What's the consensus

Status
Not open for further replies.

mopowr steve

Senior Member
Location
NW Ohio
Occupation
Electrical contractor
Have a set of plans that show a 400amp(320a continuous) meterbase w/ 2- 200amp main OCPD's serving 2 buildings as one occupancy.

Plan examiner approved plans allowing #4 cu to waterline for GEC

Inspector says that it needs to be a 1/0

I did have doubts myself but I'm not 100% sure that the #4 was in fact OK since either disconnect is only 200amp. I understand that it goes by the max equivalent size of undrounded service conductors but not sure that the intent is for the service conductors feeding the meterbase section as the factory service conductors feeding OCPD's are 3/0. I guess I would not concider them paralleled unless they landed on 1- 400 amp OCPD not in this case where they are 2- independent service disconnects. So aren't we really sizing this particular GEC to trip the 200 amp breaker if a fault would occur?
Whats the difference if some one say has a 4 gang meterbase fed with 500mcm then comes out the bottom, mounts there own 200amp disconnects, and runs #4 for GEC to those~?
 

charlie b

Moderator
Staff member
Location
Lockport, IL
Occupation
Retired Electrical Engineer
So aren't we really sizing this particular GEC to trip the 200 amp breaker if a fault would occur?
That is absolutely not the reason we install GECs, and that is not the job that they perform for us. They are not about assisting with tripping anything. There is much debate over what they really do for a living and why we install them in the first place, and I would rather not stir up that debate. So without mentioning lightning (oh shucks, I just mentioned it), let me say that having an essentially zero voltage between the neutral conductor and planet earth is a good thing.

Now back to your question. If the location of the first overcurrent device is the meter base, then the size of the feeder to that meter base is the basis for the GEC size. So I agree with the Inspector. But to make sure all is covered, have you established the neutral-to-ground bond at the meter base, and are you running an equipment grounding conductor (EGC) with each of the 200 amp feeders to the two buildings?

 

augie47

Moderator
Staff member
Location
Tennessee
Occupation
State Electrical Inspector (Retired)
I would imagine you will get different opinions and they may vary depending on where your connections are made. If the GEC connection originates at the meter or POCO service point I would think it would be based on the size of the service conductors at that point. That size is not given in the OP, but for a #4 to suffice the service conductors could not exceed 3/0 Cu or 250AL.
If your GEC originates at the disconnects I, for one, would ignore any factory installed conductors and base my decision on the NEC required conductor size. For common GEC with individual taps, I would base my decision on the sum of the CM of the NEC sized conductors to two disconnects. In this case 3/0 Cu so 167,800 X 2 or 335,600 kcmil requiring a #2 with a #4 tap to each individual disconnect.
 
Have a set of plans that show a 400amp(320a continuous) meterbase w/ 2- 200amp main OCPD's serving 2 buildings as one occupancy.

Plan examiner approved plans allowing #4 cu to waterline for GEC

Inspector says that it needs to be a 1/0

I did have doubts myself but I'm not 100% sure that the #4 was in fact OK since either disconnect is only 200amp. I understand that it goes by the max equivalent size of undrounded service conductors but not sure that the intent is for the service conductors feeding the meterbase section as the factory service conductors feeding OCPD's are 3/0. I guess I would not concider them paralleled unless they landed on 1- 400 amp OCPD not in this case where they are 2- independent service disconnects. So aren't we really sizing this particular GEC to trip the 200 amp breaker if a fault would occur?
Whats the difference if some one say has a 4 gang meterbase fed with 500mcm then comes out the bottom, mounts there own 200amp disconnects, and runs #4 for GEC to those~?

I am a little confused by this:

serving 2 buildings as one occupancy.

Does that just mean that one building is an accessory structure as you would have in 230.40 Ex #3? In that case, That is an interesting question. With a service with multiple service disconnect enclosures, you can of course use the GEC tap method, but Im not sure off the top of my head if there is specific permission to do this at separate structures where you would have the common GEC...? If you grounded it before the sets split, then you could have a "200 amp ground" at each structure and that would definitely be compliant, just not sure if you can skip the GEC at the pedestal.
 
Ok reread and it seems like your 2 service disconnects are located at the pedestal. So it seems pretty clear to me you need your service GEC there sized to the service conductors, and then you need a GES at each building (250.32) sized for the supply conductor serving that building.

Regarding the situation I first interpreted it as, that is each building being supplied by a set of service entrance conductors, I believe you could use 250.64(D)(2) and would not have to have a GEC at the pedestal. Do others agree?
 

mopowr steve

Senior Member
Location
NW Ohio
Occupation
Electrical contractor
By the way, the meterbase is fed by 500kcmil.
And it's a meter/disco combo.

I think if your going to go the route of it being sized for lightning, again explain why that matters sizing it in correlation to service entrance conductors I'm sure you'll explode a smaller service just as well as a bigger one fed from the same transformer. Which then makes me think my question really is WHY? If it's not to provide a low resistance path to trip a OCPD, then is it sized to provide low resistance path to allow opening of a high line fuse? Well then why would it be ok to use one size GEC for a service and then another service get a larger or smaller size fed from the same transformer?

Does anybody really have the dig on what the GEC to a waterline is based on?
 

mopowr steve

Senior Member
Location
NW Ohio
Occupation
Electrical contractor
Yes, GEC to waterline goes to disconnect side of meter/combo and GEC to double set of rods.
Bonding done at meter/disco with EGC ran with feeders to sub panels.
 

charlie b

Moderator
Staff member
Location
Lockport, IL
Occupation
Retired Electrical Engineer
Didn't I try my best to avoid mentioning lightning (dang it, there I go mentioning it again!)? So for the record, I believe the size of the GEC is not based in any way on the possibility of it being involved in a lightning strike. Its existence might have a bearing on that possibility, but not its size. So can we forget about lightning for a while?

Again for the record (and please feel free to check out this statement with any information source you can find), as long as we are not dealing with systems of 4160 volt and above, the current flowing through a GEC will never, never be enough to trip any overcurrent device, no matter what its setting, no matter where in the circuit it might be, no matter what might be happening to the circuit at the time of the incident. At the very least, for the purposes of the design and installation of electrical systems, we must never take credit for the existence of the GEC when we are planning to protect either personnel or equipment from the damaging effects of overcurrent. Refer to the last sentence of 250.4(A)(5).

I believe that the size of the GEC is based on the size of the wires feeding the point at which the GEC is connected because the bigger those wires are, the harder will be the task of keeping the voltage between neutral and planet Earth held to near zero.

 

mopowr steve

Senior Member
Location
NW Ohio
Occupation
Electrical contractor
Thanks for input Augie,
Answer me this, if someone has a multigang meterbase fed with 500mcm and then has 200 amp disconnects under it, would you make them try and put 1/0 to there ground bar (most bars I've seen may have lugs for service neutral, feeder neutral then a set of holes that maybe accept up to #4max).
Id say more times than not you'll find a #4 ran to these not a 1/0. Which gets me to saying that is no different than what I have...... A meter, 2 main disconnects just happen to be in the same enclosure.
 

augie47

Moderator
Staff member
Location
Tennessee
Occupation
State Electrical Inspector (Retired)
Didn't I try my best to avoid mentioning lightning (dang it, there I go mentioning it again!)? So for the record, I believe the size of the GEC is not based in any way on the possibility of it being involved in a lightning strike. Its existence might have a bearing on that possibility, but not its size. So can we forget about lightning for a while?

Again for the record (and please feel free to check out this statement with any information source you can find), as long as we are not dealing with systems of 4160 volt and above, the current flowing through a GEC will never, never be enough to trip any overcurrent device, no matter what its setting, no matter where in the circuit it might be, no matter what might be happening to the circuit at the time of the incident. At the very least, for the purposes of the design and installation of electrical systems, we must never take credit for the existence of the GEC when we are planning to protect either personnel or equipment from the damaging effects of overcurrent. Refer to the last sentence of 250.4(A)(5).

I believe that the size of the GEC is based on the size of the wires feeding the point at which the GEC is connected because the bigger those wires are, the harder will be the task of keeping the voltage between neutral and planet Earth held to near zero.



:thumbsup:
 

mopowr steve

Senior Member
Location
NW Ohio
Occupation
Electrical contractor
Didn't I try my best to avoid mentioning lightning (dang it, there I go mentioning it again!)? So for the record, I believe the size of the GEC is not based in any way on the possibility of it being involved in a lightning strike. Its existence might have a bearing on that possibility, but not its size. So can we forget about lightning for a while?

Again for the record (and please feel free to check out this statement with any information source you can find), as long as we are not dealing with systems of 4160 volt and above, the current flowing through a GEC will never, never be enough to trip any overcurrent device, no matter what its setting, no matter where in the circuit it might be, no matter what might be happening to the circuit at the time of the incident. At the very least, for the purposes of the design and installation of electrical systems, we must never take credit for the existence of the GEC when we are planning to protect either personnel or equipment from the damaging effects of overcurrent. Refer to the last sentence of 250.4(A)(5).

I believe that the size of the GEC is based on the size of the wires feeding the point at which the GEC is connected because the bigger those wires are, the harder will be the task of keeping the voltage between neutral and planet Earth held to near zero.



I know that a GE of rod, plate, pipe electrode or Ufer will never be able to carry enough current to trip a OCPD hence why there Definite sizes only needed to connect them but a waterline GE can because of its all multitude of connections to the majority of electrical systems that defiantly gives a return path back to POCO transformer. Hence why I think it's size is tied to something but what exactly? Why would it not be to the maximum setting of The OCPD serving the building and not relative to service conductor size. I may have to go back and look at the language and how it's changed over the decades.
 
Thanks for input Augie,
Answer me this, if someone has a multigang meterbase fed with 500mcm and then has 200 amp disconnects under it, would you make them try and put 1/0 to there ground bar (most bars I've seen may have lugs for service neutral, feeder neutral then a set of holes that maybe accept up to #4max).
Id say more times than not you'll find a #4 ran to these not a 1/0. Which gets me to saying that is no different than what I have...... A meter, 2 main disconnects just happen to be in the same enclosure.

I depends on where you are landing the GEC. WIth multiple enclosures, you may use the tap method and in that case each tap would be sized to the conductors serving that enclosure. OR you may hit the neutral conductor before it splits into multiple sets with a single GEC. I dont see any ambiguity.
 

charlie b

Moderator
Staff member
Location
Lockport, IL
Occupation
Retired Electrical Engineer
I know that a GE of rod, plate, pipe electrode or Ufer will never be able to carry enough current to trip a OCPD hence why there Definite sizes only needed to connect them but a waterline GE can because of its all multitude of connections to the majority of electrical systems that defiantly gives a return path back to POCO transformer.
We are discussing the wire that connects the electrode to the neutral bus at the first overcurrent device. That is, we are discussing the GEC, not the GE. For a water pipe to be used as a grounding electrode, it must be in contact with dirt for at least 10 feet, and we have to be able to connect to it within 5 feet of where it enters the building. Beyond 5 feet, the electrical distribution system no longer cares about the water pipe. The purpose of connecting a GEC to the water pipe GE is to, once again, make the voltage of the neutral bus the same as that of the dirt. It is not there to carry fault current back to the POCO transformer. Other wires are doing that job. Indeed, if the water pipe is plastic inside the building but metal up to the point of our connection, then it still does its one and only job as a grounding electrode.

If the entire water system piping is metal, then you are right in saying it is essentially electrically connected to every metal conduit and every electrical enclosure throughout the building. If an ungrounded (i.e., "hot") wire comes into contact with the metal case of the equipment of which it is a component, then there will be fault current flowing through all available paths back to the main service panel, and from there to the POCO transformer. It is not impossible that some of that current will flow along the metal water pipe, and then up the GEC to the main panel. But that would require there to be a bond wire from a point on the water pipe someplace else within the building path to a metal conduit or an electrical enclosure. We might get that happening if there is a water pump in the building. That that path is not needed. The EGC that serves the faulted equipment is enough to trip the breaker and terminate the event. The GEC is simply not relevant to this protective action.

To pursue the point even further, suppose there is a water pump with a metal case. The EGC serving that pump would be bonded to the case, and there is a metal-to-metal connection between the case of the pump and the water pipe. If the water pump's motor had the fault, then I can see two parallel paths. One will go from fault point inside the motor to the case of the motor, to the water pipe, to the point of connection of the GEC to the water pipe, and then to the main panel. The other will go from fault point inside the motor to the case of the motor, to the EGC, and from there directly to the main panel. That second path will cause the breaker to trip. We are not taking credit for that first path. Indeed, since the plumber is not forbidden to replace metal pipe with plastic pipe, we don't even know if that first path really exists.

 
Last edited:

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
The two things that sizing the water pipe GEC to the incoming conductors can do are:
A. Making sure that it can safely carry the full normal neutral current if the service neutral opens and there is a water pipe to POCO neutral path through the house(s) next door. This could just as well be sized to the OCPD. and
B. Trying to limit the fault voltage on the EGCs to only half the line to neutral voltage until the OCPD opens. (Same water pipe to neighbor's GEC and bond) This one depends on wire size not OCPD.
 

mopowr steve

Senior Member
Location
NW Ohio
Occupation
Electrical contractor
Charlie, I'm fully aware of everything you say about grounding and bonding except where sizing GEC to waterline electrode. I am looking for a much more technical answer to the sizing of a GEC to a waterline which qualifies as an GE (which must be larger than any other GE method) not an answer of because it's harder to reference neutral to ground with larger service conductors. I'm asking about the particular formula, reasons/considerations to its size as quoted in NEC.
Sorry harder doesn't cut the mustard, unless your going to tell me that the resistance and impedance of the GEC has some drastic difference from one size to another. Or as I've said before, because of its possible inherence to parallel a neutral back to transformer as more a bonding issue than GE (in which case, it would seem to be sized to some fault condition. Then what is expected to happen? open a fuse, trip a OCPD, be able to carry neutral load, other?) After all POCO's use #6-#4 everyday down the pole to bring ground reference to neutral (and yes maybe lightning mitigation).
Actually going down this road one would think it some percentage of max available current often associated to certain size service conductors.
 

mopowr steve

Senior Member
Location
NW Ohio
Occupation
Electrical contractor
The two things that sizing the water pipe GEC to the incoming conductors can do are:
A. Making sure that it can safely carry the full normal neutral current if the service neutral opens and there is a water pipe to POCO neutral path through the house(s) next door. This could just as well be sized to the OCPD. and
B. Trying to limit the fault voltage on the EGCs to only half the line to neutral voltage until the OCPD opens. (Same water pipe to neighbor's GEC and bond) This one depends on wire size not OCPD.

Now we're talking! That opens the thought process I'm looking for somewhat.
So let's say service neutral opens, now neutral current flows thru metal waterlines in turn utilizing neighbors waterline GEC to return thru their neutral back to transformer. All the while my GEC is sized large enough to carry full load current (not really or it would be the same size as required for my neutral conductor) and the poor neighbor all of a sudden find his place on fire because his little GEC couldn't handle the load. Makes one think all GEC's to waterlines qualifying as GE's should be sized to trip the largest service OCPD on the system or sized to safely bolt a secondary fault to the point of blowing the primary fuse.
What do you think about that?
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
I think it is sufficiently improbable (as well as hard to quantify) that the NEC is unlikely to consider it. And I probably could not justify the expense to a customer.
Since the neutral and the GEC are not allowed to be fused, maybe an alarm would be in order. Note that opening the neighbor's service disconnect does nothing to protect him.
A GF detector on your service wires, on the POCO side would protect you both. A relatively insensitive RCD would do just fine. Too sensitive would cause a problem since some normal current will flow back to POCO on the GEC without any fault present.
 

ActionDave

Chief Moderator
Staff member
Location
Durango, CO, 10 h 20 min from the winged horses.
Occupation
Licensed Electrician
Now we're talking! That opens the thought process I'm looking for somewhat.
So let's say service neutral opens, now neutral current flows thru metal waterlines in turn utilizing neighbors waterline GEC to return thru their neutral back to transformer. All the while my GEC is sized large enough to carry full load current (not really or it would be the same size as required for my neutral conductor) and the poor neighbor all of a sudden find his place on fire because his little GEC couldn't handle the load. Makes one think all GEC's to waterlines qualifying as GE's should be sized to trip the largest service OCPD on the system or sized to safely bolt a secondary fault to the point of blowing the primary fuse.
What do you think about that?
I went to a Soare's Grounding class a few years back and the presenters said that there is no explanation given in the code or other references for GEC sizing and the best reason they could think of was basically what you just said.
 
Status
Not open for further replies.
Top