What's the consensus

Status
Not open for further replies.
mopowr steve said:
Does anybody really have the dig on what the GEC to a waterline is based on?
Click to expand...
250.66 - which if you learn how to use it properly you will see that it is based on the size of incoming conductors and has nothing to do with overcurrent device selection.

You have a 500kcmil supply conductor - that is the basis whether you have a single 30 amp disconnect being supplied or 6 200 amp disconnects being supplied. (6 -200's is not impossible here - depends on load calculations)
 
I understand the NEC associates the underground metal waterline GEC size to the service entrance conductor size, buts that's the rub. What does sizing it to these conductors expect to accomplish? In other words how did the original author arrive to its sizing?

Is there some documentation on measured fault current flow at multiple worst case scenario accidents that drives its size. Seems odd when other factors than just the size of service entrance conductors should determine its size (such as transformer sizing, fuse sizing, and if not; then why not just size to main OCPD) I would think.

A couple of good viable points have been made in some of the previous posts though.

Sorry just being analytical.
 
mopowr steve said:
I understand the NEC associates the underground metal waterline GEC size to the service entrance conductor size, buts that's the rub. What does sizing it to these conductors expect to accomplish? In other words how did the original author arrive to its sizing?

Is there some documentation on measured fault current flow at multiple worst case scenario accidents that drives its size. Seems odd when other factors than just the size of service entrance conductors should determine its size (such as transformer sizing, fuse sizing, and if not; then why not just size to main OCPD) I would think.

A couple of good viable points have been made in some of the previous posts though.

Sorry just being analytical.
Click to expand...
Has been mentioned already, but hasn't quite sunk in yet - grounding electrodes are not intended to serve the purpose of fault clearing and typically will never do so for under 1000 volts. A water pipe electrode may have another service bonded to it on another property and have low enough resistance path but the resistance of pipe to earth if that scenario wasn't there would be higher.

I honestly don't know how or why they determined the values in 250.66, but also remember they take in some reality when they only require a 6 AWG max to a single rod or pipe electrode, or a 4 AWG max to a concrete encased elecrode, AFAIK in those cases they figure a rod will never dissipate any more energy then a 6 AWG can deliver to it, same for the 4 AWG and a CEE. This must mean they expect a water pipe or building steel to be able to dissipate more energy then a rod or a CEE does in general since they do require larger GEC run to them in some situations.
 
If I recall correctly the origin of these tables was an IEEE committee report “A Guide to Safety in AC Substation Grounding”. The committee report discussed the validity of grounding conductor sizes specified in the tables – based on a typical conductor length of 100’ and the voltage drop over the conductor based on this 100’ length.
The orig
 
On second thought (it's been at least 15 years or so) I may be remembering where GEC conductor sizing factors are discussed at length in IEEE Std 80. It takes an adult beverage or six to digest the math involved, or else your mind will really hurt. :D
 
Everybody don't get excited I know that a GEC or its system is by definition not there to carry fault current. I totally understand that. But electrons can't read. They will take all and any possible paths back to the source.

I am thinking that when a GEC lands on a metallic underground waterline that it in fact must do both, typical GEC function (grounding) and (bonding) and be able to carry a fault of some magnitude (because it will) But what is this magnitude and why is it related to service conductor size shouldn't it be some other factor(s). Otherwise why would it be sized larger than other GEC to GE's of rod,pipe,plate,Ufer,ring.

When it comes to bonding we size to the circuit that could impose a fault on that metal system, so how in the H is service entrance conductors going to do that there not even close to each other. Ahh, by the connection of that GEC wire. So let's say a fault occurs inside the mast of a service and the neutral back to the transformer is compromised and one of the hot conductors has now bolted to the neutral wire feeding the service. Current is going to flow on that GEC to waterline, and back to the transformer utilizing other people's service neutrals. Is this how they arrive to its size? Again what is its size supposed to accomplish, be able to safely take a bolted fault until a primary fuse blows? But that current flow would be based on the transformers output capability (or max available current at the location of fault) not what size service entrance conductors are. Now if this were the case I would think the GEC would be even larger so why don't we just dismiss this idea and just cover the things we can control like sizing it to the main OCPD serving a building instead and just cover faults from load side into building.

Ill have to check out that IEEE report.
 
mopowr steve said:
Everybody don't get excited I know that a GEC or its system is by definition not there to carry fault current. I totally understand that. But electrons can't read. They will take all and any possible paths back to the source.

I am thinking that when a GEC lands on a metallic underground waterline that it in fact must do both, typical GEC function (grounding) and (bonding) and be able to carry a fault of some magnitude (because it will) But what is this magnitude and why is it related to service conductor size shouldn't it be some other factor(s). Otherwise why would it be sized larger than other GEC to GE's of rod,pipe,plate,Ufer,ring.

When it comes to bonding we size to the circuit that could impose a fault on that metal system, so how in the H is service entrance conductors going to do that there not even close to each other. Ahh, by the connection of that GEC wire. So let's say a fault occurs inside the mast of a service and the neutral back to the transformer is compromised and one of the hot conductors has now bolted to the neutral wire feeding the service. Current is going to flow on that GEC to waterline, and back to the transformer utilizing other people's service neutrals. Is this how they arrive to its size? Again what is its size supposed to accomplish, be able to safely take a bolted fault until a primary fuse blows? But that current flow would be based on the transformers output capability (or max available current at the location of fault) not what size service entrance conductors are. Now if this were the case I would think the GEC would be even larger so why don't we just dismiss this idea and just cover the things we can control like sizing it to the main OCPD serving a building instead and just cover faults from load side into building.

Ill have to check out that IEEE report.
Click to expand...
Larger service conductors will carry more fault current - you still have source impedance and length of the run that can present counter effects though. But I would guess someone has determined how large of a conductor is needed to be able to withstand certain current levels for certain amount of time before that conductor starts to become damaged. If POCO happens to have oversized fuse on primary a fault could remain until something gives up, connections are first candidates, unless conductors are really small in relation to transformer capacity transformer will probably let smoke out if connections don't fail first.
 
It kind of makes sense if you think of a pipe GEC as a backup neutral. An undersized GEC would result in more current taking even less desirable paths, e.g. EGC to water using appliances, in the event of a lost neutral. I don't know if this was the rationale though.
 
charlie b said:


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.
Click to expand...

If a hot service entrance conductor faults, but the amperage of the fault is not enough to burn the transformer fuse, the service enclosure is energized and poses an electrocution risk. With a low impedance connection to earth, the risk of shock is greatly reduced since the potential of what a person is standing on is, hopefully, equal to that of the charged enclosure. The larger the service entrance conductor, the more it can deliver, so a larger GEC stands a better chance of maintaining equal potential in such a fault scenario. Since service entrance conductors are out of the control of the users, this is the most that can be done to protect them when they walk up to the box to open it because the lights are flickering.
 
jeffgreef said:
If a hot service entrance conductor faults, but the amperage of the fault is not enough to burn the transformer fuse, the service enclosure is energized and poses an electrocution risk. With a low impedance connection to earth, the risk of shock is greatly reduced since the potential of what a person is standing on is, hopefully, equal to that of the charged enclosure. ....
Click to expand...
That would only be true if you were standing right on top of the grounding electrode.
 
ActionDave said:
That would only be true if you were standing right on top of the grounding electrode.
Click to expand...
I agree.

Stand over top of it and spread feet as far as you can and you likely have some voltage between them though it may only be as high as 5-10 volts depending on conditions of soil, for a 120 to ground source, 277 to ground maybe gets it high enough you do feel it in places where you wouldn't with 120 volts. Get into over 1000 volt systems, and you just seem to get struck from out of nowhere and end up dead when walking over such a zone.
 
Granted, yes, the potential of the earth below will be less than that of the enclosure and system electrode, depending on how far away the electrode is. As well, the fact that the electrode is bonded to neutral gives a possible fault path back to the transformer of enclosure, person, earth, electrode, GEC, neutral bond which might lead to a shock. I wonder if current would take that fault path, since the fault has much easier access to the neutral directly from the enclosure. However another fault path exists through earth, to the neighbor's electrode or the transformer's. The GEC/electrode on the faulted system reduces the possibility that fault current would take the path through the person to those other electrodes, because the system electrode is far better connected to earth. Again the larger the GEC the better capable it is of handling the amperage from larger entrance conductors. But perhaps that doesn't matter for this fault path because of the impedance of the earth which will severely limit fault circuit amperage.
 
jeffgreef said:
Again the larger the GEC the better capable it is of handling the amperage from larger entrance conductors. But perhaps that doesn't matter for this fault path because of the impedance of the earth which will severely limit fault circuit amperage.
Click to expand...

Exactly, and that is why the GEC to a ground rod is only required to be 6 AWG even for a 4,000 amp service or 4 AWG for a concrete encased electrode.

The only GEC that might have to be as large as 3/0 is the waterline and that makes sense as it may have other electrical connections besides the earth back to the source.
 
jeffgreef said:
But perhaps that doesn't matter for this fault path because of the impedance of the earth which will severely limit fault circuit amperage.
Click to expand...

For 1000 volts and below the grounding elecrode is always going to be a higher impedance then the service conductors. Not that the earth isn't a low impedance conductor, but making a low impedance connection to it is difficult to do.

For a high voltage high frequency lightning strike or even over 1000 volts POCO distribution voltages - the amount of current that takes each path will be different then it is for 1000 volts and less.
 
Status
Not open for further replies.
Top