Hello,
I am new to arc flash and am a plant engineer. I have trying to understand about arc flash as we are in the process of having a study done. I have been told about not taking any credit for the main device inside a panel, like the type in your house. I can understand in that situation but can't grasp why on larger equipment.
However, I am confused on why if the main breaker is mounted in a separate bay, that it can not be used for the rest of the switchgear. Specifically, it is Square D QED-2. The main breaker is in a separate compartment at the end. Here is a picture:

There are no internal barriers that would prevent an arc flash from engulfing the main breaker.
The distance from the load side bussing to the line side connections is only a few inches. If the arc propagates along the horizontal cross bussing it could easily bridge the gap, therefore preventing the main breaker form clearing the fault.

Yes, it might be possible that the main breaker could trip fast enough if the fault originates in the furthest section, but common industry practice is to assume it will not.

I would not be surprised if, someday, manufactures figure out a way to slow down the arc propagation just enough to allow the Instantaneous function of the breaker to clear a fault before it is engulfed (some are already doing this with 480V motor control centers). This would not be true arc resistant gear (which requires special construction and testing).

This is an issue that continues to be the subject of quite a bit of discussion. I just returned for IEEE / IAS Electrical Safety Workshop in Dallas and also the latest IEEE 1584 meeting and I had a few side discussions about this. It is still a gray area.

Many in the industry use a 2 label approach for switchgear. One label for the main based on the operation of a device further upstream and one label for all of the feeders based on the main. Some have begun using this approach for other equipment such as MCC's but it is up to the judgement of the study engineer and/or the person performing the study as well as the equipment design as JBD points out.

We really don't have a good answer when it comes to the question: "will the arc flash propagate to the line side of the main" so the conservative approach is to assume it will with equipment such as panels. Does it always propagate? - No. Could it propagate to the line side? - depending on the equipment type - yes it is possible.

The clear exception would be true Switchgear construction. This is where 600V class equipment has a 60" (or close) deep construction and all the breakers are rear connected and each cubicle is a true isolated compartment. Currently it is only possible with metal clad circuit (air frame) breakers, except for Square D, they have an encased beaker version.

I reviewed a MCC arc flash incident last week. The affected section only housed a size #1 starter, however, the damage was great. Each section within this MCC appeared to be very well isolated from the the main bus. The top of the circuit was located near the top of the enclosure and it arced over without human intervention. I would assess this incident to be in the range of 10-20 cal/sq cm. My point is that the fault propagated to the main bus even though the specific section looked to be mechanically isolated. That fault could have easily gone to the line side of the incoming breaker, therefore, the breaker would not be part of the incident energy calculation. As a final point I investigated a similar MCC accident two years ago and the electrician died.

Just an FYI: This is not a "Switchgear". This is a "Switchboard". Two very different animals.

Hello,
Thank you all for your replies. However all this puts me or my plant into a difficult situation. Apparently I am told if credit is not taken for the main breaker for tripping, the entire switchboard is dangerous at 75 cal/cm. With the main breaker, it is around 4 cal/cm.
Given this situation, in order to lock out equipment fed from the switchboard we would have to shutdown by having the utility open up the fuses?? This causes severe operational issues (24/7 plant).
Any help???

Have you performed a risk analysis of your tasks?
Have your Electrical Safe Work Practices been vetted by a qualified entity, such as your insurance underwriter?

Looking through the proposed 2015 changes to 70E (there are links in other threads) it looks like your situation may be addressed in a new section 130.2(A)4 where it may allow for 'normal operation' based on 5 specific conditions.

Barring all of this, you would need to separate your main device from the distribution sections (relocate it or add an upstream device).

We have not done a risk analysis. How would that help in this situation where one needs to turn off a circuit breaker to power down equipment for lock out? apparently just operating the breaker is a danger risk of arc flash because credit cannot be taken for the main breaker in a separate compartment on the switchboard. 2015 is not going to help me in 2013.

You are required to do risk assessments on your tasks. We used to call these JHA's. This is true even for 70E. A hazard analysis looks at what could happen but does not look at the likelihood. That is what the notes in the arc flash hazard definition and under the arc flash hazard analysis are referring to when they say that properly maintained and installed equipment is not likely to cause an arc flash. So depending on what you are doing, arc flash PPE may not be necessary. 70E gives no help on this. There are separate ANSI standards for risk analysis. As somewhat of a shortcut since 70E tables do include a risk analysis you could use the tables for some tasks such as operating breakers and use IEEE 1584 for others.


As to the propagation argument, I have not seen a study done on this and I have not seen everyone doing this so its a grey area. In MV bolted metal enclosed gear that we have where we've had a rash of switch failures, I haven't seen a shred of evidence for the "propagation" argument. And this stuff is both high power and higher voltage and pretty open from one compartment to the next where you'd expect to find evidence. In these cases as long as you can justify your position you could make a case either way. My concern is that I have seen smoke and sometimes arcing but not much else. I am not interested in overprotecting just because someone a PE is trying to put their interests above what I pay them for with theoretical concerns.