found much info on this topic.

Can anyone explain in pretty basic terms what a ground loop is?  What bad effects it causes?  And how do you determine when it is necessary or not, to tie multiple grounds together?

westfw

A ground loop happens when grounds of two systems are tied together in more than one place, usually once through the chassis ground (via the third prong in your power outlet)  and once through a signal cable.  It's a bad thing if the ground potential is slightly different between the two boxes, because current will flow in places it shouldn't.

It's almost always necessary to tie grounds together; otherwise signals have no reference value.  For some types of circuit (notably audio/visual equipment), it's apparently desirable to only connect grounds together ONCE, by making most of the signal cables only connect to ground at one end.  (A concert or studio setup with a dozen mikes, plus speaker amps, plus recorders, plus lighting control, plus ... has a LOT of opportunities for bad behavior of the grounded parts of cables...)

I guess the "single point grounds" suggested for HF circuitry (high sensitivity A-D circuits, switching power supplies. etc) is a small scale version of the same thing, although in this case it is usually the inability of wires to behave "ideally" at the frequencies involved  that is the problem.

MarkT

Its a big issue for small signal analog circuitry - the power supply currents flowing through the ground wires create small voltages (V = IR) that interfere with the small signals you are trying to measure/amplify.  Typical example is microphone amplifier with mains hum.

For all digital circuitry its not usually relevant as the voltages are too small.

For when you mix analog and digital it becomes important - you can end up with the really noisy digital supply currents creating voltages across the sensitive analog parts of the circuit.  The normal approach is to keep analog and digital ground and power separate, except that the grounds are tied together at one point (not on a high-current path).  The two grounds are ideally coupled with an RF-choke to reduce noise-injection at high-frequencies.

For RF work the issue becomes one of providing a low-inductance ground plane since inductance dominates resistance.

There is a very specific issue with large area loops near a transformer - the leaking magnetic flux induces mains frequency signal in the loop.
[ I won't respond to messages, use the forum please ]

cr0sh

Here's a link to a Popular Science article which explains ground loops (and other issues), from a 1958 hi-fi electronics (tube-based) perspective:

http://goo.gl/cHJXz
I will not respond to Arduino help PM's from random forum users; if you have such a question, start a new topic thread.

Schmidtn

Scroll down to page 209, number three in the pink box.  Don't look inside your car battery using a match for light.  Thanks for the tip PopSci! :P

On a related note.  When I was on a ship in the navy we had acquisition antennas and direction finding antennas.  Long cables can act as resistors (the ship was 820 foot long and you could very rarely take a straight shot to wherever you were trying to go) those DF antenna arrays could be a pain that way with ground loops.  We did a yards period where all our antennas went off ship and new cables were strung.  Recalibrating everything sucked pretty hard and we did it with software, not even with pinching cables to make notch filters.

cr0sh

Scroll down to page 209, number three in the pink box.  Don't look inside your car battery using a match for light.  Thanks for the tip PopSci! :P


That's pretty funny - I think I missed that when I was reading that issue.

I've been reading the 1950s decade, starting from January 1950, for the past month and half or so; right now I am on August 1959. Reading the articles, you wouldn't believe some of the really insane (to our age) things people were doing. I think one of the "funniest" was the recommendation to put drops of mercury into the distribution posts of the distributor before plugging the spark plug cables in, to assure a good contact between the cable tip and the post contact! A future issue had an admonition from another reader not to do that. Why? Perhaps because the mercury would make a mess when you removed the cap? Or maybe it was an environmental hazard?

Nope - neither: Because it would corrode the brass fittings on the cable ends and in the distributor posts! Which I am sure is true, but they really weren't thinking about mercury being an issue. Same with asbestos (that was a complete unknown). Radiation had a "two-faced" reaction, at least from the vantage point of this magazine; on the one hand, nuclear power had to be carefully managed and dealt with (shielding, design, etc) - on the other hand, nuclear power (and radiation therapies, etc) were the "great-new-thing" (tm) - it seemed like it wasn't new unless it had some link to transistors or nuclear power (oh - another big thing during the mid-to-late 1950s for amateurs seems to have been hunting for uranium deposits - geiger counter kit ads and articles on how to build and use them have appeared regularly in the latter half of the decade).

:)
I will not respond to Arduino help PM's from random forum users; if you have such a question, start a new topic thread.

Schmidtn

Ha!  For some reason when you said hunting for Uranium deposits was popular in the late 1950's the first thing that came to mind was Ward Cleaver smoking a pipe and telling Beaver to be careful with his Geiger Counter and to stay out of trouble while Beaver skips out the front door with one of his buddies on a Uranium hunt. 

Can't wait for people 50 years from now to look back at us and think, "What were they thinking?! Didn't they know how toxic handling transistors is?" Or whatever we're doing now that we really shouldn't be.

Grumpy_Mike

Quote
Or whatever we're doing now that we really shouldn't be.

I think we know. Watching TV and eating fast food.

cr0sh


Ha!  For some reason when you said hunting for Uranium deposits was popular in the late 1950's the first thing that came to mind was Ward Cleaver smoking a pipe and telling Beaver to be careful with his Geiger Counter and to stay out of trouble while Beaver skips out the front door with one of his buddies on a Uranium hunt. 

Can't wait for people 50 years from now to look back at us and think, "What were they thinking?! Didn't they know how toxic handling transistors is?" Or whatever we're doing now that we really shouldn't be.


Oh, yeah - that's another thing - invariably, at least up through about 1958 or so, every man (unfortunately, that honestly is every "white" man - but that's another topic for another thread - that and the mysoginistic viewpoints from the era) picture seems to almost always have a pipe in their mouth. There's one interesting article about underground RF coaxial lines being spliced "delicately" by a repair guy; he's got a pipe in his mouth down in the trench!

Then - I guess the ball started to roll on how tobacco and smoking was causal for lung cancer and other ailments, and the number of those images have gone down past about 1958 - but they still pop up occasionally.

I'm going to continue to read the issues forward in time until the following three things occur: Model Garage disappears, Wordless Workshop disappears, and "how-to" articles disappear; I'm pretty sure they happen in that order, too, and the how-to articles likely drop out sometime around 1978 or so...

Finally - Grumpy_Mike: I think you've hit the nail on the head, so to speak; though likely more so on the latter than the former (western society loves it TV, and other than seemingly making some of us dumber, as far as we know it doesn't cause cancer or other illness).

:)
I will not respond to Arduino help PM's from random forum users; if you have such a question, start a new topic thread.

Grumpy_Mike

Quote
and other than seemingly making some of us dumber

It is a bit more that that it is a sort of passyfier that saps the will and reduces the ability to think clearly and critically. I find a night in front of the TV makes you a lot more tired than almost any other activity.
I have done a lot of work for digital TV companies designing set top boxes and distribution systems and sometimes looking at the channels that were being distributed I felt like I was a heroin dealer.
I think it was Marx who said that the Church was the opiate of the people, that's nothing to TV.

Go Up