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: Bill McIntosh on November 13, 2017, 12:07:06 PM

Title: 2 volts too much?
Post by: Bill McIntosh on November 13, 2017, 12:07:06 PM
We recently had a touring Christian Band You May Have Heard Of play at our church.  The rider called for 2 x 100 amp connections 120 volt 3 phase.  Our electrical contractor tied their tails into panels.

The band's tech lead was reluctant to connect the tails to the PA distro - he measured 2 volts between neutral and ground.  They connected after a discussion with our electrician and show went with no issues.

Is 2 volts a concern that we need to resolve? 



Sent from my LG-D800 using Tapatalk

Title: Re: 2 volts too much?
Post by: Scott Helmke on November 13, 2017, 02:03:12 PM
Rule of thumb in the live sound world is that 2 volts will mean a noisy/buzzy show.
Title: Re: 2 volts too much?
Post by: John Roberts {JR} on November 13, 2017, 02:29:11 PM
not a human hazard but evidence of a problem.

JR
Title: Re: 2 volts too much?
Post by: Mike Sokol on November 13, 2017, 07:51:12 PM
Is 2 volts a concern that we need to resolve? 

In any subpanel with a load on it, there will nearly ALWAYS be 1 or 2 volts between the ground and neutral. That's because the ground and neutral should NOT be bonded at the subpanel, only at the incoming service panel. So if you have 4 volts drop on that panel due to a heavy load, then typically 2 of those volts are on the line conductor, and the other 2 volts are the drop on the neutral conductor. If you measure between the neutral and the ground on that subpanel you'll come up with 2 volts. I get worried when I measure EXACTLY 0 volts between neutral and ground on a subpanel because that suggests a secondary neutral-ground bond on the subpanel. That 0 volts is because you've now contaminated the ground with the voltage fluctuations on the neutral due to load variations. And THAT'S what causes ground loop hums.
Title: Re: 2 volts too much?
Post by: Stephen Swaffer on November 13, 2017, 08:52:05 PM
A bit a swerve-but relevant I think?  Can you even get a UL listing for equipment that has neutral connected to a chassis?  That would create a secondary ground/neutral bond?  I know chassis tied to neutral's used to be standard fare.  If there is no connection the voltage difference is irrelevant.

I agree with Mike-a 20 amp imbalance with 1 ohm of resistance in a neutral = 2.0 volts differential.  That is pushing good design practice for voltage drop in a circuit.
Title: Re: 2 volts too much?
Post by: Mike Sokol on November 13, 2017, 09:55:37 PM
I agree with Mike-a 20 amp imbalance with 1 ohm of resistance in a neutral = 2.0 volts differential.  That is pushing good design practice for voltage drop in a circuit.
I drew this graphic up a while ago to help me better understand the various voltage drops in a typical power distro system, and what certain measurements could possibly mean. I try to go into every voltage or current test situation with at least a guess of what measurement to expect. If the voltage measured doesn't line up with my expectations, then there's a hint as to what's wrong. For instance, as you load a typical home branch circuit with more current draw and approach 20 amperes, you'll likely come close to a 5% voltage drop. So 120 volts drops by 6 volts to 114 volts. But only 3 of those volts are on the hot/line conductor. The neutral conductor does an equal and opposite "rise" in voltage, so it will create a voltage difference if you measure between it and the grounding conductor. This basic knowledge is essential to troubleshooting any grounding system. That's because all wires have resistance. And resistance always results in a voltage drop if there's any current draw. It's a fundamental law of electricity. 
Title: Re: 2 volts too much?
Post by: Mike Sokol on November 14, 2017, 05:52:33 AM
A bit a swerve-but relevant I think?  Can you even get a UL listing for equipment that has neutral connected to a chassis?  That would create a secondary ground/neutral bond?  I know chassis tied to neutral's used to be standard fare.  If there is no connection the voltage difference is irrelevant.

That is correct. However, while there's no hard connection between the neutral and ground in any modern gear, there often is some sort of capacitor based RF circuit between the neutral and ground to reduce RF generation. I believe that this high-frequency path can allow transient currents in the line and neutral to be transferred into the grounding system.
Title: Re: 2 volts too much?
Post by: Bill McIntosh on November 14, 2017, 08:53:32 AM
I drew this graphic up a while ago to help me better understand the various voltage drops in a typical power distro system, and what certain measurements could possibly mean. I try to go into every voltage or current test situation with at least a guess of what measurement to expect. If the voltage measured doesn't line up with my expectations, then there's a hint as to what's wrong. For instance, as you load a typical home branch circuit with more current draw and approach 20 amperes, you'll likely come close to a 5% voltage drop. So 120 volts drops by 6 volts to 114 volts. But only 3 of those volts are on the hot/line conductor. The neutral conductor does an equal and opposite "rise" in voltage, so it will create a voltage difference if you measure between it and the grounding conductor. This basic knowledge is essential to troubleshooting any grounding system. That's because all wires have resistance. And resistance always results in a voltage drop if there's any current draw. It's a fundamental law of electricity.
Thanks Mike.  There was no load on that sub panel but there were other loads on that service.  So the small potential he measured in the sub panel is to be expected.  A higher voltage could be a red flag requiring investigation.

Sent from my LG-D800 using Tapatalk

Title: Re: 2 volts too much?
Post by: Mike Sokol on November 14, 2017, 10:25:25 AM
Thanks Mike.  There was no load on that sub panel but there were other loads on that service.  So the small potential he measured in the sub panel is to be expected.  A higher voltage could be a red flag requiring investigation
Exactly. Most electricians and even electrical engineers don’t understand this basic principal. But now you know...
Title: Re: 2 volts too much?
Post by: Jonathan Johnson on November 14, 2017, 03:41:20 PM
A couple of things I noted in this discussion:

So, to the first point, it appears that electrical safety education may be getting out there. To the second point, it appears that more education is needed. Maybe that's covered in "Entertainment Electrics 102."
Title: Re: 2 volts too much?
Post by: Mike Sokol on November 14, 2017, 04:34:08 PM
That 0 volts is because you've now contaminated the ground with the voltage fluctuations on the neutral due to load variations. And THAT'S what causes ground loop hums.
And that's what can also cause GLID (Ground Loop Intermodulation Distortion). None of you here believe that's a real thing, but I'm going to set up a demonstration to prove it.  ;)
Title: Re: 2 volts too much?
Post by: Jonathan Johnson on November 20, 2017, 01:49:51 AM
I drew this graphic up a while ago to help me better understand the various voltage drops in a typical power distro system, and what certain measurements could possibly mean. I try to go into every voltage or current test situation with at least a guess of what measurement to expect. If the voltage measured doesn't line up with my expectations, then there's a hint as to what's wrong. For instance, as you load a typical home branch circuit with more current draw and approach 20 amperes, you'll likely come close to a 5% voltage drop. So 120 volts drops by 6 volts to 114 volts. But only 3 of those volts are on the hot/line conductor. The neutral conductor does an equal and opposite "rise" in voltage, so it will create a voltage difference if you measure between it and the grounding conductor. This basic knowledge is essential to troubleshooting any grounding system. That's because all wires have resistance. And resistance always results in a voltage drop if there's any current draw. It's a fundamental law of electricity.

Let me see if I can simplify it.

Measuring between hot and neutral, you are measuring against a loaded neutral. The load on the neutral causes a voltage drop.

Measuring between hot and ground, you are measuring against an unloaded ground. There is no voltage drop on the ground.

Voltage drop is a function of wire resistance and load. Where there is no load, wire resistance is mostly irrelevant for a high impedance voltmeter. The voltage drop on the hot wire is irrelevant in this comparative analysis; it doesn't make a difference in the math because we are not measuring the voltage differential between the beginning and end of the hot wire. (But that is what we are effectively doing with the neutral wire.)

The difference between the hot-neutral voltage and the hot-ground voltage should equal the neutral-ground voltage reading. The neutral-ground reading is equal to the voltage drop on the neutral.
Title: Re: 2 volts too much?
Post by: Mike Sokol on November 20, 2017, 08:17:37 AM
The difference between the hot-neutral voltage and the hot-ground voltage should equal the neutral-ground voltage reading. The neutral-ground reading is equal to the voltage drop on the neutral.

Exactly. And that's why you don't want to double-bond the ground wires to the neutral at the subpanel, or bootleg ground any receptacle. Because if there's significant current in the neutral, which causes a voltage drop across it's length, then the ground wire will be sharing this current and participate in the voltage drop. You can easily see this if you run a long test wire between two receptacle grounds. If both grounds are properly bonded only at the service panel, and properly isolated from neutral, and not accidentally bonded to building steel, then you should measure very close (within a few mV) to 0 volts between them, no matter what the load. However, if one of them has been improperly connected to neutral or building steel, then you can watch the ground voltage difference between them move up and down as you load one circuit or the other.

I've fought ground loop hum that would come and go, depending on loads like kitchen appliances being turned on in a church. For example, when the big coffee urn starts pulling 15 amps, there's likely a 4 or 5 volt drop in its branch circuit with 120volts from Hot to Neutral dropping to maybe 115 volts. If the ground in the sub-panel or branch circuit is contaminated with a connection to the neutral, now the receptacle you plugged your mixer into at the back of the room (and connected to the kitchen branch circuit), can have its ground wire potential shift by 2 volts or more as the loads change. Since your stage power is probably on another circuit and maybe even a different sub-panel, you can get a ground-loop hum that comes and goes as the coffee pot thermostat cycles on and off. I had this exact scenario happen once which drove me crazy for hours until I figured it out.
Title: Re: 2 volts too much?
Post by: Stephen Swaffer on November 20, 2017, 12:55:02 PM
If both grounds are properly bonded only at the service panel, and properly isolated from neutral, and not accidentally bonded to building steel, then you should measure very close (within a few mV) to 0 volts between them, no matter what the load. However, if one of them has been improperly connected to neutral or building steel, then you can watch the ground voltage difference between them move up and down as you load one circuit or the other.


If you are dealing with a steel frame bulding and a diligent inspector, you will almost never find a ground that is only connected at the panel.  J-boxes are generally attached to steel and code requires them to be bonded to any grounding conductor in them.  As a matter of technicality, usually most red iron in a building is part of the grounding electrode and as such it is an acceptable ground.  If there is current in the grounding wire/building that is causing a voltage gradient that is causing trouble, then by definition that is objectionable current and a code violation.

Theory, code and real world are often at odds, though!

The neutral of course, is another story.
Title: Re: 2 volts too much?
Post by: Jonathan Johnson on November 20, 2017, 03:17:48 PM
If you are dealing with a steel frame bulding and a diligent inspector, you will almost never find a ground that is only connected at the panel.  J-boxes are generally attached to steel and code requires them to be bonded to any grounding conductor in them.  As a matter of tecnicailty, usually most red iron in a building is part of the grounding electrode and as such it is an acceptable ground.  If there is current in the grounding wire/building that is causing a voltage gradient that is causing trouble, then by definition that is objectionable current and a code violation.

Theory, code and real world are often at odds, though!

The neutral of course, is another story.

I don't think multiple bonds of grounding wires (think of a mesh) is a problem. After all, the shields on the cable between the mixer and the amplifiers are typically grounded on both ends to the equipment chassis, which is in turn connected to the ground wire of their respective power cords, which may be on different circuits and the ground wires are connected back in the panel. That's at least two potential ground paths. And that's quite normal, and not a problem as long as there's no current or voltage potential on the ground wire.

The problem is when ground and neutral are bonded at multiple points. That causes currents to flow on the ground wire. If the current path happens to include the shield of a link between two pieces of audio equipment, you have ground current induced hum. The neutral and ground should be bonded at one point only.
Title: Re: 2 volts too much?
Post by: Mark Wilkinson on November 20, 2017, 07:00:21 PM
.................

The problem is when ground and neutral are bonded at multiple points. That causes currents to flow on the ground wire. If the current path happens to include the shield of a link between two pieces of audio equipment, you have ground current induced hum. The neutral and ground should be bonded at one point only.

Thank you guys, between Mike and you, this has become very clear.
Title: Re: 2 volts too much?
Post by: Bill McIntosh on November 20, 2017, 08:24:47 PM
Thank you guys, between Mike and you, this has become very clear.

Thanks from me too. 

Also I want to clarify that the tour tech did not make a fuss, just showed the meter readings to our electrical contractor.  They discussed, tour tech was happy and the show went as scheduled with no audible issues.
Title: Re: 2 volts too much?
Post by: Mike Sokol on December 07, 2017, 12:14:36 PM
FYI: This thread made it onto the back page of the December issue of Live Sound International magazine. We should be able to walk into any bar in the world and get a free drink.... Or maybe not.
Title: Re: 2 volts too much?
Post by: Rob Spence on December 07, 2017, 12:46:06 PM
Cool!


Sent from my iPad using Tapatalk Pro
Title: Re: 2 volts too much?
Post by: Mike Sokol on December 07, 2017, 06:28:08 PM
FYI: This thread made it onto the back page of the December issue of Live Sound International magazine.

I'm extremely thankful to you guys (and gals) for bringing your intellect and experience to these sometimes complex topics. I, for one, find them a great way to increase my understanding of electrical systems of all types.
Title: Re: 2 volts too much?
Post by: Frank DeWitt on December 13, 2017, 09:01:28 PM
while we are all thinking about voltage drop on the neutral, vs the ground think about what happens when a surge protector does its thing.  A giant surge hits the AC feed to the booth.  The surge protector connects that crazy high voltage on both hot and neutral to ground.  Current flows through the ground developing a voltage.  Let's say 100 volts for just an instant. No problem.  The equipment has been saved.

But. There is a audio cable from the mixer in the booth to the amp on the platform.  Now the amp that didn't get a surge sees 100 volts coming down pin one.  It's own ground is (or was) at zero.  At this point bad stuff happens

Answer,  a surge protector like Surge-X that clamps the hot and neutral together. 
Title: Re: 2 volts too much?
Post by: John Roberts {JR} on December 14, 2017, 09:22:43 AM
while we are all thinking about voltage drop on the neutral, vs the ground think about what happens when a surge protector does its thing.  A giant surge hits the AC feed to the booth.  The surge protector connects that crazy high voltage on both hot and neutral to ground.
High voltage on Neutral?
Quote
Current flows through the ground developing a voltage.  Let's say 100 volts for just an instant. No problem.  The equipment has been saved.

But. There is a audio cable from the mixer in the booth to the amp on the platform.  Now the amp that didn't get a surge sees 100 volts coming down pin one.  It's own ground is (or was) at zero.  At this point bad stuff happens
pin one "should" be bonded to chassis ground and safety ground...So no problemo.
Quote
Answer,  a surge protector like Surge-X that clamps the hot and neutral together.
Safety ground is supposed to be the ultimate sewer for fault currents. 

Clamping line to neutral seems OK, while it could cause neutral to rise significantly wrt ground, but neutral is floating inside products so no problemo.

JR