NFPA 70E says in 130.3 that the flash hazard boundary and PPE must be determined.

130.3(A) then gives Ralph Lee’s equation to determine the boundary, as an alternate to using 4 ft for other than 300 kA-cycles.

130.3(B) says that if work is to be done within the boundary, then exposure levels must be determined, either by the “detailed rolleyes: ) flash hazard analysis†in 130.3(A) or by the tables. There is very little further information on how to calculate the energy levels, although additional information is available in Annex D.

Annex D outlines no less then FOUR methods to determine the energy levels. (Lee, Doughty – pre-1584, one for >600V (which I’ve never seen before), and 1584). It is my understanding that the information in the annexes are not enforceable, but are for information only.

Also in use in the industry are Arc Pro, which I believe uses a physics based model, and the Duke Energy program, which uses (another) physics based model.

All of the methods are accepted within the industry as valid, yet NFPA seems to say that you must use Lee’s equations. So, the question I have comes in two parts:

1) Are we really held to Lee’s equation by the NFPA 70E. I’m talking about strictly according to the code, not enforcement.

2) If we are not held to Lee, and Annex D is not enforceable, what is even keeping me to the equations there? I personally like Robert Wilkins equations, can I use them? Can I just make up my own?

NFPA and IEEE confusion

I think you cut to the heart of one of a very confusing issue. Back in 2004 shortly after the latest NFPA 70E was released, I wrote an article for the NEC Digest that was published at the begining of 2005. The article was intended to show the relationship between OSHA, NFPA 70E, NEC and IEEE 1584 fit together when it comes to electrical safety and arc flash. [url="http://www.brainfiller.com/documents/OSHANECNFPA.pdf"]Copy of PDF of NFPA/OSHA Article[/url] My draft was reviewed by the NFPA’s top electrical safety person at that time. I was required to delete almost all references that I made to IEEE from the article including the title. I still have a copy of the reviewer’s comments that explain the relationship (or lack of) between 70E and 1584. I’ve copied some of the review comments below (from 2004) that might help with all of this.

Here is a line from my original article that was required to be deleted:

“In addition, IEEE Std.1584tm, IEEE Guide for Performing Arc Flash Hazard Calculations, contains the recommended practices for performing detailed arc flash calculations that can be used to determine the correct PPE for the amount of incident energy that could be available at a given location�.

Here is the reviewer’s response to the above paragraph:

“IEEE 1584 is a guide that is a possible method of performing arc flash calculations. IEEE 1584 was submitted in a proposal to be the only method of determining arc flash calculations. The committee rejected the proposal and chose to put the information in the annex. Annex information is informational and not an enforceable part of the standard which means it is not embraced by the committee. There are also other methods and examples in the annex from other sources.�

I also had this paragraph in the article but was required to delete it:

“The Institute of Electrical and Electronics Engineers (IEEE) has developed a standard known as IEEE Std.1584tm, IEEE Guide for Performing Arc Flash Hazard Calculations, which defines formulas and procedures used to calculate the amount of incident energy that can be released during and arcing short circuit.�

The following statement was also made by the NFPA reviewer referencing a statement I made about using the IEEE 1584 formulas

“The formula in the standard is the Ralph Lee formula. It is the only formula that is enforceable. There are formulas in the annex from different sources. None of them are recommended and are there for information only.�

During the last IEEE 1584 meeting that I attended in September 2007, one of the people in the meeting that is on the NFPA 70E committee stated that the NFPA does not want to endorse other methods including IEEE and again referenced Ralph Lee’s formulas. There was an immediate response from many people in the room about IEEE being based on more current research and models and why would someone want to use formulas that date back to the 1980’s.

There does seem to be some disconnect between NFPA and IEEE however this is presently being resolved. A joint collaborative effort has been formed between IEEE and NFPA to work together on the next generation of research and theory. This effort is presently underway with some collaborative tests (such as DC arc flash) already completed. I am hopeful that with both groups working together now, much of this confusion will finally be put to rest.

In the mean time, everyone I know that performs a detailed study uses IEEE 1584 so they can defend that they have used information based on more current research. Most of the major software manufacturers also base their programs on the IEEE methods (although some have other calculation methods as an option) Some utility based software such as ArcPro does use a physics based model which is based on a single phase arc flash.

So where does this leave us in regards to your question? For now, still confused.

So the rule is, we don't know which rule to use??

Which rule to use?

I guess Ralph Lee's equation does not cover arc flash in a box! The formula cannot hold true when trying to do arc flash hazard analysis on switchgears and panels below 1000 volts, which definitely has a more severe arc flash considering higher fault current levels and the containment of the whole arc flash/blast energy inside the compartment! Using Lee's formula at the said voltage range will give very conservative incident energy levels which you will underprotect your people.

The heat transfer DOS software FLUX by Duke cannot go near with results from other program solutions because that program was done with the open air, single phase arcs as its model. It's no wonder results do not come near with other calculations!

I was really hoping to get more response, but the board has been slow lately.

It sounds to me like if the NFPA decided to through the book at all of us using 1584 or other methods, they could. And since they are backed by OSHA, that's a pretty scary scenario.

By the way, anyone know what method the NESC (C2) uses? This is an IEEE standard, anyway. I've heard they used ArcPro for their tables.

I consider NFPA one set of opinions. I consider IEEE a different set of opinions. I consider "industry concensus" a different set of opinions to both of the previous.

For industrial plants I have yet to see a faciilty that is not using NFPA 70E with the calculations of IEEE 1584. It would require very compelling science for me to assume the liability of not following this industry concensus.

Ralph Lee's Paper

Does anyone know where I can get a copy of Ralph Lee's original paper? I have searched Google but can't seem to find it posted anywhere. Is it out of print? Does anyone have a file copy?

Thanks!

Update

As an update, there is a good article in this month's IEEE Industry Applications by Tammy Gannon and John Matthews, "IEEE 1584-2002 Arc Modeling Debate". Tammy is part of the NFPA/IEEE collaborative effort to study arc flash.

The article discusses the various methods of calculating the arc flash energy, and the reasons behind some of the controversy.

In the end, it looks as if Ms. Gannon is leaning toward the Wilkins equations, which I mentioned above. I've been an advocate of the Wilkins model for some time, as I believe it is not only a better fit of the data, but also takes the physics of the event into account, and is not a simple statistical fit.
Good job Ms. Gannon!

I do, check your PM inbox.

Ralph Lee Paper

Here is a link to PCIC's free version of Ralph Lee's Paper.

[url="http://www.ieee-pcic.org/archive/The%20other%20Electrical%20Hazard%20%20%20Electric%20Arc%20Flash%20Burns.pdf"]The Other Electrical Hazard: Electrical Arc Blast Burns[/url].

This works just fine up to about 600V but it is wrong. It has V as equal to A in prediction of arc flash boundary. This is not correct and we know it. Ralph did something no one else had done and I only hope all our work can be as useful as long as his has been but it gives particularly bad numbers the higher you go. I really look forward to IEEE/NFPA releasing a new model but even that is unlikely to fix the issue with higher voltages unless money is donated to do higher voltage work. This came straight from the folks doing the study. They are going to make a better model with lots more research but I don't know that it will really give better numbers that what we have right now. I do believe it will be more accepted by the world community IF they vett it properly but the biggest problems I see in the current models is leaning on the Lee equation for Medium Voltage on up. It is wrong because arc doesn't work like the Lee equation all the way up.

I have one research project still waiting on funding to show how far a medium voltage arc can even possibly be pushed. Some IR will go a long way but not 200 FT like some of the calculations say. They are just plugging and chugging Lee Equations. It will get better with time and if you don't do medium voltage work, it doesn't matter much.