I have been asked recently by clients of an arc flash study why the flash protection boundaries are so huge. I started looking into this and noticed that the IEEE method tends to give flash protection boundaries that are often significantly longer than those calcuated using the equations in NFPA. Does anyone know the reasons for this difference? Initial thoughts are that the longer IEEE results are based on the reflection and focusing effects of the enclosure.

Clients don't like seeing that they have to put on PPE when within a half mile (exaggerated) of their switchgear. How have you seen the flash protection boundary handled?

Welcome to the 1584 Standard! Most of the theory is based on total incident energy exposure. This means you take the energy and the duration to perform the calculations - as I am sure you have seen.

The problem is if you have a low short circuit current, often a device's time current curve will indicate a very long clearing time - often 10 or 20 seconds or even longer. If you take the low energy from the low fault current and then extend it for 10 or 20 seconds or more, the TOTAL energy is quite large. The question becomes, is it valid? There is not a perfect answer. Can the person jump away in 1 or 2 seconds? Can the arc actually sustain itself that long? There is more research yet to be done to answer these questions. IEEE allows cutting the clearing time off at 2 seconds if someone has room to jump out of the way - which can still be a long.

The problem compounds itself because the Flash Protection Boundary calculations are also based on the incident energy calculations. If you have a very large incidient energy from a very long clearing time, the FPB calculations can not distinguish between a lot of incident energy over a short period of time (exposion) or a small amount of incident energy over a long period of time (shower of sparks). Therefore what you get is a large (and often not realistic) flash protection boundary with low fault currents and long clearing times.

This will likely be cleared up in later revisions to 1584 but for now a bit of "realistic" judgement has to be made.

b.t.w. I actually have heard of FPB's hitting around 1/2 mile!

I have a bit of a different take on the problem.
Much of the problem comes from the "distance exponent", x. This is a factor that depends on the voltage level and equipment type. This is a calculated "fudge factor" that we assigned based on test results.

The test data, although large, failed to include data at any reasonable distance out. The result is that beyond ~5 feet, the calculation is an extrapolation of the data. I don't believe that any calorie measurements were taken from instruments further than 5 feet from the source.

More data needs to be taken to validate this extrapolation, but I don't see this in any of the project proposals for the IEEE/NFPA effort.

It's probably a little bit of everything. The low fault current / long clearing time is one of the problems many bump into. I'm sure the X also has a bit to do with it and there is also the Calculation Factor which is the ultimate "fudge factor". Some people I know were tracking the biggest calculated boundary and did hit close to 1/2 mile. That's requires one very long hot stick! Of course I believe they made a more logical adjustment of 2 seconds.

Boundary & PPE -> Doors Open vs Doors Closed

Are there any specific instructions regarding how to apply the boundaries and incident energy levels? Here is some background to frame the question: I am involved in an arc hazard study for an old generation plant. When the protection systems were designed (60's) plant security (long clearing times) were the driving factor. Consequently there are a lot of results >40cal.

The plant intends to do coordination studies and work to mitigate the problems. However, this will take some time (1-2 years). In the mean time the cat is out of the bag and labels with as-is incident energies and arc hazard boundaries are being posted.

In general I think the plant understands that for these high incident energy cases, de-energizing gear prior to opening up enclosures for maintenance is what needs to happen. But what about:

1- Switching operations when the doors are closed?
2- Standing outside the cubicles observing meters?
3- Live work in a switchgear feeder cubicle requiring category 2 PPE that is right next to an incoming main cubicle rated above category 4?

There are a number of situations such as these examples. There seems to be a high-stakes tradeoff between people protection and practical operation (the "been doing it for 30 yrs" argument).

Any suggestions?

1. No difference, the 2009 tables have these tasks the same HRC for racking breakers with doors open or closed. Unless it is arc rated switchgear. For switch operation, the HRC's do differ, but not for racking.
2. Again, refer to the tables, as long as you are not "Interacting" with the gear you dont need FR clothing for this.
3. Good question, and not really a clear answer. My understanding here is go with the HRC2 in this case. You shouldnt be working on the part of the system that is HRC 4 (I assume it is high due to the OCPD being the fuses on the MV sode of the transformer).

Between a Rock and a Hard Spot

Thanks all for the information. We just finished results for the generation plant I mentioned in previous posts. There are a lot of red "Danger" labels (>40 cal). The client is balking a bit because now they are between a rock and a hard place. These labels would preclude live work and non-remote switching, but because they are a generating plant shutdowns don't come often and live equipment maintenance has to be performed. It could be some time before permanent mitigation is in place. Any suggestions for this type of situation?

http://www.remoterackingsolutions.com

If you like I can give you a full list of the generation plants that already use this system, it is a long list and is becoming commonplace in generation plants.

There isn't a lot you can to reduce IE except for the basics:
1) Set your breaker settings to operate faster and reduce 't'.
2) Increase your working distance if possible.
3) Install current limiting fuses.
4) Increase impedance to reduce current, ie, reactors.

Using the NFPA task matrix may offer some relief but you have to make sure the Ibf and t in the field are equal or less than the Table notes.