Why a grounding electrode at sub-fed structures?

Status
Not open for further replies.

AJElectric

Member
Location
Iowa
We do a lot of (mostly) agricultural wiring so we commonly encounter a scenario where there is a POCO pole or pad-mount transformer with 200A 120/240V 1ph meter loop or CT metering on premises (farm yard with residence and outbuildings) and from there splitting off to feed separate buildings.

Back in the day, the standard procedure was GE bonded to N at the service disconnect for premises, and then OH or USE wire to each structure without including a ECG or GEC. At "service entrance" for each building is a GE bonded to N and appropriate main w/ OCP. Most of these drops would range from 30 to 100A.

20 years later, as many of these yard's services are being upgraded to accommodate new buildings, heavier loads, adding standby generator, etc. typically a new proper service disconnect with OCP and/or MDP is installed near POCO trans with GE bonded to N and comprises the "main service" for the premises - everything downstream being treated as a feeder and 4-wire service (L,L,N,ECG) routed to new structures.

Regarding _existing_ feeders off this new main service the question is always "What to do about the lack of ECG from bonding point at main service???" Common practice has been to leave it as-is and leave the N bonded at each structure (remember no ECG exists back to POCO trans) with existing GE in place, with the assumption that the GE "sort of" takes the place of ECG. (In other words, no one wanted the expense of adding a ECG to every building which has been there 50 years and "didn't need one" before).

It should go without saying that the GE at each building does not take the place of the EGC, and a loose main neutral would present a shock hazard in spite of the GE bonded to N (and it HAD been that way for 50 years). So for me this begs the question - on a new installation where a 4-wire feeder is run to the building, why _is_ a GE required there?
 

ActionDave

Chief Moderator
Staff member
Location
Durango, CO, 10 h 20 min from the winged horses.
Occupation
Licensed Electrician
......Regarding _existing_ feeders off this new main service the question is always "What to do about the lack of ECG from bonding point at main service???" Common practice has been to leave it as-is and leave the N bonded at each structure (remember no ECG exists back to POCO trans) with existing GE in place, with the assumption that the GE "sort of" takes the place of ECG. (In other words, no one wanted the expense of adding a ECG to every building which has been there 50 years and "didn't need one" before).
That's what you do as long as there are no metallic paths back to the main panel.

It should go without saying that the GE at each building does not take the place of the EGC, and a loose main neutral would present a shock hazard in spite of the GE bonded to N (and it HAD been that way for 50 years). So for me this begs the question - on a new installation where a 4-wire feeder is run to the building, why _is_ a GE required there?
They say it has to do with power surges and lightning strikes. I think there are just a lot of people that have an unnatural affection for ground rods.
 

Smart $

Esteemed Member
Location
Ohio
... So for me this begs the question - on a new installation where a 4-wire feeder is run to the building, why _is_ a GE required there?
Try looking at it another way. The GES has always been required. The only difference now is it is not bonded to the grounded conductor, so the EGC is required... which is the real difference.

It has been rationalized, perhaps proven to a degree, that a system's grounded conductor should be bonded to the earthing electrode system at only one point. Bonding at more than one point can result in voltage gradients appearing on the grounded conductor during line surges and/or nearby lightning strikes. Keeping the grounded conductor and electrode system isolated except for that one bond minimizes that undesirable condition.
 

kwired

Electron manager
Location
NE Nebraska
Try looking at it another way. The GES has always been required. The only difference now is it is not bonded to the grounded conductor, so the EGC is required... which is the real difference.

It has been rationalized, perhaps proven to a degree, that a system's grounded conductor should be bonded to the earthing electrode system at only one point. Bonding at more than one point can result in voltage gradients appearing on the grounded conductor during line surges and/or nearby lightning strikes. Keeping the grounded conductor and electrode system isolated except for that one bond minimizes that undesirable condition.
Yet the POCO makes a connection to earth at nearly every structure in their system on the grounded conductor.

I see where the idea of separate EGC to separate structures comes from and think it is probably the best idea, but at same time we never ran a EGC to separate buildings for many years and I am not aware of too many issues with that. Most of the issues with stray voltages at places like dairy farms were often traced to something failing more so then simply because there is no EGC to a separate building. And that failing item may be something on POCO system a couple miles away and is still a problem even if the dairy barn is supplied with an EGC.
 

AJElectric

Member
Location
Iowa
That's what you do as long as there are no metallic paths back to the main panel.

This is a very good point - I don't know of any yards that still have UG metallic water piping around here, but it was common 50 years ago and is definitely something to be considered. That's actually quite a can of worms, because if one did find metal piping it's likely the premises plumbing system is a hybrid of plastic & metal. There is probably no one-size-fits-all rule that could be applied here. UG telephone lines from building to building also a point of contention.

Try looking at it another way. The GES has always been required. The only difference now is it is not bonded to the grounded conductor, so the EGC is required... which is the real difference.

Understood. Adding a EGC to each structure and then un-bonding the N at each structure should improve safety, purportedly because without it we're relying on _only_ the neutral to carry ground fault current and being a CCC the neutral is at risk of developing loose connections over time which would energize everything bonded to it (ground fault or not).

On the flipside, a loose neutral is likely to be noticed (120V loads not working or damaged) and soon corrected - whereas with "updated" wiring a loose EGC (while less likely) would not be realized until a ground fault energizes everything creating a major shock hazard. When that happens, would the GES at that structure bring anything to the table (good or bad)? My first thought is that it would lower touch-potential of grounded surfaces (although not to a safe level) but increase step potential in the vicinity (possibly more dangerous). And as a separate issue, if tied to metal water piping that goes UG but is not contiguous back to main service, couldn't it actually create even more potential hazards?

To piggy-back on that, while not always practical would GFCI protection at the feeder entrance for each structure be an end-all solution in absence of EGC?

It has been rationalized, perhaps proven to a degree, that a system's grounded conductor should be bonded to the earthing electrode system at only one point. Bonding at more than one point can result in voltage gradients appearing on the grounded conductor during line surges and/or nearby lightning strikes. Keeping the grounded conductor and electrode system isolated except for that one bond minimizes that undesirable condition.

Exactly, and this makes me wonder - when N is bonded at only one point on the yard and 4-wire feeder correctly run to each structure, does the GES at every structure create more problems than it solves? Case-in-point, I have a customer with pad-mount POCO trans and site disconnect, N bonded to GES there, splits off to existing UG to residence and other buildings. Added new barn fed 4-wire UG about 200ft and ufer GES at barn as per code. Telephone line from utility to residence, then runs UG to barn about 200ft but separate path. The barn phone system runs on 12VDC, but it has a PE terminal on the circuit board with specific instructions to bond this to a dedicated GE (we bonded it to the ufer instead). Upon first lightning storm, the smoke escaped from the phone system. I suspect that if instead there was no GES at the building (there is a EGC back to POCO trans) this would not have happened. Of course, anything can happen in a lightning storm but why exacerbate it by having a GES at two points on the yard?

I'm not trying to argue a point here, just looking for input. To me, the jury is out on whether a GES at every building is a liability or an asset.
 

AJElectric

Member
Location
Iowa
Yet the POCO makes a connection to earth at nearly every structure in their system on the grounded conductor.

Perhaps the main difference between the POCO system and premises wiring is that they're covering many miles, and would rather see a lightning surge go to earth in the immediate vicinity than be carried back on several miles of conductor then to earth. Additionally with OH conductors in a way they're using the grounded conductor _for_ lightning protection, whereas this normally isn't applicable with premises wiring.

I see where the idea of separate EGC to separate structures comes from and think it is probably the best idea, but at same time we never ran a EGC to separate buildings for many years and I am not aware of too many issues with that. Most of the issues with stray voltages at places like dairy farms were often traced to something failing more so then simply because there is no EGC to a separate building. And that failing item may be something on POCO system a couple miles away and is still a problem even if the dairy barn is supplied with an EGC.

I agree completely - stray voltage is really a separate issue from the EGC (or lack thereof) with GES (or lack thereof) for each building. On dairy facilities we typically install a neutral isolator and then EGC to every building and only one GES on the premises bonded to N at the POCO trans. At the same time we avoid any neutral load at all, especially over long distances across the farm, preferring a step-down trans near point-of-use where 120V is needed.
 

kwired

Electron manager
Location
NE Nebraska
Problem with multiple grounded neutral system used by POCO is any voltage drop on their distribution (primary or secondary) is imposed on the grounded conductor running from their service transformer to the user's service equipment. This can leave you with a voltage to ground on any item bonded to the electrical grounding system even if you ran proper equipment grounding conductors to secondary buildings or structures - and is often a cause of shock/electrocution at places like boat docks/marinas. This is also a problem for the mentioned dairy farms (problems are observed quicker with dairy cattle because it usually effects milk production pretty quickly compared to how quickly health issues are noticed in other animals), because of the voltage that may be present on the premises grounding system to earth. Separately derived systems doesn't solve the problem as the grounded conductor of those is bonded to same primary side grounded conductor that has the stray voltage on it, neutral blocking equipment as mentioned is a possible solution - I'm not even certain exactly how they work as I have never had my hands on one.

Advantage of multiple grounded neutral system used by POCO is you have elecrodes all over the system and this results in a low impedance to ground. Create a voltage system that is isolated from the main grid of this country and you only have what electrodes you provide at that system to create an earth reference. Ground rods are lucky to get 25 ohms which is a relatively high resistance - but put thousands of them parallel to one another and the grid has a fairly low resistance to ground though there is added impedance between them depending on current flow conditions.

Yes grounding at nearly every structure of the system helps POCO with lightning mitigation by keeping it in a somewhat localized area as well.
 

AJElectric

Member
Location
Iowa
The neutral blocker is effectively an open contact that immediately closes when a particular voltage (usually 11V) is measured across it.

When implemented, the bond between primary and secondary neutrals at the POCO trans is removed (goodbye stray voltage) and the blocker is put between them. POCO still has a GE bonded to primary N and shell, and then at least 6ft away we have a GES bonded directly to transformer secondary N (not at main disc) using insulated GEC.

The next stray voltage culprit to check off the list is a weak or over-loaded neutral anywhere on-premises. One way to mitigate this is to not use a neutral at all over long distances. Maybe it's overkill. Establishing SDS's all over the farm should not be panacea, but it should be considered any time a new structure is erected. Many times it's easy to get the N load down to only convenience recepts and small things, only a small trans is needed.
 

iwire

Moderator
Staff member
Location
Massachusetts
It should go without saying that the GE at each building does not take the place of the EGC, and a loose main neutral would present a shock hazard in spite of the GE bonded to N (and it HAD been that way for 50 years). So for me this begs the question - on a new installation where a 4-wire feeder is run to the building, why _is_ a GE required there?

A grounding electrode system is required at a separate building or structure for the same reason a ground electrode system is required at the service and it has nothing to do with the presence or not of an EGC.

250.4(A) Grounded Systems.
(1) Electrical System Grounding. Electrical systems that
are grounded shall be connected to earth in a manner that
will limit the voltage imposed by lightning, line surges, or
unintentional contact with higher-voltage lines and that will
stabilize the voltage to earth during normal operation.
 

AJElectric

Member
Location
Iowa
There may be some misunderstanding about my dilemma. I do understand NEC requires GES at every structure, but my question is what it actually does when there is already a GES at the site disconnect.
old grounding methods.jpg
new grounding methods.jpg
neutral blocker.jpg
Minimize N-E-V.jpg
 

iwire

Moderator
Staff member
Location
Massachusetts
There may be some misunderstanding about my dilemma. I do understand NEC requires GES at every structure, but my question is what it actually does when there is already a GES at the site disconnect.

I don't think there is any misunderstanding. :)


Consider this, in the vast majority of cases the neutral supplying the 'site disconnect' has already been grounded many times by the power company so why does the NEC require grounding it again at the site disconnect?

I think we are all stuck with the same answer 'because the NEC says so' I don't think you are going to get a clear technical answer.
 

ActionDave

Chief Moderator
Staff member
Location
Durango, CO, 10 h 20 min from the winged horses.
Occupation
Licensed Electrician
There may be some misunderstanding about my dilemma. I do understand NEC requires GES at every structure, but my question is what it actually does when there is already a GES at the site disconnect.
It helps with this....
there are just a lot of people that have an unnatural affection for ground rods.
 

AJElectric

Member
Location
Iowa
Consider this, in the vast majority of cases the neutral supplying the 'site disconnect' has already been grounded many times by the power company so why does the NEC require grounding it again at the site disconnect?
A practical reason from the NEC's perspective is not having jurisdiction over POCO side of things - for example "vast majority of cases" is correct but not very reassuring, so NEC covers it to ensure a grounded system at each premises.

A practical reason on from my own perspective, if the site is running on Generator while the POCO has to R&R a transformer I want to ensure a grounded system.

I think we are all stuck with the same answer 'because the NEC says so' I don't think you are going to get a clear technical answer.
You are probably right :cool:
 
Am I wrong here?

A lightning strike is the ultimate in step potential. With a median 30,000 amps going into something with substantial resistance, there are large and dangerous potential differences from one point on the ground to another.

Then, isn't that potential going to get equalized through the grounding conductors that form a circuit between distant ground rods?

Then, could that create a dangerous voltage drop across the EGCs? There's more to consider than just the resistance, since lightning has significant power even well into the RF range and total impedance is relevant.

Are the neutrals safe because they're not bonded at the subpanels?

Could the EGCs arc over to the current carrying conductors?

The emphatic advice I've always seen in the radio world is that, though it's safe and prudent to have a multi-point ground field, all of the equipment grounds without a single exception must land at the same one of those multiple points.
 

kwired

Electron manager
Location
NE Nebraska
Could the EGCs arc over to the current carrying conductors?

After what I saw a couple weeks ago - I'd say yes.

Not sure exactly where the lightining strike was but it came into a splice box on a grain bin where from there it supplied tap conductors to three disconnects. Splice was done with a three pole "terminal block". That block was nearly completely melted - all three poles. Underground conductors were still fine except near point of attachment to the block, tap conductors were still fine except near the point of attachment at that block, supplied loads were all fine, just a melt down at that block, and likely arced to the grounded box it was mounted in at that point as well. Other end of underground was from a disconnect on a pole - nothing was damaged at the pole, which included a service disconnecting means, three phase meter/socket, overhead transformer bank, and primary fusing/other equipment to connect to the primary.
 

AJElectric

Member
Location
Iowa
Then, isn't that potential going to get equalized through the grounding conductors that form a circuit between distant ground rods?
That is exactly what happens, and IMO occasionally is the cause for lightning damage to equipment (while some believe ground rods protect things????)

Then, could that create a dangerous voltage drop across the EGCs?
Definitely a valid concern - then consider potential for arcing and flashover, especially if EGC is routed through/over/around building to connect to "auxiliary" GR for other equipment (telephone service, standby generator).

After what I saw a couple weeks ago - I'd say yes.
I have seen similar, and even had a case where starters on bin fans pulled in due to lightning storm (some welded closed, others not). Owners arrived next morning to find the fans running.

Lightning wrecks stuff, more ground rods doesn't help.
 

Smart $

Esteemed Member
Location
Ohio
...A lightning strike is the ultimate in step potential. With a median 30,000 amps going into something with substantial resistance, there are large and dangerous potential differences from one point on the ground to another.

Then, isn't that potential going to get equalized through the grounding conductors that form a circuit between distant ground rods?....

That is exactly what happens, and IMO occasionally is the cause for lightning damage to equipment (while some believe ground rods protect things????)


Definitely a valid concern - then consider potential for arcing and flashover, especially if EGC is routed through/over/around building to connect to "auxiliary" GR for other equipment (telephone service, standby generator)...
Don't forget that the grounded conductor is bonded to the GES. That in itself helps "stabilize" the system voltage, which is the point of bonding the GES. The downside is, yes, there can be voltage gradients across the system... but ask yourself, what is the alternative? Do you think fewer electrodes will change the voltage gradients imposed on conductors from a nearby lightning strike?
 

AJElectric

Member
Location
Iowa
I don't mean to sound facetious, but one alternative would be to not purposely create a low-impedance path across the voltage gradient by way of GES in two places. An actual lightning protection system (completely separate from the EGC/GES) would be another alternative (outside the scope of NEC though).

Do you think fewer electrodes will change the voltage gradients imposed on conductors from a nearby lightning strike?

Yes, in most scenarios.

At my home there is a GES at the service/meter on the house then from main panel a 30A subfeed running 30ft through the house then 50 ft UG to a small shed, wood construction on slab. NEC requires GES at this shed and I do not see how this second GES makes the system safer, however I do believe that if lightning strikes near the shed then the GES -> EGC thru house -> GES would give the lightning a lower-impedance path THROUGH THE HOUSE to earth to a lower point on the gradient. Something about that makes me uncomfortable.

Instead of a shed, if this were a steel structure like a pre-fab metal lawn shed anchored to ground/slab, or windmill anchored into the ground then in the event of lightning the GES at the structure would be mostly irrelevant.

On the other hand, one could argue that if the wire went OH to a wood pole near the shed and lightning strikes the pole (certain destruction) then the GES at the shed would direct some energy to earth there instead of through the GES at the house (but enough to make a difference??? really???)

I think the bottom line is that lightning is mostly unpredictable, and that's no reason to ignore it but one should not expect a ground rod to offer lightning protection.

MH has some great insight if you watch his video.
https://www.youtube.com/watch?v=YuDqXFvRv94

So what I ask myself :D how is the GES at the shed 50 ft from my house any different from a typical auxiliary GE adjacent to the house? Either way the liability as described in MH's video is the same IMHO.
 
Status
Not open for further replies.
Top