How can you garauntee an Arc Resistant Switchgear can safely operate during a Arc Fault?

[Wfg42438]

Hello Everyone,

I was reading an IEEE paper published a few years ago regarding Arc Resistant Switchgear

The paper discusses an Arc Faul event in an Arc resistant gear and a visual inspection that can be performed in time current plots to ensure the equipment can truly operate as an arc-resistant switchgear

In the paper, the authors claim that if the equipment can safely operate when within the arc-resistant design region (see attached images and paper)

The Arc Resistant design region is drawn as the intersection of a horizontal line using the Arc Current rating and another vertical line at the arc duration rating of the equipment in a TCC

This is then evaluated against the protective device curve and the arc current flowing through it

In their sample evaluation, they mention that the as long as the clearing time of protective device does not exceed the Arc duration rating and Internal arcing current then the equipment will operate correctly

Now my question is can we truly say the equipment won't behave as a arc-resistant switchgear if the Arcing duration is exceeded but the arc current rating is not?

Does anyone out there have any references or experience with this evaluation?

Any feedback would be appreciated, thanks!

Link to the IEEE Paper:

https://www.eaton.com/content/dam/e...or-control-center-for-an-in-service-fault.pdf

 

[stevenal]

Now my question is can we truly say the equipment won't behave as a arc-resistant switchgear if the Arcing duration is exceeded but the arc current rating is not?

No, I don't think anything can be said either way, since there is no testing performed beyond the 0.5s duration limit. The paper you linked did not say "won't", it said "cannot be expected to." My questions then are: Where will your worker be when the 0.5s limit expires, and what will be his IE exposure at that location in relation to his PPE ratings?

I disagree that 0.5s is normally sufficient for an upstream device to clear the fault. I gave a common example in another recent thread.

I note there is no adjustment of the 0.5s duration limit by (I^2)t in the IEEE paper as suggested by a post in that thread.

 

[dpc]

No, there is no guarantee if duration exceeds test time. Using an I^2t reference seems logical for arc energy, but I don't think that there is anything in the standard to support that. Similar scenarios are common for all short-circuit testing and ratings.

 

[DM61850]

0.5 seconds is a long time if someone buttons up the protection. Maybe, installing a faster protection scheme would be better and easier than try to find switchgear that can hold out that long.

 

[FreddyNurk]

DM61850, its obviously a compromise, but ideally it'd be that the protection scheme is set to trip within the identified timeframe, so the combination of switchgear withstand and protection trip time should hold.

Whether the 0.5 second duration is required in order to coordinate with downstream protection, or it allows for sufficient time for the upstream protection to trip as a backup within the timeframe is probably on a case by case basis, and likely the subject of much debate, noting stevenal's observations regarding the timing of backup protection as well.

The opposite argument to yours, of course, is that the if the protection is fast enough, is arc resistant switchgear really required? To be honest I'd likely still prefer both, a fast trip time and the ability to withstand the delay until the upstream (backup?) protection trips as well.


EDMS Australia

 

[stevenal]

So your primary protection trips instantaneously for downstream faults. If this breaker arcs internally the next device upstream will need to clear the breaker fault. The upstream device needs to delay long enough to ensure the primary doesn't clear the fault, so it is delayed by a suitable coordination interval of say 0.25 s. Total clearing time is less than 0.5 s, so all is well - so far. The upstream device is also arc-resistant, so it's backup protection must now be delayed by further by suitable coordination interval of another 0.25 s if it should arc. The relay for this third device trips at approximately 0.5 s, but breaker clearing time must be added putting you above the allowed duration.

Simple coordination won't get you far this way. To button up protection for this example, you will need to use communication based protection. My concern is that when the dollars are added up, the decision will be to specify non-arc-resistant gear that doesn't have the 0.5 s coordination limitation. I think the arc-resistant gear provides added safety even when durations exceed the rating, since you've added at least a half second (and probably more) to move away.

 

[DM61850]

You don't even need communication based protection. Go install differential or arc-flash light sensing relaying. I would just button it up. Can you live with someone getting hurt or the downtime due to your switchgear being totally wrecked and needing to be replaced? An arc flash sensing SEL-751A can be had for 1k. SEL-387A differential can be had for under $4k.

 

[davidbeach]

I don't believe the 751A has the optical arc flash sensing, but the 751 does.

 

[DM61850]

It does. When I was at SEL, the reason they added an A suffix was for Arc Flash. You have to specify it. If you specify a 751A with arc flash sensors, that puts you at $1.7k, not $1k. When I grabbed the number above, I just used the base number value.

 

[stevenal]

...downtime due to your switchgear being totally wrecked...

The whole concern here is what happens after the switchgear has already failed and arced over internally. The wreckage and resulting downtime already exists.

With my scenario above with three levels of coordinated protection, my assumption was that at least one level would be remote from the others. Buttoning up would therefore require communication. If you sense an arc in the downstream device the signal needs be delivered upstream.

A question for the members: A non-arc-resistant breaker is due for replacement remote from upstream protection. The existing relaying is in good shape. Would you..
A. Replace the breaker with an arc-resistant style and button up (replace) the relaying, install communication to the upstream device, and do whatever else is needed to ensure faults in the new device were cleared within 0.5s?
B. Replace the breaker with a non-arc-resistant style, rejecting any arc-resistant styles even if they are low bid? (One manufacturer I know no longer builds non-arc-resistant breakers at the voltage I generally need).
C. Replace the breaker with an arc-resistant style with no other changes, figuring the extra 0.5s can only enhance safety over the existing breaker? (no incidents to date)
D. Other?

 

[dpc]

This is probably obvious, but keep in mind that the arc-rating is not of much significance for personnel safety if a door is open when the fault occurs.

Also, the use of arc-rated switchgear doesn't necessarily reduce equipment damage for an internal fault.

 

[Wfg42438]

Thanks for all the feedback it's much appreciated!

Based on the comments it's my understanding that when only one of the arc-rated switchgear ratings (current or time) are exceeded its a gray area where it's not clear what would happen.

Based on this we can only expect the gear to behave as an arc-rated switchgear as long as both the current and time rating is not exceeded

The author seems to mention this in his example where he shows if his arc duration exceeds the rating then the gear cant be expected to operate as an arc-rated gear

 

[AVR_1.0pu]

I've seen something recently about switchgear with a so called maintenance switch, whereby a faster trip time is set before you enter the vicinity of the switchboard, hence reducing energies. Eaton Arcflash Reduction Maintenance Switch and Schneider (

https://www.se.com/us/en/faqs/FA272576/)

are examples. I've not seen it implemented.

Cheers/

AVR