Ground Fault

Status
Not open for further replies.

mshields

Senior Member
Location
Boston, MA
Ground fault is not required on a 208Y/120V service. I believe the reasoning behind this has to do with the fact that 120V line to ground cannot sustain an arc? Am I correct on this?

Be that as it may, clearly you can have a ground fault at 120V and it doesn't have to have anything to do with an arc. i.e. somebody ends up with current going through them, the GFI senses more current comming in via the hot than going out via the neutral and you get a ground fault.

I realize I'm probably mixing up my personnel protection ground fault protection with equipment ground fault protection. But in any event, I realize that there are holes in my understanding of this rudimentary subject. Can anyone help?

Thanks,

Mike
 

brian john

Senior Member
Location
Leesburg, VA
Mike:

Yes it has to do with the voltage levels that can sustain an arc, At 120 VAC to ground, arcing faults are not as major issue. Fuses and circuit breakers can and do their function at this lower level, at 277 the impendence of the arc does not allow the current levels to reach high enough levels to operate standard overcurrent protection (OCP) till excessive damage is done. I have seen many 277 arc faults ?blow ups? and the resulting divesting damage, this level of damage is seldom seen in 120/208 VAC.

Most of the 120/208 VAC electrical damage we see are melt downs, usually a loose connection when the insulating material finally fails the OCP operates to clear the fault. The same type of fault on a 277/480 VAC system can make the switchgear disappear.
 

George Stolz

Moderator
Staff member
Location
Windsor, CO NEC: 2017
Occupation
Service Manager
Mike, this is purely equipment protection, as your gut was telling you.

In addition to Brian John's excellent explanation, I'll tack on this Mike Holt graphic. It has more to do with bonding (as indicated by the section number) but the physics are also relevant in regards to the GFPE issue.

1113921936_4.jpg
 

steve66

Senior Member
Location
Illinois
Occupation
Engineer
I wouldn't say 208/120V can't sustain an arc.

The requirement for GFP on 480V equipment came about due to several incidents where a "burn down" occured. (Ie. an arc kept going without tripping the main breaker.)

It might that similar arcs didn't occur on 208V systems, or it might just be that similar arcs on 208V systems did trip the circuit breakers.
 

brian john

Senior Member
Location
Leesburg, VA
Steve:

I took a course quite a while ago the book we utilized, the name slips me but the author was Paul Gill, stated that the peak voltage of a 120 VAC sine wave was not enough to sustain an arcing ground fault. I will substantiate this when I get into the office.
 

davidr43229

Senior Member
Location
Columbus, Oh
The theory of Ground Fault was introduced into the NEC in 1989 215.10 &240.13 at 1,000 amps and more. This was introduced to protect equipment and not personnel. Exception NEC 517.17. A ground fault basically measures power in , vs power out. When power in does not equal power out, it trips the Ground Fault relay in low magnitude ground faults, not high levels. That is the job of the OCPD.
208 Volts can sustain an arc, but usually only in Industrial facilities where the KVA rating of the transformer is 125KA & above. It is the available fault current and total clearing time of the OCPD. and not the voltage that sustains an arc. You usually have more available fault currents in a 480/277 system, due to larger transformers and available fault current.

Now GFCI, while having the same theroy is for the protection of personnel, where standing water "may" reside, ie: kitchen, garage etc...
To quote Mike Holt ...."GFCI Ground-Fault Circuit Interrupter. A device intended for the protection of personnel that functions to de-energize a circuit within an established period of time when a current to ground exceeds the values established for a Class A device. Class A ground-fault circuit interrupters trip when the current to ground has a value in the range of 4 mA to 6 mA. For further information, see UL 943, Standard for Ground-Fault Circuit Interrupters. See Article 100."
 

brian john

Senior Member
Location
Leesburg, VA
David GFP was introduced into the NEC sometimes in the 70's (I believe), and not all ground fault protection is the zero sequence type, (measuring phase current and neutral current). There is ground return where a single CT is on the neutral ground bond.

Lastly why would a 125 kva limit in industrial facilities make a difference in Ground fault?
 

brian john

Senior Member
Location
Leesburg, VA
Maybe we are mixing apples and oranges, Arc flash hazard and arcing ground faults that result in a sustained arcing ground fault that leads to significant damage of the distribution equipment. I am not denying arc flash hazards exist at 120/208, I was stating (I believe and could be wrong) that arcing ground faults are not sustainable at the 120/208 volt level. I am having trouble locating a text book that I (once again) believe stated this and gave the supporting reasoning. I have an open office and someone borrowed this book (I HOPE).


Ground Fault Protection first appeared in the 1971 NEC in 230.95.

Basic types of Ground Fault protection: Zero Sequence (large window CT), Ground Return (single CT on neutral bond) and Residual (4 separate CTs). There may be one or two other methods, but we seldom see them in our area of operation.

Depending on the type of distribution effects the style you utilize. Originally in the 70's the primary style of GFP (we saw) was the ground return method, this was inexpensive to install and was typically comprised of a CT with a 5 amp secondary that operated a dash pot. It is my understanding that the large window type CT's had accuracy issues and took time to refine. The other method utilized sometimes was Protective relaying similar to that employed on high voltage systems, though this was more expensive that the Ground return method. The zero sequence style is predominately utilized on single feed services in our area.



I am not an engineer and am CAREFULLY posting this, I ran across this while trying to locate information regarding arcing ground faults. I state CAREFULLY because these are MINIMUM values. Unless someone else has any data I will continue looking.


From the IEEE Gray Book 241-1974

Page 214 partial of Table 53

Approximate Minimum Value of Arcing-Fault Currents in Per Units of Bolted Faults


Type of Fault
Nominal System Voltage

Single Phase Line to Ground Fault 570 VAC- 0.40 480 VAC-0.38 208 VAC-0


Lastly you state transformers at 125 KV, I thought we were discussing 120/208 VAC and 277/480 VAC systems. Ground fault protective relaying has been in use on higher volage systems (above 1000 VAC) for quite a while.
 
Last edited:

steve66

Senior Member
Location
Illinois
Occupation
Engineer
Again, I'm not sure if 208V can sustain a arc, but I still haven't seen anything yet that rules it out.

I would expect it to depend a lot on the distance of the arc, and the size of the distribution transformer (along with the available bolted fault current).

Like David mentioned, arc flashes are sometimes considered with distribution transformers larger than 125KVA, and often ignored below 125KVA. So I think he is just saying that is another variable you have to consider before saying "208V can't sustain an arc fault."

Also, most arc flash calculations are done inside a piece of gear. They assume the arc has to span about a inch or more between the busses. I think it would be much easier for 208V to sustain an arc between two pieces of wire in a feeder where the separation is nill.
 

davidr43229

Senior Member
Location
Columbus, Oh
Brian John,
How did I get my oranges and apples mixed up, in terms of when Ground Fault was introduced in the NEC.(joking)
I know when we do Arc Flash studies the level is mostly "0" with transformers of 120/208 below 125KVA ( I forgot to put the A in there, the last time) Transformers greater than 125KVA can have 30,000 amps depending upon the Z. Couple that with the total clearing time 6-12-30 cycles depending on where the Ground fault is set "In Industrial Applications" of the Main, this could represent quite an Arc Flash Fazard of Class 2* at 18".
Sorry for the miscommunication. I'll do better next time (honest).
 

brian john

Senior Member
Location
Leesburg, VA
David what I was saying was, you are talking about available fault current (I think) Arcing ground faults are low current faults due to the impedeance of the ground system and the resultiing arc. Hence the lower current setting of the relays (100-1200 amps normally) and time settings of Instantaneous to 1.0 seconds trip.

Arc flash is real and present I agree, what I was discussing was a sustained arcing fault that vaporizes the equipment. I have been on 100's (maybe) of these jobs over the years all 277/480. The damage is amazing, I have never seen this level or anywherer near this level of damage on a 120/208 VAC distribution system.

As for gorgetting the "A" lets hope none of us (well at least me) are judged on spelling or typing skills.


Check this site

http://expert.ecmweb.com/resistance_grounding_experts/2006/04/03/why-the-1000-amp-limit-for-gfcp/
 
Last edited:

brian john

Senior Member
Location
Leesburg, VA
It took me a while to locate my source but I finally located the book. I was surfing the web for literature and saw a book for sale and remembered I had this book in my library.


?Overcurrents and Undercurrents All About GFCIs?.
Author Earl W. Roberts,
1996, Mystic Publications

This book also covers GFPE, from chapter 12 (my abridged version)

While arcing faults can occur on 120/208 VAC systems the arc does not result in a restrike arc due to the voltage level. Studies have shown that to have a restriking arcing ground fault the voltage levels of 374 volts are necessary.

120X1.414 = 169.68, Peak Voltage
277X1.414 = 391.67. Peak Voltage

GFPE is required on 277/480 VAC distribution systems to protect against the devastating damage that occurs from these sustained restriking arcing ground faults, it is not required on 120/208 distribution systems as the restriking high impedance burn down arcing ground faults are not occurring.

What tweaked me to research the source of the information was Mike Holt made the same basic statement in his seminar that I have made here. I asked him for his source, While he could not state the source he told me to email him and he would research his source.
 

brian john

Senior Member
Location
Leesburg, VA
A perfect example is the burn down associated with wet bus duct. The fuses, do not operate as the fault current is limited by the impedance of the arc (NOT A BOLTED FAUILT). The GFPE is disconnected due to past so called nusiance trips, and the restriking arcing fault burns the bus duct up, this happens quickly.

I have seen switchgear melted to the ground and fuses were intact.

This was from water was on the load bus, the switch had opened from ground fault and someone choose to attempt to reclose the switch. I do not remember if the fuses open or not.


Click on picture above to see full-sized image.
 
Last edited by a moderator:
Status
Not open for further replies.
Top