It is currently Thu Dec 03, 2020 10:08 pm



Post new topic Reply to topic
Author Message
ekstra   ara
 Post subject: Ground Fault Protection
PostPosted: Fri Aug 02, 2013 9:39 am 
Sparks Level

Joined: Thu Jul 19, 2007 7:54 am
Posts: 201
Location: St. Louis, MO
Ground fault protection was mandated by the NFPA several years ago. At the time, it was specifically mentioned that this protection would help to protect against phase to ground arcing faults and burndowns.

What I am looking for is a comparison of a fault with and without ground fault protection. Has anyone seen such a paper or video?
We all know it works, but has it been quantified?


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Aug 05, 2013 8:17 am 

Joined: Wed Dec 02, 2009 4:46 pm
Posts: 29
Location: CT
WDeanN wrote:
Ground fault protection was mandated by the NFPA several years ago. At the time, it was specifically mentioned that this protection would help to protect against phase to ground arcing faults and burndowns.

What I am looking for is a comparison of a fault with and without ground fault protection. Has anyone seen such a paper or video?
We all know it works, but has it been quantified?

My understanding is that this requirement arose from observations of low level arcing faults that became increasingly common in the 60s after many solidly grounded 480/277V systems were installed in the late 50's and forward. Prior to that floating, or impedance grounded 480V systems were more common.
If the ground fault is not low level, i.e. it becomes a single phase to ground arcing fault at th emaximum possible current then Ground fault Protection is not, probably, what will open the overcurrent device. It is probably that short time or instantaneous protection will be faste, or similar in response. The GF detection only targets low level high impedance faults that may not be deteced by other means due to their low level or intermittency.
Papers on this were published in IEEE Journals in the 1960s. NEMA also publishes an application guide on ground fault protection of LV systems with some analysis of copper consumption rates (NEMA PB 2.2) that is, I believe, available at not cost from the NEMA website.
I would point out that if designing a GF protection system it should be kept in mind that single pole CBs are, inherently ground fault overcurrent devices and that ground fault protection can be difficult to coordinate with downstream overcurrent protection. Excessive use of multiple layers of GF devices can "decrease" overall system selectivity.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Aug 19, 2013 8:08 am 
Sparks Level

Joined: Thu Jul 19, 2007 7:54 am
Posts: 201
Location: St. Louis, MO
Marcelo wrote:
My understanding is that this requirement arose from observations of low level arcing faults that became increasingly common in the 60s after many solidly grounded 480/277V systems were installed in the late 50's and forward. Prior to that floating, or impedance grounded 480V systems were more common. ...
Papers on this were published in IEEE Journals in the 1960s. ...

I've seen several of the papers. They agree that equipment damage may be prevented or made less with ground fault protection. What I am really looking for is a quantitive comparison of with vs. without ground fault protection.

Marcelo wrote:
If the ground fault is not low level, i.e. it becomes a single phase to ground arcing fault at th emaximum possible current then Ground fault Protection is not, probably, what will open the overcurrent device. It is probably that short time or instantaneous protection will be faste, or similar in response. The GF detection only targets low level high impedance faults that may not be deteced by other means due to their low level or intermittency.

I've heard this also. I understand that the short time or instantaneous will be as fast, but why is the ground fault protection ignored in these cases by most analysis? I've even heard that ground fault protection will offer NO protection for an arcing fault. I find this hard to believe. I have investigated one case where the arc fault was cleared by ground fault protection. The relay record clearly showed it picking up, dropping out, then picking up again and finally tripping on ground fault. It picked up and dropped out once on phase protection as well.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Wed Aug 21, 2013 6:51 am 

Joined: Mon Nov 19, 2007 5:25 am
Posts: 30
Location: Titusville, Fl.
Not sure of a study or video, but from a common sense approach, the 1000A rule from NEC (Arts 215.10 & 230.95) pertaining to application of Ground Fault (GF) protection – makes good sense. In most cases operational errors (human error) result in a ground fault event. So GF protection at even less current is a good practice. Most transformer arrangements are Delta- Wye configured, so the thru phase fault from the secondary may rely on the primary protection (cross your fingers it’s not that upsized…..), but not the case with ground fault, hence a reason for GF protection downstream of the secondary. Although the above mentioned NEC rule applies to equipment operating at 600VAC, we’ve engaged the ground fault instantaneous protection at the Substation Feeder CB’s at the Substation for feeder protection. This came into play as a young laborer assigned to demolish MV cabling in a man hole cut into a live cable. You can challenge the positive identification / verification of isolation method pertaining to LOTO, the procedure used, the unqualified lineman (laborer) performing the work, and the level of PPE donned – all of which I will admit was deficient! However due to the GF protection, this young laborer was saved with minor injury as quantified by the NFPA70E – 1.2 cal/cm^2 rule. Firm believer in GF protection -indeed.
I agree, with what was mentioned above pertaining to coordination of the GF schemes and remember to properly apply the associated CT hardware on the equipment bus - and you should be OK w/r to nuisance tipping…
FWIW


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Wed Aug 21, 2013 3:18 pm 
Arc Level

Joined: Mon Jan 18, 2010 11:35 am
Posts: 524
Location: Wisconsin
Most 480Y/277V, GF relays have delays of at least .1sec (not counting pickup time of a 120VAC shunt trip coil on fusible units) this makes them relatively slow devices in the world of arc flash protection.

Yes a GF device will reduce the probability of an arcing L-G fault propagating into a 3-phase arcing fault (the risk), but it provides very little to no improvement in the arc fault incident energy level (the hazard).

For MV systems with electronic relays, the GF function may offer some relief. Residual and zero-sequence schemes need to be considered.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Thu Aug 22, 2013 11:11 am 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 2174
Location: North Carolina
NEMA PB 2 talks about this quite a bit and is freely available. Several IEEE papers written in the 1950's and 1960's (and referenced in NEMA PB 2) document this as well.

Ground fault protection is a giant can of worms because there are several ways to do it. So let's be clear what NEC (NFPA) was after. This is when you have a solidly grounded wye system. Although not the same current as a bolted fault, the currents are still substantial, equal to the bolted fault current divided by 1.732 if we ignore any additional impedances. With small loads typically seen in commercial/residential settings, this is no problem because the current is still plenty high to trip an overcurrent protection device. However as the current increases it becomes difficult or impossible to detect and trip out a ground fault especially if we assume the presence of even a few ohms of resistance in the ground path.

Still, and NEMA PB 2 should be pretty enlightening, the result is heavy ground fault currents and tripping is done simply to minimize equipment destruction.

The opposite scheme, NO grounding, is also available at higher voltages. In this scheme in a non-arcing ground fault damage is minimal but finding the ground fault can be problematic. If a second ground fault occurs without removing the first one, you get a L-G-L fault which is almost as destructive as the solidly grounded fault but there is no overcurrent protection at all. In an arcing ground fault, the system capacitance acts like a voltage multiplier and the arc usually sets off discharges inside the weakest insulation (motors) and rapidly destroys them. The result is roughly 400% reduction in motor life, with the benefit that with diligent maintenance ground faults do not trip. There are some "pulsing" systems which claim to make it possible to trace the fault but I've tried and found that they are not reliable in practical use.

Splitting the middle are resistance grounding systems. These are split into either high or low resistance systems. These use the same wiring scheme as a solidly grounded delta-wye except a resistor is inserted in between the neutral and ground. They are usually rated for either 15-25 A (high resistance) or 100-400 A (low resistance). Since the current is within normal operating conditions, again damage is almost zero. However there are more options with regards to alarming and/or tripping. Putting a CT on the resistor wiring or a PT on the resistor itself allows for tripping on ground fault measured directly. A core balanced CT (residual Earth CT, and various other names) that surrounds all 3 phase legs can also be used but can be used at ANY location within the power system. By setting the time delays on the ground fault protection, selectivity can also be done so that the fault is localized. In addition it is possible (and practical) to perform "alarm-only" as long as with ungrounded systems there is a diligent effort to find and remove the fault if a continuously rated resistor is used, although again the fault can be localized which is not practically possible with ungrounded systems. This is impossible to do with ungrounded systems. Also because the system capacitance is continuously discharged through the resistor, no "voltage multiplier" effect forms and motors are not damaged. About the only disadvantage remaining with this scheme (and ungrounded) is that everything must be rated for full line-to-line voltage.

So...if you are already going to have to spend the money to put in ground fault protection anyway, spending a little more money (<$2500 for a 480 V system, although it gets more expensive as voltage increases), there is almost no reason not to go all the way to a resistance grounded system. The additional cost is nominal compared to the advantages over a solidly grounded system with ground fault protection. Ungrounded systems are the cheapest form of "grounding" in terms of installed cost but are very costly in terms of long term costs to maintain the system.


Top
 Profile Send private message  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 6 posts ] 

All times are UTC - 7 hours


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
© 2019 Arcflash Forum / Brainfiller, Inc. | P.O. Box 12024 | Scottsdale, AZ 85267 USA | 800-874-8883