It is currently Thu Feb 22, 2018 3:35 am



Post new topic Reply to topic
Author Message
ekstra   ara
 Post subject: Isolated protective device - selection in an arcflash
PostPosted: Mon May 20, 2013 3:51 am 
Offline

Joined: Mon Oct 08, 2012 2:59 am
Posts: 24
Location: India
A fault on a bus is also a fault on the main Protective Device (PD) if the PD is not isolated. In such case, the main PD cannot be a source PD to clear an arc.
Please throw some light on the above statement. I mean, is the above statement correct? Is it mentioned in any standard?
I tried to simulate an arc flash on a bus defined as MCC in ETAP 12.0.0. ETAP is [SIZE=15px]neglecting[/size] the nearest protective device. It is selecting the upstream breaker of the nearest PD. Please comment.
Thanks in an advance


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon May 20, 2013 8:01 am 
Offline
Plasma Level
User avatar

Joined: Mon Sep 17, 2007 5:00 pm
Posts: 1245
Location: Scottsdale, Arizona
What you are describing is the current best practice when performing an arc flash study. A couple of problems could result if you base the arc flash duration on the operating time of the main device.

[INDENT=1]1) What if the arc flash begins on the line side of the main (terminals)[/INDENT]
[INDENT=1]2) What if the arc flash originates below the main and then escalates (within a few cycles) to a location on the line side of the main. [/INDENT]
[INDENT=1] [/INDENT]
Although number 2, doesn’t always happen, it can happen. The problem is we really don’t have any way of knowing with certainty whether an arc flash can propagate to the main so the conservative interpretation is to assume that it does and determine the duration by the clearing time of the next device upstream.

A few exceptions to this approach may include:
  • Switchgear - It may be unlikely for an arc flash in one cell/cubicle to escalate to the main cell/cubicle.
  • MCCs - This interpretation is sometimes used with motor control centers depending on the design.
The use of either of these possible exception are typically based on the judgment of the person performing the study. There is not a black and white interpretation of either.

As far as this being in a standards, IEEE 1584-2002 B.1.2 Page 76:

[INDENT=1]Note that if an arcing fault can be initiated on the line side of a main protective device in an enclosure, that protective device should not be utilized for the calculations, instead, the upstream protective device should be used. That is because only an upstream protective device can be considered to provide protection for an arcing fault on the line side of the main protective device in a downstream enclosure.[/INDENT]

I hope it helps.

_________________
Jim Phillips, P.E.
Brainfiller.com


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon May 20, 2013 9:19 am 
Offline
Sparks Level
User avatar

Joined: Tue Oct 26, 2010 5:00 pm
Posts: 115
Location: Maple Valley, WA.
Jim is correct. The one possible exception would be if the equipment is switchgear construction (versus Switchboard, MCC, panelboard...etc.). With Switchgear construction, the breakers are enclosed in cubicles. Arc flashes on the load side of main breaker or main bus will most like/y be protected by the main breaker.

However, we select the option to calculate the line side arc flash energy of the main breaker and put that label on the Main Breaker cubicle door. Usually, the main bus will have a lower AF energy level due to the Main Breaker tripping before the upstream device.

_________________
Robert Fuhr, P.E.; P.Eng.
PowerStudies


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Tue May 21, 2013 4:08 am 
Offline

Joined: Mon Oct 08, 2012 2:59 am
Posts: 24
Location: India
Thanks Jim, that helps.
I am precisely interested in the second point "2) What if the arc flash originates below the main and then escalates (within a few cycles) to a location on the line side of the main.".
If the arc flash happens below the main protective device(it still did not escalate to the line side), is it not the responsibility of the immediately upstream breaker to take care of it first? or is the immediate breaker be ignored for arc flash incident energy calculation?
Or is it that the commercial software (ETAP 12) is considering the worst case (the fault has escalated to the mains!) and ignoring the immediate breaker, choosing the next upstream breaker to calculate the incident energy!
Thanks in an advance.
P.S. I have simulated the same in SKM demo(limited to 15 buses, I used a simple two bus system. I defined one bus a MCC) . To my surprise the immediate breaker tripped as I would have expected naturally. Now which one is correct technically? and what is the explanation of the other?


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Tue May 21, 2013 9:02 am 
Offline
Plasma Level
User avatar

Joined: Mon Sep 17, 2007 5:00 pm
Posts: 1245
Location: Scottsdale, Arizona
Girish wrote:
Thanks Jim, that helps.
I am precisely interested in the second point "2) What if the arc flash originates below the main and then escalates (within a few cycles) to a location on the line side of the main.".
If the arc flash happens below the main protective device(it still did not escalate to the line side), is it not the responsibility of the immediately upstream breaker to take care of it first? or is the immediate breaker be ignored for arc flash incident energy calculation?
Or is it that the commercial software (ETAP 12) is considering the worst case (the fault has escalated to the mains!) and ignoring the immediate breaker, choosing the next upstream breaker to calculate the incident energy!
Thanks in an advance.
P.S. I have simulated the same in SKM demo(limited to 15 buses, I used a simple two bus system. I defined one bus a MCC) . To my surprise the immediate breaker tripped as I would have expected naturally. Now which one is correct technically? and what is the explanation of the other?


I think you nailed it with your summary. It is "all of the above" The main should trip if an arc flash / fault occurs downstream from it.

The concern is if the conducting plasma is large enough, it could reach the line side of the main and even if the main trips, the arc flash has now propagated ahead of the main.

The real problem is we just don't have any way of knowing if this would happen with any certainty. It depends on the equipment design, the location of arc flash, speed of breakers, size of enclosure, gap distance, magnitude of fault current etc. etc.

From my lab experiences, often the arc flash is just a big bang and the main trips. However, sometimes it can go ahead of the main and then it is much worse.

Since there is no way to tell for certain whether the main will end the event, most people take the conservative route (software vendors included) and assume the duration is defined by the next device upstream. About the only exceptions that I see taken are for switchgear and motor control centers depending on their design.

Great discussion!

_________________
Jim Phillips, P.E.
Brainfiller.com


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Wed May 22, 2013 3:09 am 
Offline

Joined: Mon Oct 08, 2012 2:59 am
Posts: 24
Location: India
Thanks Jim.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Wed May 22, 2013 3:01 pm 
Offline
Sparks Level

Joined: Mon Jun 07, 2010 9:40 am
Posts: 119
P.S. I have simulated the same in SKM demo(limited to 15 buses, I used a simple two bus system. I defined one bus a MCC) . To my surprise the immediate breaker tripped as I would have expected naturally. Now which one is correct technically? and what is the explanation of the other?[/quote]

They BOTH are correct, under the assumption that each makes.
Esentially the different result is a matter of the options you have selected in SKM. You can set SKM to give you the results from the Line side of a directly connected main device too.
Another factor thatcomes in to play (potentially) is miscoordination. SKM will look at upstream breakers to see which clears first. It all depends on you you set your options in SKM. I am sure there are probably different options in ETAp too, that probably just have different defaults.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon May 27, 2013 6:14 pm 
Offline
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 1878
Location: North Carolina
If the fault is in the main breaker then you'd simulate a line side event anyway. However in other sections of an MCC if we get a bus fault, the arc is magnetically propelled away from the power source (simple physjcs), until it reaches either an obstruction which allows the plasma to pool and jet out such as at fiberboard barrriers at that point or the end of the bus bars where the same thing happens. This seems to be a common scenario. The marks on the bus bars where the arc travels that look like little points (arc tracks) are not always visible without a microscope. It throws folks off that assume that the arcing fault remains stationary as it does in failures within components other than overhead lines and bus bars. However I have never seen arcs travels towards the power source nor have I seen evidence of conductive plasma clouds which would leave thin layers of metalization on everything after an arcing fault, rendering even nonconductive components highly conductive. Is there a forensic analysis of the case described somewhere? I can give lots of examples of arc propulsion, youtube videos of arc travelling down power lines, arc track pictures, etc.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Thu May 30, 2013 12:10 am 
Offline

Joined: Mon Oct 08, 2012 2:59 am
Posts: 24
Location: India
PaulEngr wrote:
If the fault is in the main breaker then you'd simulate a line side event anyway. the arc is magnetically propelled away from the power source (simple physjcs).


Well Paul, you are implying that the possibility mentioned by Jim
Jim Phillips (brainfiller) wrote:
2) What if the arc flash originates below the main and then escalates (within a few cycles) to a location on the line side of the main.

is not a possibility at all. Your point seems logical to me, its basic physics and I obviously missed that point before. Please cite appropriate literature/experimentation to backup the same.
This discussion, I think, is taking a 'U' turn. Now I think one of them (ETAP) may be wrong. The selection of immediate protective device should be the right thing.
Please comment.

Great discussion!!


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Thu May 30, 2013 8:52 am 
Offline
Plasma Level
User avatar

Joined: Mon Sep 17, 2007 5:00 pm
Posts: 1245
Location: Scottsdale, Arizona
Here are a few more things to consider.

Paul is quite correct, the arc does move away from the source. However, the hot plasma gas can reach other areas - Many of you have seen the pretty famous YouTube video from color security camera footage of the person racking in a breaker and the plasma cloud fill the equipment (and even escape out the back).

We just had a case last week where the arc was initiated at the bottom of an MCC section and the conducting plasma cloud / ionized gas rose to the top and initiated a second arcing point on the upper horizontal bus.

As far as the line side of the main issue, here is text verbatim from Page 76 Section B.1.2 of IEEE 1584:
Note that if an arcing fault can be initiated on the line side of a main protective device in an enclosure, that
protective device should not be utilized for the calculations, instead, the upstream protective device should
be used. That is because only an upstream protective device can be considered to provide protection for an
arcing fault on the line side of the main protective device in a downstream enclosure.

Practices will continue to evolve but at the present time, conventional wisdom is assume the arc can either initiate or escalate to the line side of the main unless the equipment is compartmentalized like switchgear and some MCC's Using an interpretation contrary to this is up to you but you will be in an extremely small minority.

Just more to think about.

_________________
Jim Phillips, P.E.
Brainfiller.com


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Thu May 30, 2013 4:15 pm 
Offline
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 1878
Location: North Carolina
My computer is being a pain as far as copying links but see "Understanding Arcing Faults" by Bruce Land III at APL-JHU in 2009 on the archives of bwisa.org. This was part of APL's effort to develop an arc flash detection relay for switchboards on navy ships. Bruce wrote several papers discussing forensics of panelboards on ships that were caused by arcing faults and first turned me on to finding arc tracks in bus faults to locate the source.

I've read about single phase arcs propagating into 3 phase arcs such as in metal enclosed switchgear arc flash research that Willis was doing at Mersen, but that is inside a single enclosure. I can easily make a case for this in switchboards where it is essentially all one enclosure. I would believe this happening within a single section in an mcc because outside of the buckets it is effectively the same geometry as a panelboard. There have been two cases of arc propulsion leaving very little evidence at the initial fault while pulling buckets out of live MCC in the last 10 years reported at my current employer. With the small size of the openings between sections, I have a hard time believing in propagation towards the main unless you have an obscenely high amount of thermal energy and burn through the enclosure walls. It doesn't seem likely that plasma could flow through such a small opening with enough heat/energy to initiate anything in an adjacent enclosure. It kind of looks like the JHU data might give insight into predicting this, or at least a direction.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Fri May 31, 2013 7:11 am 
Offline
Plasma Level
User avatar

Joined: Mon Sep 17, 2007 5:00 pm
Posts: 1245
Location: Scottsdale, Arizona
Paul, I completely agree. That's why somewhere towards the beginning of this discussion I mentioned MCC's and Switchgear being possible exceptions. A large arc flash (lots of plasma) in an enclosure could create this problem but again you are quite correct, a smaller plasma cloud? - not as likely but most would not want to take the risk. I believe the real problem we have is most interpretations tend towards legal / liability interpretations. Just a few years ago, many people used one calculation for switchgear fearing the "what if". Then they began to use two - one for the main and one for everything else as people thought it through. Hopefully as we get an even better understanding with more papers, research and discussions like this one, things can continue to improve. - Thanks!

_________________
Jim Phillips, P.E.
Brainfiller.com


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Jun 03, 2013 11:26 pm 
Offline

Joined: Mon Oct 08, 2012 2:59 am
Posts: 24
Location: India
The arguments presented by Jim are equally convincing. May be I should consider both devices (immediate upstream and further upstream) and prepare for the worst case.
Thank you everybody for your comments/insights.


Top
 Profile Send private message  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 13 posts ] 

All times are UTC - 7 hours


Who is online

Users browsing this forum: No registered users and 1 guest


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:  
© 2017 Arcflash Forum / Brainfiller, Inc. | P.O. Box 12024 | Scottsdale, AZ 85267 USA | 800-874-8883