Mezani
Member
- Location
- Washington
What code prohibits the use of the primary feeder equipment grounding conductor from being used as the grounding electrode for a transformer?
grounding electrode/equipment grounding conductor for a separately derived system
- Thread starter Mezani
- Start date
- Status
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
But what it doesn't say is that you can't use a grounding electrode conductor for equipment grounding purposes, and I am fairly certain there are other places that specifically allow you to acquire an equipment grounding conductor from a GEC.
I used to look at it that way before 2011... but now you have to consider that an EGC must be run with circuit conductors and a GEC no longer qualifies. If you don't want to run a wire-type EGC, the only option I see now is to run metallic raceway that qualifies as an EGC.
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
So if I have an isolated well pump in the middle of a field that operates @ 480 volt, and no other electrical equipment, but decided I wanted a 120 volt receptacle for convenience purposes and installed a small transformer for it, I either need to
A- run a GEC from the transformer secondary to the existing grounding electrode system for 480 volt system
B- if I would install another grounding electrode I would still need bonding jumpers to the GE system of the 480 volt system
C- run a GEC with the transformer supply conductors and use a metallic wiring method that itself qualifies as an equipment grounding conductor
D- if the wiring method supplying the transformer does not qualify as an equipment grounding conductor and I want to run the GEC with the supply conductors then I have to install both a EGC and a GEC in the raceway even though they may be the same size and may attach to effectively the same point on both ends.
That's pretty much the way I see it, to be compliant. FWIW, I think it's outright idiotic having to do so and am open to other ideas. I haven't looked to see if there is any change in 2014 NEC, but if not, feel free to make a proposal to rescind or modify 250.121 in the 2017 edition...
petersonra
Senior Member
- Location
- Northern illinois
- Occupation
- engineer
I run into this issue with control xfmrs.
You have to bond the transformer case and provide a GEC.
The transformer company often provides a convenient stud for making the connections. There are usually three wires there. The EGC, the N, and the GEC. In a control cabinet the EGC and the GEC go to lugs on the panel, often times the same lug. Why one needs two green wires going from a stud in the transformer to the same lug on the panel escapes me.
You have to bond the transformer case and provide a GEC.
The transformer company often provides a convenient stud for making the connections. There are usually three wires there. The EGC, the N, and the GEC. In a control cabinet the EGC and the GEC go to lugs on the panel, often times the same lug. Why one needs two green wires going from a stud in the transformer to the same lug on the panel escapes me.
- Location
- Illinois
- Occupation
- retired electrician
The 2014 code really puts it back to the way it was before the 2011 code. You could do it if all of the requirements for both the GEC and EGC were met. Not really as easy as it sounds, but it is possible.
Many will say that a GEC will always have some current and therefore the EGC would too, creating a violation of 250.6(A).
petersonra
Senior Member
- Location
- Northern illinois
- Occupation
- engineer
The objectionable current rule?
Who defines what level of current is objectionable?
In any case it is all but impossible to make an GEC in many situations that is not effectively in parallel with at least some EGCs.
Think about a typical building that has conduit supported by the steel structure of the building. The steel structure of the building is effectively bonded to the conduit that is an EGC.
- Location
- Illinois
- Occupation
- retired electrician
It appears that CMP 5's opinion is that any current on the EGC, other than fault current is objectionable current.
- Location
- Placerville, CA, USA
- Occupation
- Retired PV System Designer
Although Mike Holt teaches that the ideal is that there should only be one point of connection between the EGC "network" and the GES "network", there is nothing in the NEC that requires that. If there is current in an EGC whose source is not part of the building wiring system (e.g. earth potential gradients) I do not necessarily see that as objectionable current from the point of view of the NEC. If it does constitute objectionable current, then the only safe and practical solution (other than eliminating the source) is to never share a conductor as EGC and GEC and never to bond any EGC to ground anywhere except at the main bonding point. Given that a lot of electrical equipment can have an incidental earth connection via gas or water pipe or mounting on concrete, the latter may not even be possible.
Sent from my XT1080 using Tapatalk
Sent from my XT1080 using Tapatalk
Given that a shared GEC-EGC will not oft be employed in a residential system, with the typical witring methods utilized in commercial, institution, and industrial environs, it is next to impossible to isolate EGC current, of any kind, from bonded non-current carrying metal parts or through concrete (especially damp and metal reinforced)... and this current actually returns to the source through the GEC... and vice versa if the connected system is not the source. This is likely the main reason why the NEC does not define objectionable current... and those that say there should never be any current on an EGC, other than fault current, do not truly understand the finer points a grounded-conductor system.
Just want to add, if there is any objectionable current between terminations, it will be carried by both the EGC and GEC when separate. Under 250.6(B), you'd be permitted to discontinue one or more grounding connections, interrupt the continuity of a conductive path, etc. So technically, one conductor serving both GEC and EGC functions would be the best remedial action anyway.
I still think 250.121 is outright idiotic no matter how much they modify it. :slaphead::happyyes::happyno:hmy:
I still think 250.121 is outright idiotic no matter how much they modify it. :slaphead::happyyes::happyno:
hmy:
petersonra
Senior Member
- Location
- Northern illinois
- Occupation
- engineer
Just want to add, if there is any objectionable current between terminations, it will be carried by both the EGC and GEC when separate. Under 250.6(B), you'd be permitted to discontinue one or more grounding connections, interrupt the continuity of a conductive path, etc. So technically, one conductor serving both GEC and EGC functions would be the best remedial action anyway.
I still think 250.121 is outright idiotic no matter how much they modify it. :slaphead::happyyes::happyno:
I think what they may have meant to say is that you have to have a a wire type GEC that is separate from whatever EGC you might have.
The GEC runs from point A to point B. Point A, the GEC is attached to the SDS at the same point where the system bonding jumper is attached (250.30 A 3 or A 4). Point B is the "nearest" "accessible" grounding electrode or a busbar with a jumper to the grounding electrode. The GEC runs unspliced (or spliced with irreversible connectors) and is bonded to the ends of ferrous raceways and at points where the ferrous raceway is discontinuous. The purpose is to provide a dedicated and clean ground, referenced to the earth (where the reference is free of objectionable current on the grounding paths).
The EGC runs from point A, the metal frame of the load, to point B, the source breaker grounding busbar. The purpose is to facilitate breaker tripping in ground faults by providing a low impedance path back to the source, to increase the ground fault current seen by the breaker so it trips faster.
The most common violation I see is the omission of the GEC. You can survey any building, open the transformers looking for the GEC and it's not there.
In the lower probability case of a transformer fed directly from service equipment, points A and B for both the GEC and the EGC can be the same. In the common case of transformers fed from distribution sub panels in a large facility, points A and B for the GEC and the ECG are not the same and not close to being the same.
In the common case where 250.6 applies, the EGC must be run to the source breaker grounding busbar regardless of how noisy or how many amps are found flowing on the other nearby EGC's and their conduits. The GEC however can be run, and imo is required to be run, to a grounding electrode that is "nearby", "accessible", "in contact with the earth", and imo free of objectionable current found elsewhere in the building, especially away from dual fed switchgear and large UPS, VFD's that may have their filter cap sections internally Y connected and grounded (introducing objectionable current that may flow on the grounding paths).
The EGC runs from point A, the metal frame of the load, to point B, the source breaker grounding busbar. The purpose is to facilitate breaker tripping in ground faults by providing a low impedance path back to the source, to increase the ground fault current seen by the breaker so it trips faster.
The most common violation I see is the omission of the GEC. You can survey any building, open the transformers looking for the GEC and it's not there.
In the lower probability case of a transformer fed directly from service equipment, points A and B for both the GEC and the EGC can be the same. In the common case of transformers fed from distribution sub panels in a large facility, points A and B for the GEC and the ECG are not the same and not close to being the same.
In the common case where 250.6 applies, the EGC must be run to the source breaker grounding busbar regardless of how noisy or how many amps are found flowing on the other nearby EGC's and their conduits. The GEC however can be run, and imo is required to be run, to a grounding electrode that is "nearby", "accessible", "in contact with the earth", and imo free of objectionable current found elsewhere in the building, especially away from dual fed switchgear and large UPS, VFD's that may have their filter cap sections internally Y connected and grounded (introducing objectionable current that may flow on the grounding paths).
Last edited:
- Location
- Placerville, CA, USA
- Occupation
- Retired PV System Designer
Which is of dubious value given that all electrodes other than auxiliary electrodes must be bonded together to form a single GES.
I would never advocate or specify driving a ground rod for the electrode of a SDS. I would call that hokey or dubious (but it's a common spec).
Typically the grounding electrode is the water main entrance point or in a steel framed building, any of the existing I beams that are in contact with the earth, typically sitting on concrete piers and bolted to the concrete with a substantial quantity of heavy threaded rod and rebar, all encased in the concrete pier pour. The I beams would be assumed to be all bonded together. The steel frame is not to be used as a grounding conductor, but it is bonded together, so you may pick one of many. Don't have the code reference for this and I could be wrong, but field bonding of new electrodes has a low probability of occurring compliantly (no new ground rods).
This is where the NEC offers a remedy for noisy source grounding and neutrals. The SDS gives you the opportunity to create a clean neutral, referenced to the earth at a point where earth contact is substantial and free of noise present elsewhere in the facility. You may choose, and imo are required to choose, an electrode that is a clean reference to the earth.
Last edited:
And that's not a problem where they run essentially different routes... but what about when they run the exact same route and are essentially connected at both ends. I would call that paralleled typically, but that doesn't even meet the requirements of parallel conductors, as the GEC will likely be larger than a wire-type EGC. If a line to ground fault where to occur, more of the fault current would be handled by the GEC in this case, which is exactly the opposite of either's purpose.
I think what they should have said...
OR
Last edited:
kwired
Electron manager
- Location
- NE Nebraska
- Occupation
- EC
A GEC for a service will carry current because the supply side is also grounded and there is a parallel path through earth, water pipes, other metal objects. But a properly installed separately derived system only has one point with ground references attached to it and is much more difficult to end up with parallel potential current paths by design. If you had an electrode at the transformer and at another point further downstream then you would have a situation more like what you have with services and much more likely to have current flow on the GEC.
If any current flows on the GEC of a properly installed SDS it is likely current coming from the supplying service and not the secondary of the SDS.
- Status