GFI breaker protection

Status
Not open for further replies.

ohmhead

Senior Member
Location
ORLANDO FLA
Question any breaker 1000 amp and larger must have GFI protection .

So if I had a switchboard with a service main breaker at 4000 amp with GFI protection and had two other 1000 amp breakers in that same switchboard they would need gfi protection in addition to the main.
 

texie

Senior Member
Location
Fort Collins, Colorado
Occupation
Electrician, Contractor, Inspector
Question any breaker 1000 amp and larger must have GFI protection .

So if I had a switchboard with a service main breaker at 4000 amp with GFI protection and had two other 1000 amp breakers in that same switchboard they would need gfi protection in addition to the main.

GFP is only required if more than 250 volts to ground. So a 480/277 service would need it but a 208/120 would not.
The feeders would only need GFP if coordination is required, such as a hospital.
 

ohmhead

Senior Member
Location
ORLANDO FLA
GFP is only required if more than 250 volts to ground. So a 480/277 service would need it but a 208/120 would not.
The feeders would only need GFP if coordination is required, such as a hospital.

This is a large preforming arts center we have 4# 4000amp services most of the switchboards have a few 1000 amp or 1600 amp sections a few 1200 amp branch .
there is coordination on the system . Were is that in the NEC what article I was thinking a few years back it was 800 amp but I guess that changed .
thanks for your help .
 

texie

Senior Member
Location
Fort Collins, Colorado
Occupation
Electrician, Contractor, Inspector
This is a large preforming arts center we have 4# 4000amp services most of the switchboards have a few 1000 amp or 1600 amp sections a few 1200 amp branch .
there is coordination on the system . Were is that in the NEC what article I was thinking a few years back it was 800 amp but I guess that changed .
thanks for your help .

215.10 and 230.95. As I recall, it has always been 1000 amp.
 

ohmhead

Senior Member
Location
ORLANDO FLA
215.10 and 230.95. As I recall, it has always been 1000 amp.

So what I read is 1000 amps and larger below 600 volts now if we had a series rated system with a GFI installed at the service main meaning the 4000amp main service disconnect we don't need to have GFI on any breaker down stream .
 

hurk27

Senior Member
215.10 and 230.95. As I recall, it has always been 1000 amp.

Correct, my 1999 NEC on this computer has the same wording as my 2008 in 230-95 (99) if I remember right it was also in my 1996 NEC

I have seen a few engineers wrongly try to spec them on deltas as well as 208/120 volt systems.
They are only required for solidly WYE grounded services or feeders of more then 150 volts to ground and up to 600 volts phase to phase.

Exception 2 to 215.10 eliminates the need for a GFP on a feeder if the service has GFP, but it also requires a GFP on a feeder if the feeder is fed from a SDS transformer even if the service has GFP, that is because the primary side of a transformer will only see a line to line current from a line to ground fault on its secondary.

I have had a few inspectors get confused with exception 2 saying you can only use it if you have both supply side GFP and a GFP after a transformer, but the words "and on the load side of any
transformer supplying the feeder." was added in 2008 which really confused the exception as it was in 2005: "The provisions of this section shall not apply if ground-fault protection of equipment is provided on the supply side of the feeder"
To me it should say "The provisions of this section shall not apply if ground-fault protection of equipment is provided on the supply side of the feeder except when feeder is supplied from the load side of a transformer a GFP shall be require to meet 215.10."

The commentary to the 2011 makes it clear that the intent is to not require GFP on a feeder if the service has GFP unless the feeder is supplied through an SDS transformer:

Ground-fault protection of feeder equipment is not required if protection is provided on an upstream feeder or at the service.
 

ohmhead

Senior Member
Location
ORLANDO FLA
Thanks for the input.
We have a 4000 amp main with GFI protection installed in that one switchboard there is a 1000 amp and a 1200 amp breaker each of these in fact all breakers in all our switchboards are 100 % rated typical AIC which matches the 4000amp main its 100K AIC rated and each breaker from 60 amp to 1200 amp has a 100k AIC rating so there is no breaker less than the main breaker AIC . Its 100 percent rated gear the coordination is the low settings down stream per engineer .

So can the inspector require me to put a GFI on the down stream breakers in switchboard has something change I may not be aware of .
 

augie47

Moderator
Staff member
Location
Tennessee
Occupation
State Electrical Inspector (Retired)
Two questions, if I may.
One for ohmhead... I may have missed it in a post, but is your system 480Y/277 ?

2nd question for folks familiar with coordinated systems. Can you have adequate selective coordination with a main that has GFP where none of the branch have GFP ???

(I just inspected an installation where the main and each feeder has GFP and I was told it was necessary on a coordinated system in case of a ground fault on a feeder)
 

hurk27

Senior Member
Thanks for the input.
We have a 4000 amp main with GFI protection installed in that one switchboard there is a 1000 amp and a 1200 amp breaker each of these in fact all breakers in all our switchboards are 100 % rated typical AIC which matches the 4000amp main its 100K AIC rated and each breaker from 60 amp to 1200 amp has a 100k AIC rating so there is no breaker less than the main breaker AIC . Its 100 percent rated gear the coordination is the low settings down stream per engineer .

So can the inspector require me to put a GFI on the down stream breakers in switchboard has something change I may not be aware of .

The inspector shouldn't require GFP for the feeder breakers as I said above 215.10 exception 2 does not require feeder breakers to have GFP if there is already GFP protection ahead of these breakers.

Now there are parts of the NEC that does such as see 517.17, which requires an additional
level of ground-fault protection for health care facilities, but from your OP it shouldn't apply to your siduation.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
Can you have adequate selective coordination with a main that has GFP where none of the branch have GFP ???)
If your definition of adequate includes requiring the branch breaker to trip rather than the main in the event of a line to ground fault, irrespective of whether the fault current is above the instantaneous or thermal trip point of the branch breaker, then I would have to say NO.
If you look at the time characteristics of the GF trip versus the instantaneous trip, then the answer requires information I do not have.
 

ohmhead

Senior Member
Location
ORLANDO FLA
Two questions, if I may.
One for ohmhead... I may have missed it in a post, but is your system 480Y/277 ?

2nd question for folks familiar with coordinated systems. Can you have adequate selective coordination with a main that has GFP where none of the branch have GFP ???

(I just inspected an installation where the main and each feeder has GFP and I was told it was necessary on a coordinated system in case of a ground fault on a feeder)

Well Augie according to the engineer who did the coordination study its not needed and yes its 480/277v system but the inspector is looking for it down stream .

Square-D has designed the system and the engineer who designed the job has signed it off approved so iam asking the question is the inspector correct is there something wrong with this install.
There are other switchboards down stream from the one were discussing .
The main breaker in this has a AIC of 65 k and all there breakers internal in each switchboard is rated at 65K.
The breaker which feeds these switchboards from the main switchboard are 100K as stated . So were feeding a branch circuit with a 1600amp breaker rated at 100K that feeder goes to a remote switchboard which has a main breaker rated at 65K and all the breakers in that remote switchboard are rated at 65K .
 

ohmhead

Senior Member
Location
ORLANDO FLA
The inspector shouldn't require GFP for the feeder breakers as I said above 215.10 exception 2 does not require feeder breakers to have GFP if there is already GFP protection ahead of these breakers.

Now there are parts of the NEC that does such as see 517.17, which requires an additional
level of ground-fault protection for health care facilities, but from your OP it shouldn't apply to your siduation.


This is not a hospital just a performing arts center I guess this is what one calls a fully rated system yes or no due to the fact its a match for match on the AIC rated breakers .
The reason is the inspector I know and he is a very qualified NEC code guru so iam questioning our engineering and scope on this one .
Plus folks due make mistakes some times so if this was a series rated issue we would need a GFI if the AIC were lower in the branch .
 

augie47

Moderator
Staff member
Location
Tennessee
Occupation
State Electrical Inspector (Retired)
Some inspectors have "off days" and, I even know one who has "senior moments":D. You might ask him to review 215.10 Exception 2 with you.
 

hurk27

Senior Member
Two questions, if I may.
One for ohmhead... I may have missed it in a post, but is your system 480Y/277 ?

2nd question for folks familiar with coordinated systems. Can you have adequate selective coordination with a main that has GFP where none of the branch have GFP ???

(I just inspected an installation where the main and each feeder has GFP and I was told it was necessary on a coordinated system in case of a ground fault on a feeder)

For the second question, I had a 16 plex theater that we had a problem with a GFP that the engineers had wrongly required on the 208/120 volt MDP main the problem we were called on was that all the normal lighting was on the 208/120 system which was fed from a 480-208/120 750kva transformer, the service to the building was 1600 amp 480/277 which also had a GFP on it's main, the original contractors were fired for other problems so a coordination study was never done and all the breaker settings remained at there factory settings of the minimum that they could be set at, all the emergency lighting was all supplied from the 480/277 volt system because all the normal lighting was controlled by the branch circuit breakers which were the PLC controlled GE 2k smart system series so the emg lighting couldn't be put on the same circuit as the normal lighting, what was later found out (after the coordination was done) that a 208 volt popcorn machine had a damage cord and when they moved the machine out for cleaning it would fault phase to ground which tripped out the GFP on the 208/120 MDP main, this caused all the regular lighting to go dark and because the 480/277 system didn't go down no emg lighting would come on, while we worked with the state fire marshal for a solution because we had only two choices to either re-pipe all the emg lights to the 208/120 volt panels which there was hundreds of them, or install a system that if the 208/120 volt system ever went down it would shunt trip out the 480/277 volt main so all the emg lights would come on, well the cost of moving all the emg lights to the 208/120 volt system was way to costly so the owners opted for the shunt trip system since it didn't really matter as they were down if they lost the 208 system anyway so killing the 480/277 volt main would not make any difference, you just had one more handle to pump up to charge the main switch.

Anyway we also had GE engineers come in and do a coordination study and coordinate all the breaker settings, while they were there they showed me that without the coordination that even a fault at a 120 volt 20 amp receptacle the GFP would still trip, they also did a line to neutral fault that also tripped out the 208/120 volt main, because the settings were still at factory minimum, after the coordination was done and they had set all the settings on the breakers, they showed us that even a 200 amp fault on one of the 208/120 volt local panels only the feeder breaker for that panel tripped, the next night when they went to move out the pop corn machine we found the problem with its cord since now it tripped its BC breaker which before it didn't because only the GFP main would trip, so as far as 200 amp and below I know that a correctly done coordination study as well as correct breaker settings you don't need a GFP on the feeders to achieve this, as long as you can adjust the instantaneous setting just a little faster on the down stream breakers then what the upstream breakers are set at, it is possible to keep the upstream breakers from tripping, but the problem is the GFP, if the engineers are restricted to a lower GF current setting on the GFP or a faster instantaneous time setting because of weaker ground fault current paths then you might not be able to set the down stream settings any lower and you risk that a GF will trip the GFP on the main, luckly the wiring methods used by this contractor by using RMC with bonding bushings as well as installing an aditional green #3 EGC in each of these feeder conduits to each 200 amp down stream local panel the GE engineers determined that the ground fault paths would take more fault current and were able to set the main GFP at a higher setting and longer time setting for the instantaneous setting giving more flexibility for the settings of the downstream feeder breakers, so in the end we wound up with only the down stream breaker feeding the GF that would trip and not the main GFP.

But also since we still had the problem of the emg lighting being on the 480/277 volt system and the normal lighting being on the 208/120 volt system, so we still had to install the shunt trip system so the 480/277 main would trip and turn on the emg lights if for any reason the 208/120 volt system was to ever fail so the theaters would not be left in the dark, also since the elevators had UPS backup that would take the car to the ground floor and lock it out in the event of a power failure it was not part of the problem.
 
Last edited:

ohmhead

Senior Member
Location
ORLANDO FLA
Its my theory that a fully rated system at 100% AIC is for a fault only . And has nothing to do with coordination of a breaker tripping down stream if the settings are lower than the up stream breakers and a study is made by a engineer who is paid well . But I don't like it as if any branch has a ground fault the main breaker will trip meaning everything out of the main is off shows over folks . I think a build with people sitting in seats that cost 500 bucks a pop and the show stops do to a fully rated system is not a smart design or plan . Which is better series with a GFI or fully rated with coordination of breakers what do you think would be your choice .
 

hurk27

Senior Member
Its my theory that a fully rated system at 100% AIC is for a fault only . And has nothing to do with coordination of a breaker tripping down stream if the settings are lower than the up stream breakers and a study is made by a engineer who is paid well . But I don't like it as if any branch has a ground fault the main breaker will trip meaning everything out of the main is off shows over folks . I think a build with people sitting in seats that cost 500 bucks a pop and the show stops do to a fully rated system is not a smart design or plan . Which is better series with a GFI or fully rated with coordination of breakers what do you think would be your choice .

I think we have a misunderstanding of what the AIC rating is, I might be wrong but your reply seems to me that your thinking the AIC rating is the level at which a breaker will open in a bolted fault?

The AIC rating is nothing more than the level at which the breaker can safely open a fault at that level without destroying it's self, it has nothing to do with the level of current or how fast in which the breaker will open.

The current level and how fast the breaker will open at a given amount of current is the level set in the breakers design operating trip curve, with adjustable breakers some parts of this curve can be modified to provide a selective coordination, and or GFP's can be added to down stream breakers, but this will only provide ground fault coordination not normal current path coordination (line to neutral or line to line faults).

A breaker trip curve has three areas that are of concern when selecting or setting the breaker to protect the conductors that it supply's, the long term region, short term region, and instantaneous region.

Long term is the handle rating of the breaker that can open as fast as 60 seconds or much longer not uncommon to see a breaker hold for hours a few amps over its rating.

Short term is the area at which a breaker will pick up at extreme overloads that are much higher then the rating of the breaker handle but not as high as a bolted fault.

Instantaneous is the area we are concerned when we do selective coordination, as this is the area where a bolted fault can not only exceed the Instantaneous part of the curve for both the down stream (BC) OCPD as well as the up stream (feeder) OCPD it can also exceed the Instantaneous part of trip curve on the service main OCPD and cause all three to trip which is not good.

For a more in depth article about this here is a good EC&M article that covers it a little more then I can:

Selective Coordination Using Circuit Breakers

In all the years I been working as an electrician, I never fully understood the the difference and or had allot of mis-information that was picked up from others over the years, until I ran into the problem at that theater, it was something the engineers had always handled and I never really thought about it much other then I often wondered why a bolted fault on a 30 amp BC would trip the 100 amp main also which I found out the hard way when I accidentally faulted an AC condenser unit I was working on and was embarrassed when the home owner came out to tell me all his power went out to which I didn't have a clear answer for him, but now I know.

Also the reason why this is a job for an engineer is there is allot of info that must be known such as the thermal characteristic's of conductors to know how many cycles a conductor can withstand at a given amount of current before damage to the insulation of the conductor happens, and a few other things that most electricians never deal with, without this info you can't just start adjusting the dials on breakers or you could have some liability issues that could land you in a court of law after the smoke clears.
 
Last edited:

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
I think we have a misunderstanding of what the AIC rating is, I might be wrong but your reply seems to me that your thinking the AIC rating is the level at which a breaker will open in a bolted fault?

The AIC rating is nothing more than the level at which the breaker can safely open a fault at that level without destroying it's self, it has nothing to do with the level of current or how fast in which the breaker will open.

The current level and how fast the breaker will open at a given amount of current is the level set in the breakers design operating trip curve, with adjustable breakers some parts of this curve can be modified to provide a selective coordination, and or GFP's can be added to down stream breakers, but this will only provide ground fault coordination not normal current path coordination (line to neutral or line to line faults).

A breaker trip curve has three areas that are of concern when selecting or setting the breaker to protect the conductors that it supply's, the long term region, short term region, and instantaneous region.

Long term is the handle rating of the breaker that can open as fast as 60 seconds or much longer not uncommon to see a breaker hold for hours a few amps over its rating.

Short term is the area at which a breaker will pick up at extreme overloads that are much higher then the rating of the breaker handle but not as high as a bolted fault.

Instantaneous is the area we are concerned when we do selective coordination, as this is the area where a bolted fault can not only exceed the Instantaneous part of the curve for both the down stream (BC) OCPD as well as the up stream (feeder) OCPD it can also exceed the Instantaneous part of trip curve on the service main OCPD and cause all three to trip which is not good.

For a more in depth article about this here is a good EC&M article that covers it a little more then I can:

Selective Coordination Using Circuit Breakers

In all the years I been working as an electrician, I never fully understood the the difference and or had allot of mis-information that was picked up from others over the years, until I ran into the problem at that theater, it was something the engineers had always handled and I never really thought about it much other then I often wondered why a bolted fault on a 30 amp BC would trip the 100 amp main also which I found out the hard way when I accidentally faulted an AC condenser unit I was working on and was embarrassed when the home owner came out to tell me all his power went out to which I didn't have a clear answer for him, but now I know.

Also the reason why this is a job for an engineer is there is allot of info that must be known such as the thermal characteristic's of conductors to know how many cycles a conductor can withstand at a given amount of current before damage to the insulation of the conductor happens, and a few other things that most electricians never deal with, without this info you can't just start adjusting the dials on breakers or you could have some liability issues that could land you in a court of law after the smoke clears.
One place where the AIC could come into play would be if the downstream breaker were forced to try to open against a fault higher than its AIC because the upstream OCPD was delayed.
Hopefully the available fault current calculation at the downstream device will be lower than its AIC though.
 

ohmhead

Senior Member
Location
ORLANDO FLA
INSPECTORS RESPONCE

INSPECTORS RESPONCE

First thanks for the input guys today we had a pre inspection we talked about the GFI,s down stream issue and he agreed to no GFI,s in this case it was not needed I just thought
you would like to know the out come from my post . Thanks again
 

Jraef

Moderator, OTD
Staff member
Location
San Francisco Bay Area, CA, USA
Occupation
Electrical Engineer
If your definition of adequate includes requiring the branch breaker to trip rather than the main in the event of a line to ground fault, irrespective of whether the fault current is above the instantaneous or thermal trip point of the branch breaker, then I would have to say NO.
If you look at the time characteristics of the GF trip versus the instantaneous trip, then the answer requires information I do not have.
But you CAN get what is called "Zone Selectric Interlocking" to coordinate the Instantaneous and GF protection scheme with downstream breakers so that up stream devices are restrained until the fault can be cleared as close to the source of the fault as possible.

One caveat to the OPs question:
Do we know for sure that the 1000A breakers are FEEDERS off of the 4000A main? Theoretically they could ALL be "Service Mains" if there are less than 6 in the switchboard. If that were the case, they ALL would need to have GF included.
 

iwire

Moderator
Staff member
Location
Massachusetts
One caveat to the OPs question:
Do we know for sure that the 1000A breakers are FEEDERS off of the 4000A main? Theoretically they could ALL be "Service Mains" if there are less than 6 in the switchboard. If that were the case, they ALL would need to have GF included.

The '4,000 amp main' has been cut from its source of supply and re-fed from a feeder that has GFP.

Everything supplied from that 4000 amp gear is now GFPed.
 
Status
Not open for further replies.
Top