Grounding Site lights

[George Stolz]

Jtester, is it your recommendation that the neutral be reterminated to the poles...

Yes, or get an EGC to them.

...and left alone?

No. You need to find the location of the open neutral.

Since these poles are on a multiwire branch circuit, then they would continue to operate without a neutral. If there were an equal number of poles per circuit, the voltage measured from line to neutral would even look right; the voltage from line-to-neutral in an open (source) neutral situation is determined by the resistance of the loads. If there is even resistance, there would be a roughly even voltage on phase A and phase B. So the open neutral would be less evident.

It is statistically improbable that the resistance from line to neutral would be identical on both phases, so there would be some unbalanced current trying to find it's way back to the earthed transformer. Therefore, people were receiving a shock when they touched the poles. They were filling in for the missing neutral.

Leaving the poles unbonded (without a path for fault current) is a dangerous fix. For the next five minutes, it'll work ok. The symptoms went away. But right now, you likely have ballasts getting grumpy from an imperfect supply, which could cause one to burn up and fault to the casing.

• De-energize the poles.
• Tie an ungrounded supply conductor (at the source of supply) to the neutral.
• Walk to each pole, from first to last, seperate all the conductors, and measure for continuity between the two.
• Tie them back together as you go. Leave the loads disconnected. When you come across a pole with no continuity between the phase and neutral, you've found the problem.

One problem with this is, you've already touched the connections. If there was a bad connection, you might have noticed it (or not) and (perhaps inadvertently) fixed it when you were unbonding the poles.

One other potential problem is that there may be some continuity through the loads, which could mess with you. But I don't think that will be a problem.

My 2?.

 

[fc]

It would be nice if 250.32(B)(2) was eliminated from the NEC. I think it tends to be unsafe way to install.

 

[George Stolz]

Rumor has it that a proposal has passed to delete it from the 2008 NEC.

 

[fc]

:lol: Wow my wish may come true!

Now that said I am going to wish for million dollars instead of my good looks! :wink:

 

[peter d]

IMO It does sound like an open neutral somewhere in the circuits as the grounded conduct should not be at that much difference of potential from earth assuming a fairly balanced load.

I agree with Bob. Think of a range or dryer that used the old method of bonding the frame to the grounded conductor. You don't get a shock every time you use these appliances, do you? I don't. But you will if the grounded conductor opens up.

 

[jtester]

Jtester, is it your recommendation that the neutral be reterminated to the poles and left alone?

Sorry I didn't get to reply sooner. I agree with Bob. Disconnecting the neutral made a bad situation worse. My professional recommendation would be to determine if the NEC or NESC is applicable and if the NEC is disconnect the lighting until a functioning egc is established. If the NESC is applicable then the system was legal and functioning properly, and it should be returned to its original state.

While the neutral bond may be a code violation, at least it functioned as a return path for ground faults. Now there is no adequate grounding and a line-ground fault will energize a pole at 277 volts to ground.

Jim T

 

[cripple]

It seems that the NEC does consider a light pole structure, and requires all building or structures to have a disconnecting means and grounding electrode. But 225.32 exception 4 does permit the disconnect means to be located elsewhere on the premises. 250.32(D) addresses the grounding of the disconnecting means that are located remote from those buildings or structures. It is every clear that if the disconnecting means is located remote from the light pole an equipment grounding conductor is required, and it must be bonded to the electrode of that structure as required by 250.50. :roll:

 

[don_resqcapt19]

fc,

It would be nice if 250.32(B)(2) was eliminated from the NEC. I think it tends to be unsafe way to install.

Are you saying that all services are unsafe, or that the electrons change their behavior on the load side of the meter?
Don

 

[fc]

Don


Are you saying that all services are unsafe, or that the electrons change their behavior on the load side of the meter?


Answer: NO

I don't agree with allowing no EGC to a separate structure.
I have a problem with 250.32 (B)(2) (2) Metallic paths. May not be one when we install but what about later when maybe a water line or phone line or cable is installed from the other building? Do you think they will call us back to install a EGC when that takes place?
Just a bad design. I would never install without EGC.

 

[don_resqcapt19]

fc,

I don't agree with allowing no EGC to a separate structure.
I have a problem with 250.32 (B)(2) (2) Metallic paths. May not be one when we install but what about later when maybe a water line or phone line or cable is installed from the other building? Do you think they will call us back to install a EGC when that takes place?
Just a bad design. I would never install without EGC.

My point is that we don't have EGCs with the service and the NEC requires parallel paths at the service. Why is that ok and the same installation to a second building is not ok?
Don

 

[fc]

Don

My point is that we don't have EGCs with the service and the NEC requires parallel paths at the service. Why is that ok and the same installation to a second building is not ok?


NEC says it's OK. I see your point it's the same Yes it is but the NEC also says no metallic paths.

I said I don't like that design to a second building or structure. JMO
I like to eliminate the unless there is no Metallic paths to our customers. I would hate to tell our customers sorry you can't install phone line, water line, gas line, etc. unless I add a EGC. I am sure there reply would be why did you not install EGC in the first place.
I see it this way what is the chance that they will never have a metallic path to the other building or structure?
It's just a bad design not to have EGC.

 

[George Stolz]

fc, it's worth noting in this discussion that no metallic paths bonded to the remote building's electrical system are allowed.

While we know that a phone service to a building, and a metallic water pipe must be bonded per the NEC, that's not a guarantee that it will happen in the future. A homeowner adding a water line to a detached building might not bond it, unaware of the requirement.

All said, I agree with the reasoning in eliminating the remote structure's exception from the otherwise normal practices. It's simpler to understand that way, and the benefits of an EGC are greater than the costs, IMO.

 

[peter d]

My point is that we don't have EGCs with the service and the NEC requires parallel paths at the service. Why is that ok and the same installation to a second building is not ok?
Don

I think it's because of the high fault current available at a service.


This is only a guess and I have no evidence to support it, only my own logic.

 

[iwire]

250.32(D) addresses the grounding of the disconnecting means that are located remote from those buildings or structures. It is every clear that if the disconnecting means is located remote from the light pole an equipment grounding conductor is required, and it must be bonded to the electrode of that structure as required by 250.50. :roll:

I am not sure I follow you here.

250.32(D) only comes into play when we use Exception No. 1 or 2 of 225.32

In this case we would not be using either Exception No. 1 or 2 of 225.32 we would be using exception 3 of 225.32 and 250.32(D) does not apply.

 

[mpd]

iwire

225.32 exception 3, says the disconnecting means shall be permitted to be located elsewhere on the premises.

250.32 (B) (2) says the grounded circuit conductor shall be connected to the structure disconnecting means

if we do not have a disonnecting means at the pole can you use 250.32 (B) (2) and ground the pole with the grounded conductor

 

[jtester]

My point is that we don't have EGCs with the service and the NEC requires parallel paths at the service. Why is that ok and the same installation to a second building is not ok?

Without trying to seem argumentative, and without opening another can of worms, I would offer these thoughts.

What does the neutral-ground bond at a service disconnect perform?
1. It provides a good ground fault circuit connection from our unigrounded system to the utility multiple grounded system. Not applicable in our second structure example.
2. It lessens the effects, in our system, of a utility hv wire contacting the utility neutral near the service. This most often won't occur in the second structure scenario.
3. It mitigates the lightning effects riding in on the utility neutral. This also doesn't occur much in our second building example.

Since these benefits of a neutral as a ground wire to our second building are moot, what are the disadvantages of not running a separate ground.
1. Our ground wire/neutral becomes a current carrying wire which inherently isn't bad, but the NEC frowns on that.

I can't seem to come up with an argument for eliminating the egc in these cases based on the above thoughts.

Jim T

 

[iwire]

Money.

Eliminating a conductor saves money.

 

[ryan_618]

fc,
My point is that we don't have EGCs with the service and the NEC requires parallel paths at the service. Why is that ok and the same installation to a second building is not ok?
Don

I agree. And if this proposal passes, I know someone that will try to change the code again to at least allow services to be be bonded with the main (or system) bonding jumper at the transformer and an EGC to the service, which currently is prohibited. If this type of installation is OK for seperatley derived systems, why not services?

 

[iwire]

change the code again to at least allow services to be be bonded with the main (or system) bonding jumper at the transformer and an EGC to the service, which currently is prohibited.

Ryan you have me interested how would that be set up.

This EGC would have to be at least the same size as the service conductors I would imagine.

 

[ryan_618]

This EGC would have to be at least the same size as the service conductors I would imagine.

No, not really. I think a person could size it based on 250.102(C), which really is the same concept as and EGC on the supply side, isn't it? So I guess ultimatley Table 250.66.