Continuous current on an equipment ground conductor.

[wilson9254]

We have an equipment ground conductor, 500kcmil, extended with a feeder from the service disconnect to a MLO switchboard. The feeder is sized for 4000 amperes all though the load is less than 500 amperes. The feeder consists of parallel sets of conductors that are layed within the wire way. The conductors are haphazardly placed within the wire way, that is to say the conductors are not grouped A-B-C-N. We are measuring 140-200A on a continuous basis on the equipment ground conductor. I am looking for comments as to whether this may be induced or is it the result of a cross connected neutral-ground in the facility distribution system or is it something I may be totally not paying attention to. The system voltage is 480/277. swilson@ktagroup.com

 

[charlie b]

Re: Continuous current on an equipment ground conductor.

Do you mean to say that you have all phase conductors, all neutral conductors, and the single EGC in the same raceway? If so, you must have a minimum of 9 conductors per phase (taking into account the derating for more than 3 current-carrying conductors). If the EGC is closer to the conductors of one of the phases than it is to the conductors of the other two phases, then the current you describe may very well be the result of electromagnetic induction.

There may be several ways to address this issue. But let?s start with a more detailed description of the setup. How many conductors of what size are run in how many raceways? Have you measured currents on the phases or on the neutral? Is the 500 amp load a measured value or the calculated service load? Have you confirmed that the neutral bus is not bonded to the ground bus at the MLO switchboard? Is this a new installation?

 

[brian john]

Re: Continuous current on an equipment ground conductor.

Steve:

It is my understanding this issues was resolved, if this is so could you please inform us what was the issue.

Brian Gaquin

 

[wilson9254]

Re: Continuous current on an equipment ground conductor.

Originally posted by charlie b:
Do you mean to say that you have all phase conductors, all neutral conductors, and the single EGC in the same raceway? If so, you must have a minimum of 9 conductors per phase (taking into account the derating for more than 3 current-carrying conductors). If the EGC is closer to the conductors of one of the phases than it is to the conductors of the other two phases, then the current you describe may very well be the result of electromagnetic induction.

There may be several ways to address this issue. But let?s start with a more detailed description of the setup. How many conductors of what size are run in how many raceways? Have you measured currents on the phases or on the neutral? Is the 500 amp load a measured value or the calculated service load? Have you confirmed that the neutral bus is not bonded to the ground bus at the MLO switchboard? Is this a new installation?

All of the conductors are in a single metallic wire way. The feeder is made up of 10 sets: (3)600kcmil, (1)350kcmil. There is one equipment ground conductor, size 500kcmil. Appearently I was mistaken about the facility load being about 500 amperes, it is actually 900-1000amperes measured. It was confirmed that the neutral bus was not bonded to ground at the MLO switchboard and yes it is a new installation.

 

[wilson9254]

Re: Continuous current on an equipment ground conductor.

Originally posted by brian john:
Steve:

It is my understanding this issues was resolved, if this is so could you please inform us what was the issue.

Brian Gaquin

The condition has been diagnosed as an induced current on the equipment ground conductor. The fix has been to remove the equipment ground from the wireway. This is only permissible though if the wire way is UL listed for use as an equipment ground. I do not know the status of that aspect.

 

[don_resqcapt19]

Re: Continuous current on an equipment ground conductor.

Something is not right here. I don't think we have all of the details. I find it very hard to believe that there is an induced current on the EGC that is 20% of the load current.
Don

 

[wilson9254]

Re: Continuous current on an equipment ground conductor.

Originally posted by don_resqcapt19:
Something is not right here. I don't think we have all of the details. I find it very hard to believe that there is an induced current on the EGC that is 20% of the load current.
Don

See response to Charlie B.

 

[karl riley]

Re: Continuous current on an equipment ground conductor.

Ordinarily I would snake a flexible ammeter probe around all conductors except the EGCs and read any net current. I would then measure the current on the EGCs and see if it accounts for the missing neutral current. Then I would look for the point of connection, N/G.

But I have one question to put out to Brian John or others who deal with these situations: if induction is the cause of current on an EGC, will this be at the expense of the conductor doing the inducing? It would seem to be necessary, electrically. So then would this change the amount of the balancing neutral current?

Example:
300A phase A, 200A phase B, 200A phase C, 40A induced on the EGC, so what does the Neutral carry?

Karl

 

[don_resqcapt19]

Re: Continuous current on an equipment ground conductor.

Wilson,
The response to Charlie does not change my statement. If 20% of the load current is showing up on the EGC something is wrong. I don't belive that under normal conditions that this is possible.
Don

 

[roger willis]

Re: Continuous current on an equipment ground conductor.

Back in the 60's I saw this very thing happen. We had 350 amps being induced. The cause was from a ground loop connect. We cut the loop and that fixed the problem. We had a feeder going to, lets say panel A, and a feeder going to panel B. These were 225 amp feeders. These feeders went from the back of the store to the panels located in the front of the building, about 250 ft. The eng. drawings showed connecting a ground wire between the panels. This is the wire we cut to fix the problem by opening the circuit. Since that time I have been very careful about creating ground loops.

Roger

 

[charlie b]

Re: Continuous current on an equipment ground conductor.

Originally posted by don_resqcapt19:If 20% of the load current is showing up on the EGC something is wrong. I don't believe that under normal conditions that this is possible.

Don: I agree that something is wrong, but I do not see a violation of physics here. What is wrong is the way the cables were installed. More on that in my next posting.

Suppose, just for the sake of discussion, that you lay the EGC in a large cable tray, then lay all 10 Phase A conductors (600 MCM) on top of it, and finally lay the rest of the conductors close to each other but some distance away from Phase A. Probably not a good installation, but we do not know how this installation was really performed. The electromagnetic field generated by 10 conductors of the same phase all lying on top of a single EGC could easily produce a ?huge? current (the word ?huge? being a technical term that means that I don?t want to calculate a real number).

 

[charlie b]

Re: Continuous current on an equipment ground conductor.

Originally posted by wilson9254: All of the conductors are in a single metallic wire way. The feeder is made up of 10 sets: (3)600kcmil, (1)350kcmil. There is one equipment ground conductor, size 500kcmil.

You may have a serious problem on your hands, and it is not related to the current on the ground conductor. Presuming that your 10 sets of conductors are made of copper and have an insulation system rated for 90C, the ampacity is 10 times 475 times 45% (i.e., the derating factor for 30 current-carrying conductors). Your 4000 amp system is therefore being fed by 2138 amps worth of conductors.

Apparently I was mistaken about the facility load being about 500 amperes, it is actually 900-1000amperes measured.

That is some relief, although I suspect that you meant that this was a one-time measurement, and not a 30 day load reading. At least your measured load does not (yet) exceed the ampacity of the feeder. However, I revert to a previous question: Do you have a value for calculated load? If the calculated load on this switchgear exceeds 2138, you have a code violation.

[ November 24, 2003, 02:06 PM: Message edited by: charlie b ]

 

[charlie b]

Re: Continuous current on an equipment ground conductor.

Originally posted by karl riley:Example: 300A phase A, 200A phase B, 200A phase C, 40A induced on the EGC, so what does the Neutral carry?

100 amps. There is no complete path for current to flow through the EGC and through a phase or grounded conductor. Therefore, the usual formula still applies:
(In*2 = Ia*2 + Ib*2 + Ic*2 - IaIb - IaIc - IbIc).

. . . if induction is the cause of current on an EGC, will this be at the expense of the conductor doing the inducing?

Yes, in the same way that a transformer?s primary current will vary with the load being supplied by the secondary. To complete the story, start with your example of 300, 200, 200, 40 amp currents. Now cut the EGC (from a safe distance, of course), so that the 40 amp current disappears. The new question is ?What happens to the currents in the three phases and in the neutral?? The answer is that they will decrease, and we would need detailed information about the installation to predict how much. But whatever happens to the four currents, they will still obey the formula that I cite above.

[ November 24, 2003, 02:25 PM: Message edited by: charlie b ]

 

[karl riley]

Re: Continuous current on an equipment ground conductor.

Thanks, Charlie, I follow that. And I assume that the EGC will only pick up induced current if there is a loop configuration, such as connection to building steel via a subpanel, with the loop being EGC from main panel to subpanel, with bonding to building steel taking the loop back to the main.

A second question: if a phase conductor lays exactly parallel to an EGC, with, say 1/8th inch separation, and the other phases are kept well away (not realistic - this is a theoretical limit question), if the phase carries 100A, how many amps could be induced? Assuming the EGC is in a loop configuration.

Karl

 

[pierre]

Re: Continuous current on an equipment ground conductor.

The current that the EGC is/was carrying is considerable. What does that do to the rest of the grounding system? When removing the equipment ground and allowing the wireway as the EGC, how does one go about disconnecting the 500kcmil EGC without getting hurt?

Pierre

 

[charlie b]

Re: Continuous current on an equipment ground conductor.

Originally posted by karl riley: A second question: if a phase conductor lays exactly parallel to an EGC, . . . and . . . if the phase carries 100A, how many amps could be induced?

If you have only one conductor per phase, the maximum would be 100 amps, and the actual current would be less. It would be a transformer with one turn in the primary and one turn in the secondary. If the installation being discussed in this thread really had 10 phase A conductors next to the EGC, with the other phases far away, you would have a transformer with 10 turns in the primary and 1 turn in the secondary. That would be a 10-1 step down in voltage and a 10-1 step up in current, so that the maximum current would be 1000 amps.

[ November 25, 2003, 11:22 AM: Message edited by: charlie b ]

 

[earlydean]

Re: Continuous current on an equipment ground conductor.

Would you have 10 turns? If this worked, then transformenrs would not need to be wound, but just parallel a bunch of conductors. I don't believe it works like that.

 

[charlie b]

Re: Continuous current on an equipment ground conductor.

It does work that way. The magnetic field created in the primary is based on the current in each wire. The current induced in the secondary is based on the strength of the magnetic field.

Put current through a long, straight wire, and you will create a magnetic field. Put another wire close by, and the magnetic field will create a voltage in the second wire. Now put the same current in each of two side-by-side wires, and you will get double the magnetic field. Put a third wire close by the pair of current-carrying wires, and you will get double the voltage.

But a long, straight wire does not make for a very good (i.e., efficient) primary side of a transformer. That is because of the shape of the magnetic field. The field around a straight wire looks like a series of cylinders that surround the wire. The cylinders closer to the wire have the stronger field, and the cylinders farther from the wire have weaker fields. But if you wrap a single wire in a loop, then the magnetic field is concentrated towards the center of the loop. Wrap it in a pair of loops, and you double the field. Wrap the wires in many loops around a bar of iron, and the field gets much bigger (iron is a better ?conductor? of magnetic fields than air). Finally, wrap 20 turns of wire around one side of a ring of iron, then wrap 10 turns around the other side of the same ring, and you get a 2:1 step-down transformer. The fact that the two coils share the same core (i.e., the iron ring) means that the magnetic field created by the primary windings will be carried (with very small loses) to the center of the secondary windings, thereby inducing a voltage in the secondary.

The short answer is that you can create a transformer by laying the primary and secondary wires side by side in long, straight lines. But you might need a couple of miles of wire to get enough secondary power to light up your desk lamp.

 

[karl riley]

Re: Continuous current on an equipment ground conductor.

Let's carry this thinking back to the real life induction situation. To avoid inducing large amps in the EGC, the three phases, neutral, and EGC need to be bundled together. The phase fields cancel each other, with the neutral taking up the vector imbalence, giving the EGC very little to pick up as induced current. A twist in the cable, such as in service drops, also prevents induction. Right?

Also makes it clear why when paralleling phases it is important to run all three together plus neutral and EGC in each raceway, rather than segregating the phases in different raceways.

I'm thinking out loud.

Back to one phase and an EGC running together in a straight line, the amps in the EGC would be dependent (critically) on the space between the two conductors. I don't know that 100A would induce 100A with even 1/8 inch separation, since you don't have the iron core to carry the field across. Also, it would depend on the impedance of the ground loop, since current will not flow unless the ground circuit is a circuit.

This is all IMHO, so any contrary thoughts are welcome.

Karl

 

[don_resqcapt19]

Re: Continuous current on an equipment ground conductor.

Charlie,

But you might need a couple of miles of wire to get enough secondary power to light up your desk lamp.

Then how are we getting 140 to 200 amps on the EGC in this installation? Or are we getting amps without voltage?
Don