This week’s question may set off a bit of discussion. It is about the qualifications for performing an arc flash study.

The results of an arc flash study can affect personal safety. An incorrect study could contribute to sever injury or worse in the event of an arc flash. There seems to be some debate about the qualifications of persons performing arc flash studies.

From IEEE 1584.1 IEEE Guide for the Specification of Scope and Deliverable Requirements for an Arc-Flash Hazard Calculation Study in Accordance with IEEE Std.
The arc-flash study should be performed by, or under the direction of, a qualified person with the necessary knowledge about power system analysis and arc-flash hazard analysis or experience in performing power system analysis and arc-flash hazard analysis. It is recommended that engineers who are new to performing the studies obtain peer or third party reviews from a more experienced engineer in this specific subject.
NOTE—Engineering licensing requirements of individual jurisdictions (states and/or provinces/nationalities) may require the analysis to be performed by, or under the direction of, a registered professional engineer.

From the National Society of Professional Engineers:
“that all engineers who are in responsible charge of the practice of engineering as defined in the NCEES Model Law and Rules in a manner that potentially impacts the public health, safety, and welfare should be required by all state statutes to be licensed professional engineers. NSPE recommends the phasing out of existing industrial exemptions in state licensing laws.�

There is sometimes an industrial exemption from PE licensing if the work is done for the person’s company.

Also from NSPE, The practice of engineering is defined as:
any service or creative work requiring engineering education, training, and experience in the application of engineering principles and the interpretation of engineering data to engineering activities that potentially impact the health, safety, and welfare of the public. The services may include, but not be limited to, providing planning, studies, designs, design coordination, drawings, specifications, and other technical submissions; teaching engineering design courses; performing surveying that is incidental to the practice of engineering; and reviewing construction or other design products for the purposes of monitoring compliance with drawings and specifications related to engineered works.

Having PE license certainly does not in itself qualify someone to perform an arc flash study. A person should have experience and understanding of electrical power systems, power system analysis and arc flash/studies

So, after all of that, here is the question.

Should Arc Flash Studies (for clients) be performed by or under a licensed P.E.
Yes
No
It depends

Feel free to share your thoughts about this!

This is a difficult question. First, I investigated an arc flash accident where the medical cost exceeded 12,000,000.00. the facility has an arc flash analysis performed, or at least they thought they did. The analysis company just made up the incident energy numbers. The labeled the switchboard where the accident occurred with an incident energy of 28cal/cm2. My analysis showed that the incident energy was between 120 and 140 cal/cm2. Unfortunately I was not allowed to perform a full analysis.

In another situation a company performed the arc flash analysis using the table method without considering the fault current or the clearing time.

In another case a major electrical manufacturer performed an arc flash analysis for a hospital. I had a change to review their report and methods and couldn't believe the short cuts they used.

I occasionally work with a company who has an engineer as one of the owners but he does not have his PE. This company does a good job.

So, where do we go? There definitely needs to be a standard on who can perform an arc flash analysis. Should a PE license be required? I think so but that will not solve the problem. I would be happy to help develop a standard.

That is an understatement - as you know! Thanks for your comments.

The legal side/licensing is just one issue. You have to begin somewhere if this is a person's first study. It is preferred if you work with a mentor for a while. That's how I did it back in 1981 - working with a mentor at Square D performing studies for about a year, then slowly moving out on my own. I was quite fortunate to be able to learn that way.

Like you pointed out, an arc flash study is not just your run of the mill engineering study. People's lives ultimately are at stake.

A colleague and I wrote the initial draft of IEEE 1584.1 and the PE licensing language - although it was then moved to a note. We are about to begin revising this standard to sync up with the soon-to-be released next edition of IEEE 1584.

There is certainty not a magic answer.
As Facebook would say "It's Complicated"

I voted no and seem to be alone but to put it politely, my position is that all the standards (IEEE 1584, IEEE 1584.1, and 70E) need to stay out of the PE argument.

Reasons: First, it's not going to fix the problem of poor workmanship. IEEE 1584-2002 is really just a bunch of equations when you get down to it, and the draft "2nd edition" is not much better in this regard. NFPA 70E is more like the high level framework, but leaves the details completely open. Similarly IEEE 1584 is just one piece of the overall picture but doesn't address for instance available fault current. Other standards address that but since most were written to address AIC, they are inappropriate for use with IEEE 1584. If as Jim stated, "It is preferred if you work with a mentor for a while", then obviously we have a set of standards that is very open to interpretation. Lacking clear guidance, you get lots of opinions and no way to discern the correctness of one vs. another. While I understand the concern since people's lives are at risk, this also points out why having a set of standards that is so open to interpretation is a fundamental problem. Many other safety standards have either a "prescriptive" or "performance" standard, or both. I would say that arc flash right now is kind of a performance standard but we aren't doing proof tests. IEEE 1584 is more in support of a prescriptive type of standard, but there are still large missing pieces, at least from the end user point of view.

Second, there are a lot of industries that don't need or require it. For example electric utilities almost across the board don't require PE's. All federal agencies neither require nor recognize PE's (it's state law, not federal law). If you add this requirement, at best it discourages them from following the standard.

Third issue, it's not just the "industrial exemption". The vast majority of "internal" engineers don't have PE's because the system is set up for third party engineering firms. Despite all the platitudes coming out of NSPE and NCEES they strongly discourage "internal" engineers from getting a PE due to the way the model law is structured. It pierces the agency status by making "independence" mandatory. What I mean by this is let's say I'm working for a large manufacturing company and we decide as a company that we're going to do an arc flash study and get compliant with 70E. The in house engineers at a minimum can look up results if the labels are unreadable, or missing, or some special circumstance didn't call for a label. If I know what I'm doing or say I go to Jim Phillip's training school and maybe one or two SKM training classes and I do the arc flash study in house using in house personnel, the company can use in house resources to do it, or at least maintain it in the interim (between 5 year updates) as the almost continuous plant modification projects proceed (business as usual). Then the liability for the company doesn't really change much. They still get sued if an employee gets hurt, and OSHA is going to issue a citation to the company, not an engineering firm. As an engineer I would just be doing my job so I would enjoy liability protection. This is actually the ideal situation for large firms. Maybe they go outside for the 5 year "update" to get an outside third party evaluation to be sure their internal engineers are doing it correctly, but I think we can agree then that this is the utopia that we would like to see happen.

If however I am a licensed engineer in the exact same scenario, then I need to be "independent". This means that the employer cannot influence what I do, and that I need to be fundamentally in control of the work product. I couldn't use the company's vast resources of technicians, electricians, etc., to do the inspections, which pretty much forces them into using an outside engineering firm. And at the end of the day unless I have my own company, I'm stamping the results with my personal stamp, so if anyone gets hurt, they're coming after me personally. I'm no longer "just don't my job". Despite all the platitudes and nice things coming out of NSPE and NCEES about making engineering a "profession" akin to doctors and lawyers, the reality is that most engineers are not working for third party engineering firms. and having a PE is actually opening them up to a major liability issue rather than giving them some kind of protection. All their efforts to "kill off" the engineering exemption are actually making this situation worse instead of better because they haven't fixed the ethics rules which destroy agency status.

In fact this situation has gotten so bad that Schneider/Square D in the Knightdale, North Carolina office has basically gotten rid of their PE's for liability reasons. Did you catch that Jim? They "moved" them all to an independent firm that they pay to do engineering studies for this very reason. It's gotten so crazy that they are questioning if a customer calls and asks for a starter for say a 25 HP motor, should they spec it out or just send them a catalog and tell them they have to pick it out? Or would sending specs to the customer actually constitute providing engineering which in a strict sense, it is.

Fourth, you are saying to 80% of the people that hold engineering degrees that they can't do arc flash studies. That 80% figure by the way is not my number, that's straight from NSPE. I work in the electrical services business and the vast majority of my customers are municipal or industrial. I haven't done any kind of scientific study but I can tell you MOST of them don't have arc flash studies, even doing as little as following the table based methods in 70E or NESC, something that requires very little in terms of implementation costs. Given the high cost of compliance, why would we want to set the bar even higher and discourage them further? The "bar" in this case has nothing to do with confidence or training or anything improving the results but is simply a guild mentality pushing everyone who works as an engineer out of society and into trade unions (engineering firms). Like doctors and lawyers, everyone knows that even picking up the phone to talk to one will cost you several hundred dollars and like doctors and lawyers, the general word on the street is to avoid it if at all possible. This will only continue to encourage the lack of compliance.

Finally as to the egregious and simply bad engineering involved in the arc flash studies, making a PE license mandatory will NOT fix this. The idea of using the table based version without consideration for clearing times and available fault current has already been mentioned but consider the flip side of this...I'm routinely working in customer facilities where they haven't done an arc flash study. So what then? Do I just do what the customer is doing (nothing)? Do I follow the table knowing it might not be sufficient? If I don't use the table based method, then what do I use?

What about the whole idea of being "conservative" and using "engineering judgement" (actually NOT using judgement), on the flip side? For instance 70E talks about the condition of maintenance but never clarifies what it means to properly maintain equipment. So as a PE, then I just assume that all electrical equipment is inherently not maintained properly and it's inherently a death trap, so to be conservative, you are now in a condition where the whole engineering study is garbage because as 70E states, the clearing times may be less or GREATER than calculated. So what is the incident energy? Infinite. So do we put a "get out of jail free" clause in the engineering study or simply put in all kinds of stupid and absurd results, and make wearing arc flash PPE while performing normal operations mandatory? Yep, we get really stupid but hey, it's "conservative".

Or along the same lines let's say the customer has a 34.5 kV distribution system. Following IEEE 1584-2002 we use Lee. The resulting incident energy might by 5 to 10 times higher than reality and makes it operationally impossible to work on, but hey, at least it's "conservative".

I can go on and on with dozens of examples from engineering firms, even if they know what they are doing. Theoretically it becomes client decisions but then we have that pesky "independence" clause where the engineering firm can't be corralled from doing stupid things simply because that violates their "independence".

The fact is that NSPE and NCEES have laid out a set of rules and guidelines that are similar to but actually more restrictive than doctors and lawyers. Unlike either one, we are expected to do something, not just "practice" at it. Our decisions are far from above reproach. We as a profession are for the most part "integrated" into other organizations. Most of us are not "independent" and never will be. It is a mistake to assume that this is the norm, or even desirable. In fact, I will go so far as to say that it is definitely NOT desirable. Let's face it, most industrial plants are effectively almost continuously under construction. They are constantly making small modifications, revamping ore replacing equipment, adding capacity, and a myriad of other "small" activities, all of which usually affect arc flash results at least in some small way. The idea of a "5 year update" as codified in 70E is a huge mistake. These firms need to really be doing updates to their arc flash studies "as needed", or in other words almost continuously. It is clear that if they have the in house resources, they should be using them to at least keep the power system model up to date, if not outright maintaining it themselves, but the best way to support and encourage this behavior is not by requiring PE's because as previously mentioned, this isn't a viable business model for internal engineers.

Amen on the comments about maintenance. As money gets tight, maintenance, especially long interval maintenance, tends to disappear. EPRI has much better maintenance documents than the basic manuals supplied by most manufacturers, but rarely would you find anywhere outside of nuclear facilities using these. Even there, for the non-safety related equipment, that is questionable. How many horror stories can we throw out about out of service or damaged relays, low voltage power breakers where sensors have been changed (not something typically checked during a arc flash study), relays pulled out for maintenance and then replaced in the wrong cubicle, MCCBs over 40 years old where the internal lube turned to cement a decade ago, mismatched fuses, overloaded panels being cooled by box fans, etc.?

The unknowns about maintenance are enough to convince me that the NFPA 70E table 130.7(C)(15)(A)(a) that asks if equipment is properly maintained is basically worthless. I have a fair amount of confidence after a decade of being a system engineer in charge of electrical equipment maintenance that that answer is almost always no. Arc flash PPE is always required in my world.

I hear what Paul is saying but I still voted yes. My reasoning is, that limiting who can do the studies to PE's or people working under the direct supervision of a PE probably, not certainly but probably, will increase the odds of the study being done properly.

If anyone can do the study, maybe even Joe Maintenance with not a shred of education or engineering background but who thinks they can properly apply the tables and save his company a lot of money, I think the odds are less the study will be done properly.

Do all PE's "know what they are doing". While I'm sure we all believe we are enlightened, the answer is no. I think there is probably a higher percentage of electrical PE's though, whatever that number is, who are capable and qualified to perform short circuit, coordination and arc flash studies than the percentage of "engineers" who are not.

I have seen reports done by several large firms and there were some serious errors in the report. One report was done by a PE and reviewed by a PE but they had used the SLG fault current for the 3 phase, pulled a SLG fault current apparently out of thin air, used 18" working distance on 46kV (I think they used Lee method as SKM defaults to that) which gave a value of 641 cal/cm2, and other issues.

So being a PE does not necessarily guarantee a good job.

Great conversation and some great views! This is obviously not a simple situation. I was just stating what is out there and my own experience. The comments that everyone made were very interesting and shed additional light on all of this. Thanks!

My point was pointing out a common fallacy that PE's make based on the idea of being "conservative", not trashing maintenance departments. Thanks for making my point.

OK, so let's take this to the logical conclusion. NFPA 70E in the informational notes in Article 130 that was already mentioned states that opening times can be greater than predicted if proper maintenance is not done. So you stated that your risk assessment concludes that PPE must be worn for all tasks. I'm with you so far. But one question that I'm just dying to know the answer to is this. IEEE 1584 requires a clearing time, and so do the 70E tables. But you already stated that you make the assumption that no equipment is being properly maintained, so thus at least using these standards, we can't determine the opening time. So what do you use as a standard for determining incident energy, knowing that IEEE 1584 and NFPA 70E do not apply?

I said yes per the definition of engineering in most states, and the client qualifier to the question. California might be an exception, since the industrial exemption extends to contractors and consultants.

Hi All

The Best to you from North of the Border. Obviously we are experiencing the same dilemma here in Canada as yourselves in the US with AFRA. Regardless at to whether one is a PE or not, who is policing the studies?? If it is like up here in Southern Ontario, nobody. It is the wild west.

Could we not consider some form of certification as a person or company recognized as in certified in this area of expertise. For example, Jim you have your 2 day course in which I have enrolled in many times over the years. Once someone has met the admiting requirements (PE for example) and successfully completed a course such as with yourself, one is now certified and deemed competent to perform AFRA. In other words a certain accreditation. Indeed competing the course would not be the only requirement, like the CESCP and other designations, one would have to obtain credits and auditing in order to remain qualified. The challenge is a governing or regulatory body. The PE associations dont seem to have the enforcement capabilities, at least up here in Canada

Just a thought

PEs and P.Engs have a duty to report. We police ourselves.

Is that "Duty to Report" working?

This discussion reminds me of many of the other issues surrounding what a journeyman/master/industrial electrician is qualified to do versus technologists and engineers. From a purely practical standpoint of availability of labor and economic resources, it would be prudent to use the scope of the job in combination with a rigorous training and certification process to determine who is allowed to certify a study and post labels on equipment. Electricians in our jurisdiction are more than capable of performing a straight forward calculation and table based study for smaller jobs or additions to existing installations if properly trained. As clearly illustrated in a previous comment, having an incompetently engineered study is possibly more dangerous than no study at all. Like all professions, self policing is only effective as long as nothing goes badly wrong - then the courts and lawyers take over. For the record, I am not an engineer, but I have performed my own studies just to determine my own protection when working for employers who neglect to have professional studies done.

With regards to the subject matter at hand, I haven't seen anything pursued by my state's board and there's good reason for that.

To begin with all of the underlying standards except NEC (which for arc flash plays a relatively minor role) are voluntary. Not only that but you really can't pin someone down on some standards. For instance with mechanical equipment it has been well documented that about 60% of the factors that determine equipment reliability are environmental and difficult if not impossible to quantify. Thus any kind of maintenance standard speaks about but not directly to a given piece of equipment. Electrical is similar but so far I haven't seen any similar documentation quantifying a result. This is somewhat alarming considering that there is wide agreement that condition of maintenance is a primary consideration in terms of arc flash likelihood.

Second, all of the standards as used stop short of giving exact and specific recommendations. They are performance based rather than prescriptive for the most part. In other words the key one, NFPA 70E, as typically used, gives a requirement to determine risk, hazard, and PPE. It gives annex information on PPE which is not a requirement. IEEE 1584 gives specific information on how to calculate incident energy but leaves the subject of available fault current calculations as well as enclosure determination wide open to judgement calls. IEEE 1584.1 fills in some of the gaps but it is much lesser known. NFPA 70E Annex F is purportedly a risk analysis standard but it is based on the wrong standard for the kind of hazard that arc flash is and what's more, it is completely incomplete. It has you determine a bunch of factors but doesn't document what to do with them. While I'm quite aware of the underlying standard it is based on, you could never use Annex F in it's current form and you really shouldn't be using even the underlying standard which is based on exposure to moving machinery in an assembly line setting, not rare but catastrophic events. In fact although this gets into my third point, it becomes a matter of judgement (ie, opinion) as to what standard to use for risk analysis and I'd also say emphatically that almost every single arc flash engineering study out there is totally inadequate because they pull things out of their rear ends and do not even bother using a standard for risk analysis. So it's pure opinion at best.

Third, and this is the big problem. If we have an expert opinion A and an expert opinion B, who is right? You can try to convince a jury or judge by trying to give an explanation but at the end of the day it is nearly impossible to prove problems with decisions that are based on opinions and judgements from a legal standpoint. Fundamentally we are asking the courts to make a decision based on two different opinions that nominally have equal weight. This is akin to when your kids get into one of those arguments where it's "he did it, no she did it" kind of thing where short of DNA and finger print tests, as parents we can't tell who did it and either pick one based on how convincing their accusations are or simply punish both, or neither. The courts are not interested in getting sucked into a legal morasse of trying to discern differences of opinion amongst expert witnesses and that's a trap that does nothing but create appeals.

Finally let's face it, everything is based on a huge pile of assumptions. Running the IEEE 1584 calculations, even the new ones, or one of the various calculations determining available fault current and opening time is so straight forward that we have software programs that do it but based on inputs that we provide. But there's the rub...the assumptions that are baked into the calculations in the first place. The arc flash report will document the source of the assumptions and there are often many. It would be impossible for the courts to hold an engineer responsible for using the schematics and one lines PROVIDED to do calculations without doing a visual inspection, even if that information was completely wrong, because the report should state the source of the information. Similarly the engineer might go as far as doing a selection of inspections without specifying which equipment or how thoroughly the selection is done, and whether it is random or based on convenience. If the specific equipment that is inspected is not documented, the statement amounts to little more than a meaningless statement meant to add credibility to the content but is without support of any kind. Even if a through inspection is done, there is a big difference between finding and reading labels or using calipers to measure cable size compared to a 100% visual inspection where someone is just "eyeballing" the cables and bus work. So when we bake all these things in, the arc flash engineering study amounts to little more than a compilation of some calculations done on a bunch of assumptions and opinions, and the calculations in turn are based on application of incomplete standards that leave most of the key elements open to "judgement" (opinion).

The alternative to this is that the standards become highly prescriptive to the point where the engineer judgement is really only necessary when the actual equipment doesn't match what's in the prescriptive requirements. This environment doesn't even require an engineer. For instance electricians frequently determine required cable size. It's a mostly straight forward set of calculations and tables that are well documented in the NEC. At times, particularly buried cables, resorting to Neher-McGrath software is necessary but most of the time the NEC procedure suffices, and it is a regulatory requirement in most jurisdictions, and a recognized standard in all of the others in the U.S. It is kind of hard to argue about the correct cable size because the determination is so cut-and-dry. It would be quite easy to hold electricians accountable to that standard. Thus inspectors don't need to have engineering licenses. This is a far cry from where we are at with NFPA 70E / IEEE 1584 / IEEE 1584.1 / NESC / say AiChE LOPA today.

https://www.oregon.gov/osbeels/rulesstatutes/Pages/Disciplinary-Actions.aspx

There are two unlicensed practice violations here. Not sure how the board became aware of them, those details are left out. But since the board has no police on staff, and the volunteer board members themselves are busy with their own jobs; I suspect duty to report may have played a role.

Most of the time someone calls the board to verify a license and when the board says no such person, that's when they get tipped off that there's no license. But this has absolutely nothing to do with enforcing any kind of quality control with respect to engineering studies performed. This is nothing more than trade guild activity. It's equivalent to the writer's guild or the actor's guild going after someone in the entertainment industry for writing screen scripts or acting in a film without a guild membership, without regard for the workmanship involved.

I've been a PE for a little over 20 years now. In all that time I can think of only one instance when we (PE's in my state) received a letter from our board about a disciplinary action taken on an engineer for stamp-for-hire and a warning to the rest of us not to stamp drawings prepared by other entities. In the discipline case, I think the engineer was stamping plans prepared by a contractor as I recall. The engineer had no involvement in or the supervision of the design preparation. He just took the money and stamped the drawings and submitted them to the State for design release.

I don't even remember now what the penalty was. It was many years ago.

Our board is pretty active but you can't really tell what the results are.

Spring newsletter:
http://www.ncbels.org/newsletters/spring2018.pdf

There were several violations and board actions mostly related to practicing without a license but let's narrow it down to the two that refer to competence:

"VIOLATION: Performed services outside area of competence [.0701(c)(3)]; affi xed certifi cation to inadequate design documents, failing to protect the public [.0701(b)]; produced defi cient, substandard or inaccurate reports, failing to protect the public [.0701(b)]; and failed to properly certify documents [.1103]." It does not call out specifically what the issue is but the board action is requiring the individual in question not to sign off on "high risk structures" (with details explaining what that means) until passing the structural engineering exam. I don't believe this one could hold up in court though since state issued PE licenses are either revoked in total or not and there is nothing in the regulation or the law referring to specific disciplines except stating that engineers may not work outside their areas of competence. The newsletter account leaves much to be desired.

"VIOLATION: Produced a deficient, substandard or inaccurate report, failing to protect the public [.0701(b)]." No further detail. The guy got mad and refused to renew his license so they left it at that. So no further details on what the violation is.

They usually publish 3-4 of these type of cases per year. The vast majority are protectionism stuff (practicing without a license) and land surveyors also have lots of board actions for drawing deficiencies since they have a detailed, written standard unlike engineering.

However that being said, I have never seen one dealing with arc flash or really any electrical engineering study other than "practicing without a license" violations.

We can test that duty to report easily enough. Anyone here wish to fess up to unlicensed practice or practicing outside their area of competence? Please provide name and jurisdiction.

I also need to dispute that Canada has no enforcement. I distinctly recall the Ontario board going after Microsoft over their network engineer certification. I believe the fine was $1000 Canadian.