Proper Value Setting for GFI Relays When No Coordination Study Exists.

[Michael4444]

Very often during our Field Work we find GFI Relays set at their Minimum (Amps) Pick-Up value. This could lead to Nuisance Tripping.

On other instances, we have seen the GF Relays set at Max (1200 A ), which, depending on the actual electrical load at the Facility, may in fact Defeat the purpose of the GF Relay.

We must remember that GFI protection has One Main Purpose in life, and that is to protect Plant Personnel from the results of Catastrophic Fire. The Protection of ancillary equipment and the Facility in general is Secondary in nature to that of saving a life or to keep someone from suffering personal Injury

It is very important to set the GFI Relay to a proper value, as dictated by the Electrical Load at the particular Facility.

Ideally, a recent or updated Coordination Study exists on File; however more often than not, this is not the case.

If we are working with insufficient infromation, in order to advice the client properly, we must make a definitive, accurate decision for choosing a proper Pick-up value on the GFI .

Ths Pick-Up setting Value is never a "Guess" but rather should always comprise of a conservative estimate based on the most recent existing Electrical Load Data at the Particular location.

This value must be low enough so as to Not cause a GFI Defeat Condition; and yet High enough not to cause Nuisance tripping of the Ground Fault Relay.

In our experience, when setting the Pick-Up function on a Facility's Main Circuit Interrupter Device's GFI Relay, and where no coordination study exists, a conservative value is to set the Pick-Up at 20% of existing maximum electrical load at the particular Facility..

If that specific 20% value is not one of the Fixed Ranges available at the GFI Relay Control Module, go UP to the next available Fixed Range Value.

For example, let us assume that the Max Load is 800 Amps. Ideally, the GFI Pick-Up should be set at 160 Amps. But the control Module settings are 100, 200, 300, --> 1200 Amps..

In our example, we would set the GFI Relay Pick-Up at 200 Amps.

Time Delay, if present as an available Function, would be set at Minimum. NOT Instantaneous, but Minimum.

This Time Delay setting is adequate in order to compensate for our expanded Ampere Range setting with a faster Time Delay response.


Again, a 20% calculation based on a Facility's actual Max Load for setting the GFI Value Range is a conservative value which will protect personnel and equipment adequately, yet will not be likely to cause unnecessary Tripping of the Main Circuit Interruptor...


Have a Happy Ground-Fault Day..

 

[iwire]

Very often during our Field Work we find GFI Relays set at their Minimum (Amps) Pick-Up value. This could lead to Nuisance Tripping.

I also see this often.

It is very important to set the GFI Relay to a proper value,

I agree.

If we are working with insufficient infromation, in order to advice the client properly, we must make a definitive, accurate decision for choosing a proper Pick-up value on the GFI .

Ths Pick-Up setting Value is never a "Guess" but rather should always comprise of a conservative estimate based on the most recent existing Electrical Load Data at the Particular location.

This value must be low enough so as to Not cause a GFI Defeat Condition; and yet High enough not to cause Nuisance tripping of the Ground Fault Relay.

In our experience, when setting the Pick-Up function on a Facility's Main Circuit Interrupter Device's GFI Relay, and where no coordination study exists, a conservative value is to set the Pick-Up at 20% of existing maximum electrical load at the particular Facility..

I always refuse to guess or change the settings, it is not up to me to take on this liability for company I work for. I simply tell them that they need to have an engineer do a study.

Consider this, if no real study has been done up to this point if and I go ahead and change some settings what will be the chances be that they ever decide to have the study done?

On the other hand if I write it up as 'GFP settings on minimum, may cause nuisance tripping of main. A coordination study should be done to determine settings.' They now have a real reason to spend the money on having an engineer look at it.


And welcome to the forum.

 

[Michael4444]

I understand what you said. Your concerns are valid. But do rest assured that the 20% figure is not something pulled out of thin air..

Many years ago, we were employed by one of the two largest Electrical Switchgear manufacturers in the USA as Field Test Engineers.

At that time, the 20% calculated range for GFI range setting was what we were instructed to apply by Lead Project Engineers at said Company.

This figure is a GFI Relay range value which has served us very well back then, and for the past 26 years during which we have been an independent shop.

During all that time we have encountered many instances where a coordination study was absolutely not a valid option for designating GFI Range value. Typically, the need for a proper GFI Setting was imperative. Every Hit was costing millions.

Once the Ground Fault cause is cleared, it is time to adjust the GF relay to reflect existing Real Life conditions, which as we know, change over time as equipment is added or deleted.

Time is always of the essence, and when the client cannot stand for any Nuisance Tripping, as is very often the result of a GFI set at Minimum, ( so that is not a valid option, unless the actual Max Amp Load calls for 100 Amps or less). And Of course we would not recommend Defeating the GF Protection as an option.

So what is left when the Facility's Max Load is at or above 500 Amps, is the 20% Solution.

The 20% of-Max-Load figure as used to resolve the proper range setting for the Pick-Up value at the GFI Relay was, and still is, our standard application. It has remained so because it works.

It is an actual, viable calculated Range particular and applicable to just about any Facility. It is a Range Setting which virtually eliminates Nuisance Tripping, and also does Not Defeat the GFI Relay.

Admittedly, the 20% value is not an absolute. It does not replace a proper Coordination Study, but it does perform much closer to the mark than does a GFI Relay set randomly at either Minimum or Maximum Range.

And, once safely in the Comfort Zone, if the end client still wishes for a Coordination Study, he / they can now afford to do one on their own timeline..not under emergency conditions.



The reason I posted this thread is that earlier this evening I became concerned as I was reading a < Closed > thread where, IIRC, someone mentioned using 400 Amps as their standard setting for All GFI Pick-Up range settings.

I feel, based on our experience, that the application of this 400 Amp arbitrary range value for GFI Relay settings may not be a good idea; unless of course, the actual max Load at the Device is 2000 Amps.



My advice is Free, please consider it for what it's worth...



Thank you for the welcome, BTW..


Take care.

 

[tkb]

So the loss of millions doesn't justify a few thousand for a coordination study?
Seems silly to me.

 

[kingpb]

So the loss of millions doesn't justify a few thousand for a coordination study?
Seems silly to me.

What is even sillier is that for a few thousand dollars owners are willing to risk lives by not having an arc flash study done.

That, in my opinion, is criminal :rant:

 

[iwire]

I understand what you said. Your concerns are valid. But do rest assured that the 20% figure is not something pulled out of thin air..

Of course it is pulled out of thin air if a study has not been done.


Place yourself in court, answering a lawyers questions.


Did you decide to change the values on the breaker?

Yes.

Who determined the values you set to?

I did.

Did you do a coordination study?

No

Are you an electrical engineer?

No

What qualifications do you have that made you qualified to determine the values?

Well Many years ago, we were employed by one of the two largest Electrical Switchgear manufacturers in the USA as Field Test Engineers.
At that time, the 20% calculated range for GFI range setting was what we were instructed to apply by Lead Project Engineers at said Company.

And exactly how does that have anything to do with the electrical installation that failed?

Well ... err ...

This figure is a GFI Relay range value which has served us very well back then, and for the past 26 years during which we have been an independent shop.

How would you know if the figure is low enough unless there were a number of faults?

During all that time we have encountered many instances where a coordination study was absolutely not a valid option for designating GFI Range value. Typically, the need for a proper GFI Setting was imperative. Every Hit was costing millions.

What do you mean 'It was not a viable option?' It really is not an option, they are required to do so.

Once the Ground Fault cause is cleared, it is time to adjust the GF relay to reflect existing Real Life conditions, which as we know, change over time as equipment is added or deleted.

Yes, which means an updated study.

Time is always of the essence, and when the client cannot stand for any Nuisance Tripping, as is very often the result of a GFI set at Minimum, ( so that is not a valid option, unless the actual Max Amp Load calls for 100 Amps or less). And Of course we would not recommend Defeating the GF Protection as an option.

It sure sounds like you are saying money comes first safety second.

 

[kingpb]

I missed where the time delay is discussed. I also assume this "rule of thumb" pertains to LV systems that are solidly grounded.

I can say that as an engineer, if someone could not give me a fault study to go by, it would be set to the lowest amp setting possible, and the shortest, i.e. instantaneous time. Even then I would say there's is no guarantee that it will protect anything.

 

[Michael4444]

Iwire, tkb and kingpb, I am an Electrical Field Test Engineer, with 30 years' experience Calibrating and Load Testing Circuit Breakers and Relay protection in Low as well as Medium and High voltage applications. I /we have never suffered accidents or experienced lawsuits.

Ok, enough with the Curriculum Vitae.


Gents, given a typical scenario where the Client has Nuisance Tripping, when the luxury of time to have a proper coordination study is Not a Valid option, and where the client desperately insisted to have the GFI defeated by his electrician ( ie: pulling control voltage fuses ), then the 20% solution becomes a Valid Option.

Better that option than leaving the setting on the GFI Relay range on Minimum or Maxed-out at 1200A, or picking any random Amp setting in between just because the value "felt good"; or worst yet, walking away with the knowledge that the Client was planning on defeating the GFI protection altogether.

And at this point I would assume that sometime or another, all of us here have shown up at a Plant and discovered from Field Testing and Inspection that their otherwise perfectly functional GFI Protection had been defeated in the past by some means by plant personnel as a quick response to the GFI Relay causing Nuisance Tripping.. In such a case, it is incumbent upon us to offer as many valid options as possible to arrive at a safe solution in re-activating the GFI protection..

Obviously, leaving the GFI "as is" is NOT one of the solutions. If we leave it so, the client WILL continue to defeat the GFI protection.

And similarly allowing the GFI to remain temporarily in its existing "defeated" condition while the Client awaited days for the results a proper Coordination Study also suffers the same fate.

The 20% solution in this case Is a viable option.

Naturally, the GFI Relay Range change to a 20% solution must be first explained to the Client as a verbal recommendation in full detail, and further, requires Client's written approval prior to any actual re-setting of the GFI Relay's Range setting, as calculated by examination of Client's documented maximum Load Current.


Gents, again, this conservative 20% Solution was arrived at by some very competent people at the Manufacturer; these people being profesional engineers with years of design and actual field experience with GFI design applications. It is a Conservative option arrived at with SAFETY being the Primary Concern, especially when contrasted to the options considered in the highlighted sentence above.


I thought I would share this information to contrast an earlier post ( dating back to 2008 ) which had recommended setting the GFI at a Constant Range (regardless of actual Load), a Range Setting which seemed to be a personal "Guesstimate" at best, and not a range arrived at by calculated results based on Actual Loads.

That post was greeted with approval and gratitude by quite a few here, a fact that alarmed me for its evident lack of consideration in regards to Safety.


I was imparting simply Real World information which we have found to be both proven and helpful to us over the past three decades.



You may want do keep in mind the 20% solution as a point of recommendation to the Client. Or you may choose to not.


It is your choice, ultimately.


Summing up; in the scenario described above, some level of valid GFI Protection Must be re-enabled. Walking away without having established all available options to the Client for his consideration is the real application of the term "Criminal". In my professional opinion, it would be Negligent, at best.

Some of You may feel comfortable doing so, but we most certainly do not.






Have a great day.

 

[ron]

Often the ground fault protection setting is set based on a percentage of the available line-ground or line-line-ground fault current available, not the facility load.

If the building has a 4000A service and the building load is 100A load, it wouldn't make sense to set the GF to 20A. It may trip as the load grows closer to the building potential and nobody pays attention to the low GF setting.

 

[jdsmith]

Gents, given a typical scenario where the Client has Nuisance Tripping, when the luxury of time to have a proper coordination study is Not a Valid option, and where the client desperately insisted to have the GFI defeated by his electrician ( ie: pulling control voltage fuses ), then the 20% solution becomes a Valid Option.
.........

Naturally, the GFI Relay Range change to a 20% solution must be first explained to the Client as a verbal recommendation in full detail, and further, requires Client's written approval prior to any actual re-setting of the GFI Relay's Range setting, as calculated by examination of Client's documented maximum Load Current.


Gents, again, this conservative 20% Solution was arrived at by some very competent people at the Manufacturer; these people being profesional engineers with years of design and actual field experience with GFI design applications. It is a Conservative option arrived at with SAFETY being the Primary Concern, especially when contrasted to the options considered in the highlighted sentence above.

You may want do keep in mind the 20% solution as a point of recommendation to the Client. Or you may choose to not.

If the client has nuisance tripping they need to have a coordination study done - period. By changing one ground relay setting and not examining the settings as a group in relation to the available fault current you may be fixing one nuisance tripping issue and causing miscoordination, which is still a form of nuisance tripping. As a field engineer you should have a relationship with some consulting engineer that you can engage to assist your client with the study. As Ron indicates below, the ground fault settings, and in reality ALL protective relay settings, are based on the available fault current, which cannot be obtained from the load current or Switchgear/switchboard/panelboard ratings or from load data. The ground fault settings throughout the separately derived system should all be examined from the main, feeders in the main panel, down through any devices with adjustable trips to the largest device with a fixed trip. A power engineer can complete a simple, isolated study fairly quickly if you collect a complete set of field data.

I work with field engineers on a regular basis and I respect the experience that you have accumulated over the years. I do seriously question an engineer who would set ground fault settings based on load current without considering available fault current or at least some proxy for available fault current, like transformer size and impedance or something. Load current is an important consideration for protection settings, but the available fault current is crucial.

Often the ground fault protection setting is set based on a percentage of the available line-ground or line-line-ground fault current available, not the facility load.

If the building has a 4000A service and the building load is 100A load, it wouldn't make sense to set the GF to 20A. It may trip as the load grows closer to the building potential and nobody pays attention to the low GF setting.

 

[steve66]

I completely agree that in many cases a coordination study is the only way to go.

But in some cases, I have to play the Devils Advocate. Consider a farily small system, not an emergency system, and not a critical system. Maybe its just "John's Welding shop".

Why would "John" have to have a coordination study if John doesn't want to pay for one? On the occasional trip, Maybe John is just fine with the main service going down, and maybe he likes giving money to electricians everytime they show up to make sure the fault has been cleared, and to reset the main breaker.

After all, I think clients and customers are allowed to be "silly" with their money (as long as its not posing a danger to someone or violating any codes)

Why can't an electrician adjust a GF setting without a study? (Not that I don't appreaciate your trying to send us engineers the buisness ).

If the GF has tripped, and everything checks out, and the setting is obviouslly just set too low, why not tweak it up a little? As long as the settings comply with 230.95, hasn't the code been satisfied?

If the client wants us to pull a number out of the air, 20% seems farily conservative. More likely to cause another false trip than to cause a fireball that gets someone hurt. If a customer said they weren't going to pay for a fault study, I'd at lest consider turning the dial up to 20% and telling them it may trip again.

What happens if I refuse to adjust the breaker, and after I leave, the Welding shop's part time maintenance guy decides to take matters into his own hands and cranks the settings? Someone gets hurt the next time there is a fault. Next thing I know, the lawyers are blaming me for not turning up the settings. Could I be liable for what I didn't do? I have no clue - just throwing it out there for discussion.

 

[iwire]

I completely agree that in many cases a coordination study is the only way to go.

But in some cases, I have to play the Devils Advocate. Consider a farily small system, not an emergency system, and not a critical system. Maybe its just "John's Welding shop".

Why would "John" have to have a coordination study if John doesn't want to pay for one?

Stictly speaking it is not an option.

110.10 Circuit Impedance, Short-Circuit Current Ratings,
and Other Characteristics. The overcurrent protective
devices, the total impedance, the equipment shortcircuit
current ratings, and other characteristics of the
circuit to be protected shall be selected and coordinated to
permit the circuit protective devices used to clear a fault to
do so without extensive damage to the electrical equipment
of the circuit. ........

Why can't an electrician adjust a GF setting without a study?

I guess they could if they were quilfied, IMO pulling a figure out of my rear does not make me qualified. :happyno:

What happens if I refuse to adjust the breaker, and after I leave, the Welding shop's part time maintenance guy decides to take matters into his own hands and cranks the settings? Someone gets hurt the next time there is a fault. Next thing I know, the lawyers are blaming me for not turning up the settings. Could I be liable for what I didn't do? I have no clue - just throwing it out there for discussion.

I have no control or responsibility for what the next guy does, maybe the next guy replaces 125 amp fuses with 200s because that is what he had. We can what if anything to the point we have to stop using electricity all together.

 

[ron]

Stictly speaking it is not an option.

Bob,
I'm not sure what coordinated means as related to GF settings. Of course selectively coordinated is defined in the NEC.

I would say that the only time GF settings need to be selectively coordinated is for hospitals, elevators (if present), and Article 700/701 applications (if present).

 

[iwire]

Bob,
I'm not sure what coordinated means as related to GF settings. Of course selectively coordinated is defined in the NEC.

If the GFP is set too high, will that not lead to 'extensive damage to the electrical equipment' and therefore be a violation of 110.10?

 

[Crionics]

I think if you are not qualified or have the appropraite information to perform a coordination study, set the GF device to the lowest instantaneous PU - I believe manufacturers ship devices this way for a reason. I don't buy your 20% rule. This is dangerous and unethical IMO.

Rob Wolf, PE

 

[jim dungar]

A percentage of load current as part of the decision for a GF setting, how un-scientific can you get? This rationale seems to be 'hundreds of amps of current flowing on the ground' is an acceptable part of life.


In almost all of the GF relay (i.e. main only) adjustments I have recommended, I would not be surprised that the majority dealt with the operating time setting as more critical than the current pickup point.
In reality it is almost impossible to get a 'service entrance' GF to coordinate with any branch device larger than 100A. So the settings typically become more about limiting the amount of damage, not preventing a trip. My normal practice is to try to get the main GF to coordinate with 30A branch devices, I figure this value represents the majority of branch likely to experience 'routine' ground faults.

Determining settings of a single GF device, has no where near the complexity of a full coordination study.

 

[steve66]

If the GFP is set too high, will that not lead to 'extensive damage to the electrical equipment' and therefore be a violation of 110.10?

I think that is a valid point. But 230.95 seems to be more specific. It says GFP shall not be set higher than 1200 amps, and shall not have a delay more than 1 second for currents higher than 3000 amps.

It just seems to me like that rule allows any setting that is less than 1200 amps and 1 sec. Maybe the rule should be changed? It does seem odd to have a single max. setting for all different service sizes and capacities.

I have no control or responsibility for what the next guy does, maybe the next guy replaces 125 amp fuses with 200s because that is what he had. We can what if anything to the point we have to stop using electricity all together.

Probably true in your case. But I know of at least one instance where engineers were considered accountable for what they didn't do, and for what happened when others took matters into their own hands.