It is currently Fri Aug 07, 2020 11:02 pm



Post new topic Reply to topic Go to page 1, 2  Next
Author Message
ekstra   ara
 Post subject: Low arcing current?
PostPosted: Tue Jan 29, 2019 2:44 pm 

Joined: Tue May 29, 2018 8:19 am
Posts: 29
I was involved in an arc flash incident where I am questioning the minimum likely arcing current.

An arcing fault initiated in a fusible switch in a distribution section of a 480-volt/3-phase/4-wire switchboard. Arcing apparently spread to multiple locations throughout the distribution section and settled to the bottom of the vertical bus in that section. The ends of all three bus bars were significantly eroded and the adjacent neutral bus sustained significant melting.

The switchboard was served directly from a pad-mounted 2,500 kVA utility transformer. The overcurrent protection consisted of a service entrance main bolted-pressure switch with 3000-amp class L fuses and ground fault relay, and a downstream power circuit breaker with 1600-amp trip serving the distribution section. None of these devices operated. An electrician apparently had sufficient time to attempt to manually open the main switch while the arcing was present. The main switch was found partially open and also was severely burned. The distribution section where the arc initiated was severely damaged throughout.

My question is obvious: why did none of the overcurrent devices operate? The breaker was tested and found to be good. The fuses were intact and the ground fault relay was burned but did not trip. The calculated fault current was about 51kA symmetrical. The end of the burned bus bars was about 50mm to the neutral. IEEE 1584 equations give an arcing current of about 20 kA for this spacing. At 20kA, the 3,000A fuses should have cleared in about 1.8 sec and the 1600A breaker in 0.15 sec. Yet the arc apparently persisted many minutes. With the settings that were in place on the breaker, a one minute trip time would imply an arc current < 4,000A. This is the second time in my career I’ve been involved with an apparently long duration arcing fault. How is this possible? Are the equations in 1584 that far off of reality or are there other circumstances at play?
Thanks for any thoughts on this.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Tue Jan 29, 2019 3:31 pm 
Plasma Level
User avatar

Joined: Wed May 07, 2008 5:00 pm
Posts: 822
Location: Rutland, VT
You are basing your trip times on infinite bus fault current. It is highly unlikely that you are close to that. You need to get the available fault current from the utility and then figure out what the fault current was. Let's assume that your actual fault current is 50% of the infinite bus fault current. Using that would put the arcing fault current in the 3-4kA range. Where is that on the time current curve? Probably in a location that would yield very long trip times. And if the available fault current is 25% of infinite bus, it is possible the arcing current would be low to maybe not even be on the time current curve.

This is one item that would be clear if an actual study was done at your facility. Protective device settings could be optimized for both coordination and/or incident energy.

BTW...I hope the electrician operating the switch was wearing 40 cal/cm^2 PPE

_________________
Barry Donovan, P.E.
www.workplacesafetysolutions.com


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Tue Jan 29, 2019 3:43 pm 

Joined: Tue May 29, 2018 8:19 am
Posts: 29
Thanks for the quick response.
We did a study (post-incident) and the 51kA was calculated at the fault location based on a utility-provided value of about 42kA. Motor contribution makes up the rest. The calculations were done with SKM Powertools.

The outcome for the electrician was unfortunate.

_________________
Alan Lehman


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Tue Jan 29, 2019 3:57 pm 
Plasma Level
User avatar

Joined: Wed May 07, 2008 5:00 pm
Posts: 822
Location: Rutland, VT
I would suspect that the utility gave you the infinite bus at the secondary transformer terminals. This is usually what you will get unless you are very specific in your request for the available short circuit current. I start at riser pole and model in.

If you want, pm me with the info received from the utility. Can probably tell from that what they sent you.

_________________
Barry Donovan, P.E.
www.workplacesafetysolutions.com


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Wed Jan 30, 2019 9:32 am 
Arc Level

Joined: Tue Jan 13, 2009 5:00 pm
Posts: 532
wbd wrote:
I would suspect that the utility gave you the infinite bus at the secondary transformer terminals. This is usually what you will get unless you are very specific in your request for the available short circuit current. I start at riser pole and model in.


Did you mean primary perhaps? If someone simply asks me for available fault current, they'll get a number that has zero source impedance ahead of the transformer, and a low transformer impedance. Equipment sized this way should continue to work through system upgrades, transformer change-outs, etc. Not what you want for IE calculations at all. You need to state when you need the info for arc flash, and ask to speak with someone in engineering if they persist in giving you a tabulated value.

Very sorry to hear the outcome was unfortunate.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Wed Jan 30, 2019 12:36 pm 

Joined: Tue May 29, 2018 8:19 am
Posts: 29
Yes, the fault current is probably based on "infinite primary". In my experience most utilities have a table by transformer size. I haven't had much luck getting anything other than that, even when I say that I need an accurate value for an arc flash study. I'll try asking about the primary fault current, but then I need to know the transformer impedance which means getting someone to open it up to read the nameplate. Do you normally try to get an actual number for the transformer or assume something?

_________________
Alan Lehman


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Wed Jan 30, 2019 12:45 pm 

Joined: Tue May 29, 2018 8:19 am
Posts: 29
... and this is a big fallacy in many arc flash calculations. The utility values are frequently not accurate and tend to err on the high side because they assume that is the conservative approach. They do this because (a) it's troublesome and time consuming to calculate real numbers, (b) their systems change frequently, (c) fault values depend on the distribution switching configuration, (d) historically engineers have only cared about equipment ratings so giving high values was the safe approach.

One utility engineer told me he always looks up the table value for transformers one or two sizes larger than installed in case they need to change it out later.

_________________
Alan Lehman


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Wed Jan 30, 2019 1:41 pm 
Plasma Level
User avatar

Joined: Wed May 07, 2008 5:00 pm
Posts: 822
Location: Rutland, VT
Early on I would have difficulties with getting the necessary available fault current and X/R from utilities but that is less of an issue now. Usually the first answer you get is from a customer service rep and based on a table of transformer sizes, voltage and impedance. This is the infinite bus which is usually for sizing equipment.

You have to be persistent with the customer service person and eventually they will connect you to engineering or past the work request to engineering. Use of the infinite bus fault current is well known not to be reliable for arc flash. I will get the riser fuse, riser cable, transformer nameplate data, bayonet fuse (if any) to model to the service entrance equipment. This is the way it should be done.

_________________
Barry Donovan, P.E.
www.workplacesafetysolutions.com


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Thu Jan 31, 2019 12:26 am 

Joined: Wed Dec 12, 2018 12:59 am
Posts: 3
Location: Germany
What we usually do if there is no way to get the short-circuit values from the utility is to at least take in the max rated short-circuit current of the MV-switchgear plus the dampening from the cables running to the transformer and from the transformer to the LV main switchgear. Of-the-shelve MV switchgear usually has 16/20/25/31.5 kA rated short-circuit current for 10/20 kV.
Also with motor contribution you have to be careful. The part coming from downstream of your switchgear that you are calculating for does not flow through the upstream breaker and thus does not contribute to the tripping (I hope that makes sense). I would probably also consider looking at both cases (with and without motor contribution) and see which of the two produces higher incident energies because of the different tripping times.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Thu Jan 31, 2019 9:30 am 

Joined: Tue May 29, 2018 8:19 am
Posts: 29
Thanks both of you. I wonder if a valid approach where we can't get meaningful fault values would be to run a couple of cases or a range of values and take the worst case.

I re-ran the calculations with the utility fault current reduced to 50% of the value I was given. This resulted in a calculated bolted fault current of 31.6 kA and "protective device arcing fault current" of 10.8 kA. The maximum breaker clearing time with settings that were in place at the time for this current is 16 seconds and the fuse clearing time is 45 sec.

_________________
Alan Lehman


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Thu Jan 31, 2019 1:16 pm 
Plasma Level
User avatar

Joined: Wed May 07, 2008 5:00 pm
Posts: 822
Location: Rutland, VT
Are you using the IEEE 1584-2018 or IEEE 1584-2002 equations to determine your arcing current?

_________________
Barry Donovan, P.E.
www.workplacesafetysolutions.com


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Thu Jan 31, 2019 1:19 pm 

Joined: Tue May 29, 2018 8:19 am
Posts: 29
2002

_________________
Alan Lehman


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Thu Jan 31, 2019 2:33 pm 
Plasma Level
User avatar

Joined: Wed May 07, 2008 5:00 pm
Posts: 822
Location: Rutland, VT
IEEE 1584-2002 has been superseded by IEEE 1584-2018. I think you should be using the 2018 equations.

_________________
Barry Donovan, P.E.
www.workplacesafetysolutions.com


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Thu Jan 31, 2019 3:57 pm 

Joined: Tue May 29, 2018 8:19 am
Posts: 29
Isn't 2018 still designated as "draft" status? 2018 produces somewhat higher arcing fault values in SKM for this system.
I ran this particular system with four scenarios and here are "protective device arcing current" values reported by SKM:

Infinite primary (42kA): 2002: 23.3 kA; 2018: 24.4 kA
50% (21kA): 2002: 11.7 kA; 2018: 18.2 kA

_________________
Alan Lehman


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Thu Jan 31, 2019 4:05 pm 
Plasma Level
User avatar

Joined: Wed May 07, 2008 5:00 pm
Posts: 822
Location: Rutland, VT
IEEE 1584-2018 was released at the end of November 2018 so is not in draft status anymore. The release was announced on this forum by Jim Phillips.

can you send me a one line? Also have you gone back to the utility to get the information I mentioned previously?

_________________
Barry Donovan, P.E.
www.workplacesafetysolutions.com


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Sun Feb 03, 2019 11:06 am 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 2174
Location: North Carolina
Keep in mind here that IEEE 1584-2002 and -2018 assumes that arcs are self-sustaining out to 2 seconds (for the general case) and coordination calculations assume that both bolted and arcing faults are self-sustaining in perpetuity without regard for consumption of bus bars or other effects that would naturally extinguish the arc.

Obviously the assumption of a self-sustaining arc is a conservative approach but I'm not sure if many people are aware of just how quickly an arc can actually self extinguish. By way of example the recently published paper "Arc Flash Hazards of 125 Vdc Battery Systems" had maximum arc durations ranging from 0.166 to 0.202 seconds with arc currents ranging from 1825 to 3808 A and no the shortest time was not due to the highest current. Most tests involved a 1/4" bus gap that gradually opened up to 3/4" as the bus bar eroded before the arc self extinguished. Granted the paper was dealing with DC arcs and your scenario is dealing with AC arcs so I would expect the bus bar erosion rate to be significantly slower but in any event it would not surprise me in the least given the scenario with extremely large fuses and breakers that the bus would be eaten away and extinguish the arc well before the protective devices ever tripped. Certainly a 2 second time interval seems fairly reasonable and would result in hitting the recommended time limit in IEEE 1584-2002/2018.

Based on your description alone with a 480 V arc I'm guessing it would exceed 40-50 cal/cm2 at the bare minimum for a couple reasons. Trying to maintain coordination amongst large distribution breakers and/or fusing while providing arc flash protection is an exercise in futility above 1500 kVA. It might be possible but it's not easy. If you have instantaneous protection turned on, you will automatically miscoordinate with downstream device-level protective devices unless you have an adjustable delay making it a definite time device. Traditionally devices are spaced at about 0.350 seconds between layers so by the time you reach the distribution level at 1600 A you have at least 1 or 2 layers below that level so trip times are out to close to or exceeding 1 second...not very conducive to doing much of anything about an arc flash. Modern installations lately have tended towards making instantaneous protection user-selectable, the so-called "maintenance switch" which is very effective at solving this very real problem at very low cost, much more so than the very costly solutions of 87 relaying, ZSI, or arc flash relays.

So I would not be surprised if nothing tripped and I would not be surprised if wearing even multilayer flash suits didn't do any good at all. That's just my experience. I've seen plenty of cases of something very similar to what you described in the mining and chemical industry. The only difference is that most of the time I was tasked to repair damaged equipment and people weren't in the line of fire.

In saying that I just recently replaced a bunch of switchgear for a customer. The specification was to maintain it in original state which consisted of a 3000 A class L fused switch and a bunch of large distribution breakers in a switchboard but fed by a 1500 kVA transformer. I convinced them to at least go to an all breaker option and added a "maintenance switch" to comply with NEC 2017. I used a 3000 A frame breaker as per their requirements but set it down to 1200 A long term trip settings and by the way they only have about 300 A of connected load so no issues from doing this. The maintenance switch then engages a 2400 A instantaneous trip setting (200% of pickup) but this is adjustable if needed. They don't have an arc flash study completed either but with this configuration I'm reasonable certain the incident energy is down in the teens or less based on past experience, stiff utility bus or not.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Mon Feb 04, 2019 5:42 pm 
Arc Level

Joined: Tue Jan 13, 2009 5:00 pm
Posts: 532
Thanks both of you. I wonder if a valid approach where we can't get meaningful fault values would be to run a couple of cases or a range of values and take the worst case.

I re-ran the calculations with the utility fault current reduced to 50% of the value I was given. This resulted in a calculated bolted fault current of 31.6 kA and "protective device arcing fault current" of 10.8 kA. The maximum breaker clearing time with settings that were in place at the time for this current is 16 seconds and the fuse clearing time is 45 sec.


So when will the software companies take your normal inputs along with a range of fault currents, and provide an IE versus fault current curve that shows at a glance what and where the highest IE is?


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Mon Feb 04, 2019 7:45 pm 
Sparks Level

Joined: Sat Nov 21, 2009 10:23 pm
Posts: 110
Location: Ohio
Yes, the fault current is probably based on "infinite primary". In my experience most utilities have a table by transformer size. I haven't had much luck getting anything other than that, even when I say that I need an accurate value for an arc flash study. I'll try asking about the primary fault current, but then I need to know the transformer impedance which means getting someone to open it up to read the nameplate. Do you normally try to get an actual number for the transformer or assume something?


Just an opinion, however, I will not do a study without the transformer impedance taken from the the nam eplate. There is usually a way to get the nameplate data. On the flip-side you can usually do what-ifs and get a conservative calculation without an accurate utility source impedance, you just have to be aware of what they submit and how it compares to reality.

Another utility issue is them giving the secondary short circuit current, that is totally unacceptable. That means they can be guessing at both the primary impedance and the transformer impedance.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Wed Feb 06, 2019 3:31 pm 

Joined: Tue May 29, 2018 8:19 am
Posts: 29
wbd wrote:
IEEE 1584-2018 was released at the end of November 2018 so is not in draft status anymore. The release was announced on this forum by Jim Phillips.

can you send me a one line? Also have you gone back to the utility to get the information I mentioned previously?


I see his announcement. Xplore was still giving me the draft version when I do a search. I'm waiting to hear back from the utility regarding real fault current values. I don't have a lot of hope but will wait and see what I can get. I'll post back here when I have something.

_________________
Alan Lehman


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Low arcing current?
PostPosted: Wed Feb 06, 2019 3:39 pm 

Joined: Tue May 29, 2018 8:19 am
Posts: 29
stevenal wrote:
So when will the software companies take your normal inputs along with a range of fault currents, and provide an IE versus fault current curve that shows at a glance what and where the highest IE is?

This paper was published in IEEE IAS Transactions last February about a graphical approach to find conservative IE values that sounds somewhat like what you are suggesting. SKM said they would look at it.

_________________
Alan Lehman


Top
 Profile Send private message  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 40 posts ]  Go to page 1, 2  Next

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