ProSoundWeb Community

Sound Reinforcement - Forums for Live Sound Professionals - Your Displayed Name Must Be Your Real Full Name To Post In The Live Sound Forums => AC Power and Grounding => Topic started by: Kevin Graf on December 26, 2013, 02:14:54 PM

Title: Ground Loops and Tester
Post by: Kevin Graf on December 26, 2013, 02:14:54 PM
Ground Loops and Tester

A recent Belden Blog article:
'Ground Loops'
Posted by: Steve Lampen on November 12, 2013

****************************
From above blog a link to a rather expensive Ground Loop tester:
http://www.loopslooth.com/index.html

The 'Tutorial on Interference'  and 'Scientific Article' are interesting:
http://www.loopslooth.com/Tutorial-intro.html
http://www.loopslooth.com/Article.html
Title: Re: Ground Loops and Tester
Post by: Mike Sokol on December 26, 2013, 02:23:23 PM
Ground Loops and Tester

A recent Belden Blog article:
'Ground Loops'
Posted by: Steve Lampen on November 12, 2013

****************************
From above blog a link to a rather expensive Ground Loop tester:
http://www.loopslooth.com/index.html

The 'Tutorial on Interference'  and 'Scientific Article' are interesting:
http://www.loopslooth.com/Tutorial-intro.html
http://www.loopslooth.com/Article.html

Wow, doesn't look too simple. I have a way to accomplish the same thing with a $50 clamp ammeter and a hair drier or space heater load. Of course, my method doesn't look quite so fancy with an o-scope and everything else.

I'll see if I can write up my ground loop test procedure in a few weeks and publish a link here. Then you guys can try it for yourself and offer any suggestions on how to make it better. 
Title: Re: Ground Loops and Tester
Post by: Greg_Cameron on December 26, 2013, 02:25:43 PM
The thing is, ground loops exist everywhere and are inherent due to the nature of AC wiring. The problem in our world is when those loop currents induce noise into the audio interface due to improper circuit design. I don't see how these devices are going to help since you cannot eliminate ground loops in AC wiring. They only tell you what we already know to be true. You can "induce" ground loop current into any AC system when applying an AC inductive field around any ground conductor. The issues is preventing it from being rectified into audible noise in the audio circuit. These devices won't help with that.
Title: Re: Ground Loops and Tester
Post by: Mike Sokol on December 26, 2013, 02:34:13 PM
The thing is, ground loops exist everywhere and are inherent due to the nature of AC wiring. The problem in our world is when those loop currents induce noise into the audio interface due to improper circuit design. I don't see how these devices are going to help since you cannot eliminate ground loops in AC wiring. They only tell you what we already know to be true. You can "induce" ground loop current into any AC system when applying an AC inductive field around any ground conductor. The issues is preventing it from being rectified into audible noise in the audio circuit. These devices won't help with that.

Greg, in fact you can eliminate ground loops due to double-bonded G-N connections in sub-panels, which are a violation of the NEC. The fix can be as simple as taking out the green "bonding" screw which didn't belong there to begin with, and sometime installed by electricians and inspectors who don't know any better. Also, adding audio isolation transformers between gear will eliminate all ground loop currents and hum.

Also, ground loop hum is not "rectified" in the electrical sense of the word, but it is "coupled" magnetically or sometimes resistively inside gear with the pin-1 problem (which is sadly a lot). I speak from a lot of experience since I have a ground-loop maker which consists of a Glo-Melt resistance soldering transformer good for 3 volts and 40 amps output. I can create ground loop currents and measure the hum at will. So in the last few years of playing with this I've found a lot of new ways to measure and eliminate ground loop.

But more on this later. I'll detail my experiment then and provide data for all of you to review.
Title: Re: Ground Loops and Tester
Post by: John Roberts {JR} on December 26, 2013, 02:54:39 PM
Greg, in fact you can eliminate ground loops due to double-bonded G-N connections in sub-panels, which are a violation of the NEC. The fix can be as simple as taking out the green "bonding" screw which didn't belong there to begin with, and sometime installed by electricians and inspectors who don't know any better. Also, adding audio isolation transformers between gear will eliminate all ground loop currents and hum.
This is not an audio signal integrity issue. Outlet grounds are purely for human safety in case of faults. Single ground neutral bond at panel is again a safety consideration wrt to open neutral "and" open ground faults.
Quote
Also, ground loop hum is not "rectified" in the electrical sense of the word, but it is "coupled" magnetically or sometimes resistively inside gear with the pin-1 problem (which is sadly a lot). I speak from a lot of experience since I have a ground-loop maker which consists of a Glo-Melt resistance soldering transformer good for 3 volts and 40 amps output. I can create ground loop currents and measure the hum at will. So in the last few years of playing with this I've found a lot of new ways to measure and eliminate ground loop.

But more on this later. I'll detail my experiment then and provide data for all of you to review.
Giving this some thought I speculated about testing for this with a HF signal to hopefully differentiate between a local G-N bond and distant G-N bond due to inductance.

Hopefully pin 1 issues are slowly going away as older legacy gear gets replaced.

JR
 
Title: Re: Ground Loops and Tester
Post by: Greg_Cameron on December 26, 2013, 03:19:01 PM
Greg, in fact you can eliminate ground loops due to double-bonded G-N connections in sub-panels, which are a violation of the NEC. The fix can be as simple as taking out the green "bonding" screw which didn't belong there to begin with, and sometime installed by electricians and inspectors who don't know any better. Also, adding audio isolation transformers between gear will eliminate all ground loop currents and hum.

Agreed you reduce ground loop currents by eliminating double bonding, but the devices linked to here don't appear to be checking for that issue.


Greg, in fact you can eliminate ground loops due to double-bonded G-N connections in sub-panels, which are a violation of the NEC. The fix can be as simple as taking out the green "bonding" screw which didn't belong there to begin with, and sometime installed by electricians and inspectors who don't know any better. Also, adding audio isolation transformers between gear will eliminate all ground loop currents and hum.

Also, ground loop hum is not "rectified" in the electrical sense of the word, but it is "coupled" magnetically or sometimes resistively inside gear with the pin-1 problem (which is sadly a lot).

Yes, wrong terminology. Rectification to the audio band can occur with RF signal infiltration. Not ground loops. My mistake.

Greg
Title: Re: Ground Loops and Tester
Post by: Mike Sokol on December 26, 2013, 04:04:25 PM
Giving this some thought I speculated about testing for this with a HF signal to hopefully differentiate between a local G-N bond and distant G-N bond due to inductance.

I have a B&K 309 Ground Resistance Tester which does a Fall of Potential test using a higher than line frequency signal (600 Hz, I think) so it can be used on an live ground rod without having to power down. I suspect these guys are doing something similar. Also, I used to do resistance tolerance testing using a Kelvin bridge and a DC supply. But there's nothing to stop you from using any AC signal you like instead of a DC current source. You could use ultrasonic or even into the low RF range if you like, I think. But I'm not sure of the advantage of a higher frequency except to be able to gather data without shutting down a system. Got to think about this some more though.

Quote
Hopefully pin 1 issues are slowly going away as older legacy gear gets replaced.

My tests with current Mackie and Behringer active floor wedges suggests that the pin-1 problem is still alive and thriving in modern equipment. That's a shame since it causes so much grief in typical installations, and it's not terribly difficult to eliminate with proper design.

Title: Re: Ground Loops and Tester
Post by: Mike Sokol on December 26, 2013, 04:19:23 PM
This is not an audio signal integrity issue. Outlet grounds are purely for human safety in case of faults.

I agree, but my real problem with ground loops is that it makes musicians and engineers do crazy things to stop the hum, such as cutting off the safety ground pins on power cords and ground-lifting rack gear using those $1 ground eliminators from the big box stores. One of the churches I was in last year had the ground pins cut off of EVERY extension cord and quad box in the place. They warned me not to touch any microphone while leaning on the brass rail around the edge of the stage or I would get a "good shock". I measured around 100 volts from the PA system "safety ground" to actual earth potential. Yikes!!!

They claimed that was the only way to stop the PA system hum, but they created a death trap in the process and thought nothing of it.

On a slightly different tact I had a Tel-Com student at Ball State University tell me her parents had a refrigerator in the basement with a "messed up ground". She said you would get a shock if you touched the refrigerator door while standing on the concrete floor. Their fix was the "jump in the air" while opening the door so as not to get shocked (no kidding). When I told her she needed to fix it before an unsuspecting repair guy or visitor was electrocuted, she told me it had been like that for years and "everybody" knew not the touch the refrigerator. I see and hear crazy stuff like this all the time.

IMHO Ground Lift Adapters are EVIL...

Title: Re: Ground Loops and Tester
Post by: John Roberts {JR} on December 26, 2013, 04:29:01 PM


My tests with current Mackie and Behringer active floor wedges suggests that the pin-1 problem is still alive and thriving in modern equipment. That's a shame since it causes so much grief in typical installations, and it's not terribly difficult to eliminate with proper design.
That is disappointing.. I used to manage an engineering group and new engineers are not born (or graduated) with full knowledge, but engineering management should know this stuff and inform them.

JR
Title: Re: Ground Loops and Tester
Post by: Mike Sokol on December 26, 2013, 04:29:56 PM
Yes, wrong terminology. Rectification to the audio band can occur with RF signal infiltration. Not ground loops. My mistake.

But I think you might be right about rectification for switch-mode ballasts in some new lighting technologies. I did a gig in a church a few years ago with some sort of modern arc lights in ceiling pots. They generated so much broad-band RF that my wireless mic receiver would do a full level noise burst whenever the  belt pak RF was turned off or experienced a drop-out when I walked too far from the antenna. We proved the lights were doing it by shutting them all off with the RF problem going away. Now, I didn't experience any other RF ground loop issues that I know of, but now I'm going to be on the lookout for them. Also, as I'm sure you know, any corrosion on a ground/shield connector can act like a radio detector (rectifier) and demodulate AM radio stations near your sound system. Been there, and got the t-shirt (from the radio station).   

Gosh, now I've got to worry about RF ground loop rectification. When does it all end?
Title: Re: Ground Loops and Tester
Post by: John Roberts {JR} on December 26, 2013, 09:04:59 PM
I agree, but my real problem with ground loops is that it makes musicians and engineers do crazy things to stop the hum, such as cutting off the safety ground pins on power cords and ground-lifting rack gear using those $1 ground eliminators from the big box stores. One of the churches I was in last year had the ground pins cut off of EVERY extension cord and quad box in the place. They warned me not to touch any microphone while leaning on the brass rail around the edge of the stage or I would get a "good shock". I measured around 100 volts from the PA system "safety ground" to actual earth potential. Yikes!!!

They claimed that was the only way to stop the PA system hum, but they created a death trap in the process and thought nothing of it.

On a slightly different tact I had a Tel-Com student at Ball State University tell me her parents had a refrigerator in the basement with a "messed up ground". She said you would get a shock if you touched the refrigerator door while standing on the concrete floor. Their fix was the "jump in the air" while opening the door so as not to get shocked (no kidding). When I told her she needed to fix it before an unsuspecting repair guy or visitor was electrocuted, she told me it had been like that for years and "everybody" knew not the touch the refrigerator. I see and hear crazy stuff like this all the time.

IMHO Ground Lift Adapters are EVIL...

For accuracy those are 2 to 3 circuit adapters. When the screw tab is attached to a grounded junction box via the faceplate screw they (are supposed to) complete the safety ground path. I thought I was being a good citizen by using those on the 2 pin outlets on my living room where the back line plugged in for my house parties. Only years later did i realize that the box and screw was floating so my effort was a complete waste of time. :-( 

The good news is no musicians were lost to my ungrounded outlets.

JR

PS: I just now recalled a halloween party at a co-workers house in the early '70s (in CT) where there was a ham and party snacks presented with some dry ice and a back light under some glass. As I recall there was also a light shock present if you touched the ham.. I suspected at the time that maybe jim did it,  so the ham would last longer.  :o
Title: Re: Ground Loops and Tester
Post by: Mike Sokol on December 27, 2013, 12:21:07 AM
PS: I just now recalled a halloween party at a co-workers house in the early '70s (in CT) where there was a ham and party snacks presented with some dry ice and a back light under some glass. As I recall there was also a light shock present if you touched the ham.. I suspected at the time that maybe jim did it,  so the ham would last longer.  :o

Wow, electric ham is a new one on me. But it was the 70's, so a lot of other "electric" stuff was probably happening.  ;D

If you want to see some really crazy threads, snoop around in the RV forums where some old guy who seems to know everything tells you to expect to feel a little tingle from your RV while standing on the wet ground. Does nobody know how to test a receptacle ground?
Title: Re: Ground Loops and Tester
Post by: Jonathan Johnson on December 27, 2013, 01:46:13 AM
In another thread (I can't remember which one) someone (might've been Mike) mentioned that conduit used to be considered an acceptable ground path, but that there can be ground loop currents present because the conduit may be grounded in multiple places where it's attached to the metal framing of a building, rather than a hub-and-spoke (or star) configuration where the conduit is grounded only at the service panel.

Many (if not most) new conduit installations include a green insulated or bare grounding wire. However, this may subject to the same ground loop problems as conduit, because that grounding wire will be bonded to nearly every device and electrical box in its path, unless it is a properly installed isolated ground conductor.
Title: Re: Ground Loops and Tester
Post by: Lyle Williams on December 27, 2013, 03:31:41 AM
Gosh, now I've got to worry about RF ground loop rectification. When does it all end?

Stay away from RF.  Knowing the truth will send you mad. 

Title: Re: Ground Loops and Tester
Post by: Mike Sokol on December 27, 2013, 09:15:21 AM
In another thread (I can't remember which one) someone (might've been Mike) mentioned that conduit used to be considered an acceptable ground path, but that there can be ground loop currents present because the conduit may be grounded in multiple places where it's attached to the metal framing of a building, rather than a hub-and-spoke (or star) configuration where the conduit is grounded only at the service panel.

Yeah, that would be me. The problem with metallic conduit is that there can be HUGE currents circulating through building steel, I'm guessing in the hundreds or even thousands of amperes at times. Of course, this is normally spread out over many parallel paths (all those I-beams) so the individual current paths are pretty low. If you have two sub-panels bonded to different areas of the building, then there will almost certainly be a differential voltage between them. And connecting an XLR cable between gear plugged into these two different grounding sources will result in the XLR cable trying to equalize that voltage differential. Because this is a very low-impedance current source, the only limiting factor will be the resistance/impedance of your XLR cable. I've measured about 1 amp of current per volt of ground loop differential in a typical XLR cable, and up to 5 volts ground loop differential voltage. Once those equaling currents are circulating in your XLR cable shields, then each piece of gear that has the pin-1 problem to any degree will hum. 
Quote

Many (if not most) new conduit installations include a green insulated or bare grounding wire. However, this may subject to the same ground loop problems as conduit, because that grounding wire will be bonded to nearly every device and electrical box in its path, unless it is a properly installed isolated ground conductor.

Yeah, but I've found a brand new installation (a year ago) with hard-bonded conduits feeding sub-panels, the sub-panel bolted to the building steel, and the Neutral-Ground bonding screw installed in the sub-panel isolated neutral bus (a big code violation). The sound-tech swore that the panel grounding couldn't be causing his hum because they had installed orange ground-isolated receptacles, and it had just passed inspection a few months back. However, he was also a journeyman electrician and promised to open up the panel the following week for a look-see. He then reported back that every ground bonding possibility had been done on purpose by the local inspector who insisted that all those ground bonds would make a "better ground".

Rather than fight the inspector (and local code) he just added audio isolation transformers in the ground loop paths he was able to verify with a clamp ammeter, and all was well.
Title: Re: Ground Loops and Tester
Post by: John Roberts {JR} on December 27, 2013, 10:05:31 AM
Does nobody know how to test a receptacle ground?

I ASSumed my cold water pipes were ground, but further testing reveals that isn't so reliable.

After my water main broke, we replaced it with PVC. There is a lot of iron in my tap water... My water sediment filters turn red/brown after only a few months so perhaps the water is mildly conductive. I can measure that too... Sticking my VOM probes into a cup of tap water measures 1 MOhm. The water from my RO filter measures 5 MOhm so the filter is doing some good. Still not sure my pipes are well grounded. I can measure that too... between my cold water feed to my washing machine and a screw on the front of my fuse box, I get a momentary reading then flashing OF with my VOM. This suggests some capcitance and some AC voltage, but no dominant resistance path.  :o :o :o  But I already knew my outlets are not grounded, that's why I installed GFCI in bathroom and kitchen, but now maybe I don't need to worry as much about the plumbing providing a dangerous ground path.   8)

Thanx, I just learned a little more about my crappy house wiring.

JR

PS: At least I determined that my power drop is grounded to the actual earth, several months ago when I received a shock from using the killer (hot ground swap) extension cord in my damp yard.
Title: Re: Ground Loops and Tester
Post by: Mike Sokol on December 27, 2013, 11:44:34 AM
Thanx, I just learned a little more about my crappy house wiring.

Since I've really been looking at AC power for sound systems over the last several years, I've learned not to trust ANYTHING. Sometimes the most professional looking wiring can give you a big surprise, while other times an outlet that looks very sketchy will test out perfectly fine. I now take my SureTest or Amprobe INSP-3 Ground Loop Impedance Testers (GLIT) out on sound gigs just to snoop around on all the outlets. Since I've also seen a few shady looking receptacles at hotels, I'm going to put a GLIT and a NCVT in my luggage and do some testing just for grins. If I find anything interesting I'll take pictures and post them here.
Title: Re: Ground Loops and Tester
Post by: Jonathan Johnson on December 27, 2013, 11:48:44 AM
My water sediment filters turn red/brown after only a few months so perhaps the water is mildly conductive.

Lucky you. I have to change my sediment filter element every 3 weeks. (It's a standard 2.5" dia. x 9.75" long string wound cartridge, 10 micron.) I haven't measured the conductivity of the water, but about a year and a half ago the jet in the pump plugged up with some larger metallic chunks.

There is copper, some steel, and some plastic pipe in my house. Besides a grounding rod, the service is also bonded to the cold water pipe at the water heater. However, about 15 feet away, the cold water pipe transitions to a length of plastic. Somewhere beyond the plastic is more copper pipe, which is not bonded. Somewhere there's a chunk of steel in the mix. Neither is the copper hot water pipe bonded. Where the pipe enters the earth, it's plastic.

Really, all the sections of metal pipe should be bonded, not to provide a ground path for the electrical system, but to equalize the voltage potential on those sections. This can be especially important with gas piping, to reduce the chances of a static electric discharge creating a spark in the presence of a gas/oxygen mix. (At least, I think that's the theory behind bonding all that stuff.)
Title: Re: Ground Loops and Tester
Post by: John Roberts {JR} on December 27, 2013, 12:32:24 PM
Lucky you. I have to change my sediment filter element every 3 weeks. (It's a standard 2.5" dia. x 9.75" long string wound cartridge, 10 micron.) I haven't measured the conductivity of the water, but about a year and a half ago the jet in the pump plugged up with some larger metallic chunks.

I probably need to change them more often than I do... I have a little rube-goldberg plumbing setup where the bypass water from my RO filter gets piped two rooms away into the toilet flush tank (I feel green). I can see this clear tubing turn rusty when filter gets clogged.
Quote

There is copper, some steel, and some plastic pipe in my house. Besides a grounding rod, the service is also bonded to the cold water pipe at the water heater. However, about 15 feet away, the cold water pipe transitions to a length of plastic. Somewhere beyond the plastic is more copper pipe, which is not bonded. Somewhere there's a chunk of steel in the mix. Neither is the copper hot water pipe bonded. Where the pipe enters the earth, it's plastic.

Really, all the sections of metal pipe should be bonded, not to provide a ground path for the electrical system, but to equalize the voltage potential on those sections. This can be especially important with gas piping, to reduce the chances of a static electric discharge creating a spark in the presence of a gas/oxygen mix. (At least, I think that's the theory behind bonding all that stuff.)

I am not sure if I would see any practical benefit from ground bonding my water pipes.

Bonding gas pipes is outside my area of expertise. Does flowing gas create static, that meeds to bleed off? If not grounding the pipes could make them a spark magnet from any nearby charged up individual (while the floating pipes may draw a spark anyhow if large enough capacitance). 

JR