It is currently Thu Feb 22, 2018 9:19 am



Post new topic Reply to topic
Author Message
ekstra   ara
 Post subject: Absurd Incident Energy Value
PostPosted: Mon May 02, 2016 5:09 am 
Offline
Plasma Level
User avatar

Joined: Wed May 07, 2008 5:00 pm
Posts: 711
Location: Rutland, VT
When reviewing Arc Flash Hazard reports for developing safety programs, I find I can still be amazed at the values I see that are not questioned by the engineer doing the study. It seems that whatever the computer spits out is gospel. The latest is a report done by a major manufacturer of electrical equipment.

The study included the 115kV switchyard as well as the lower voltage equipment in the facility. Since the study used SKM, the 1584 equations were applied to the open air 115kV yard. A 2 second cutoff time, 153 mm gap, Eqpt Type Air and Working Distance of 18 in were used. This resulted in an AFB of 1,834.07 in and 12,406.99 cal/cm2.

Just as a comparison, using the same values but utilizing ArcPro, I get 113 cal/cm2 (18 in SLG) and 248.6 cal/cm2 3 ph using ArcPro multiplier.

Of course the working distance used is not appropriate for 115kV as hotsticks are used to test and ground. But regardless, the value was never questioned.

_________________
Barry Donovan, P.E.
www.workplacesafetysolutions.com


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Absurd Incident Energy Value
PostPosted: Mon May 02, 2016 6:21 am 
Offline
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 1878
Location: North Carolina
That's because SKM followed IEEE 1584 as written.

IEEE 1584 contains 3 sets of equations:
1. A set of current limiting fuse equations. These actually produce the lowest incident energy values of the 3 cases.
2. The empirically derived equation, the one and only model that IEEE 1584 is "known for", and
3. The Lee theoretical equation.

IEEE 1584 does not necessarily give more or less validity to one model over another but if you plug in 115 kV data since the fuse models are only applicable to medium and low voltage fuses and the empirical model is only applicable below 15 kV, you will automatically get the Lee model since this is the only one of the three sets of equations that is valid.

Unfortunately many end users are totally unaware of this. I understand the reasons (development costs) but part of this may be for one obvious fact. Most engineers have seen and probably have at least one copy of NFPA 70E of some vintage. I think it currently costs around $90 for nonmembers. On the other hand IEEE 1584 costs around $600-$800 last time I checked. Very few engineers actually have a copy and have read IEEE 1584.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Absurd Incident Energy Value
PostPosted: Mon May 02, 2016 6:47 am 
Offline
Plasma Level
User avatar

Joined: Mon Sep 17, 2007 5:00 pm
Posts: 1245
Location: Scottsdale, Arizona
wbd wrote:
When reviewing Arc Flash Hazard reports for developing safety programs, I find I can still be amazed at the values I see that are not questioned by the engineer doing the study. It seems that whatever the computer spits out is gospel.


We are well over a decade into performing arc flash studies and this is still a problem. Often it is the result of one of two issues:

1) The engineer (or other person performing the study) does not have a sufficient understanding of the models used and significance of proper modeling and questionable results. Paul's comments are quite appropriate

2) Item 1 coupled with liability issues. "If the results are X, why should I question this and take the risk" It is "Conservative"

The industry focuses on training the electrical worker but training and experience is also important for the person performing the study. Otherwise..... GIGO Not just training on how to use the software but understanding the equations, methods, etc.

What is really concerning is when someone ask's me:

"If I buy a copy of XYZ software, can I be an arc flash expert too? By the way, where do I get short circuit data"

Yes, I have heard this statement too many times. It is good the client has you to review the reports. Too often a client simply "checks the box" - Study Complete.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Absurd Incident Energy Value
PostPosted: Mon May 02, 2016 8:40 am 
Offline
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 1878
Location: North Carolina
It's not just a problem for engineering power studies. When is the last time you've seen an engineering drawing that has 3 blanks for signatures/initials at the bottom, "draftsman", "engineer", and "checker", and only one of the 3 blanks is used or that all 3 blanks have the same initials?

I've also seen the same thing with environmental modelling, climate modelling, weather modelling and worse still, financial analysis/modelling. At one time we would observe something, develop a "model" in engineering terms or a "hypothesis" in scientific terms. Then we would develop computer code to test the model where we would do several runs of both valid and invalid test cases and verify that the output was correct (correct being a hand calculation), and then and only then, we'd use the computer code to validate or invalidate the model (or hypothesis). Somewhere along the way we stopped doing all the cerebral work and started accepting the results of the computer code as the same thing as the model. So by the time we get to this point we have an unchecked computer code which we now accept as "the model" with no support whatsoever for it being used as a substitute for what we used to do.

The problem here as I see it comes from two sources. First and foremost, we have somehow gone down a road where we are now accepting poor workmanship. It is well known in quality circles that absent other agendas or external stressors, humans have an error rate of approximately 10%. Put another way, we screw up about 10% of the time. Having that extra "checker" in the process reduces this to 1% because now both the checker and the engineer have to screw up 10% of the time to let something through. If we have those external stressors/agendas, then the error rate goes up even more to the point where it's basically a coin flip. Somehow a 90% success rate has become the new normal and acceptable. Other than a short fad in the 1980's with regards to quality control especially in the automotive world and a more recent fad in the petrochemical industry similar to nuclear where probabilistic safety analysis has become in vogue, very little concern has been shown for this.

I think there is a very basic error in judgment that got us here. I think we all admit pretty quickly that humans make mistakes with a pretty high rate, with the major damage being pride. We also recognize that machines by and large make errors but the error rates are usually very, very low relative to human error rates (by orders of magnitude). The fallacy here is missing the fact that machine produced results are based on inputs of which 10% are faulty, and those inputs include not only key punch errors but also include errors in judgement as well as errors in interpretation even if the machine produced results are indeed valid.

The second issue goes back to education. After 8 years at a university that is well known as an engineering school and makes the "U.S. News honor roll" among other places as one of the top engineering schools in the country, as I recall I had perhaps one or two "power" classes where basically we beat the power triangle to death, learned how to do conversions to impedances (complex numbers), and I think we might have covered the per-unit system. The basics of power systems analysis (short circuits, coordination, grounding) to say nothing of sequence current/voltage analysis were never covered. I recall some basics of safety but they pretty much had to do with low voltage and a little tiny amount about medium voltage, at a school which did some of the original research on insulator design!

From what I know of former interns and others I'm not alone. Power system analysis is simply not taught at the university level except as advanced classes or electives. I even happen to know that at the same university mentioned earlier there are in depth classes in National Electric Code taught in the "technician" (not ABET, EET programs) and extensive course work on Codes when it comes to civil engineering. So to put it mildly, basic knowledge of safety and requirements (Codes) is simply not taught in engineering school.

What I find interesting as the excuse for engineering schools at least is that electrical engineering is a very broad topic so their goal is to teach the first principles and that engineers are expected to learn the practical application side of things outside of school or in advanced/post graduate classes. The theory here is that for instance graduating VLSI engineers are going to go on to develop things like cell phones and watches and have very little if any exposure or need for such mundane topics as safety codes or analysis of "high power" equipment...whatever that means.

It goes further than this. Most engineers get a class in engineering ethics (I didn't have to have this one to graduate). By and large most do not get a class in quality control (that's a business school thing) and fewer yet apply it to anything outside of a basic model of producing widgets.

Besides which, putting both together, we're engineers. When we graduate, we can do anything and everything. We can build anything, and we can solve any problem that the world throws at us. Not only that but we're the best of the best, because we are told that over and over. We don't make mistakes. And after a few years, a couple tests, and paying some money, whatever words slip from our mouths are accepted as gospel in the court system as professional engineers, to say nothing of MBA's that worship the feet of an engineer that graces their presence as a consultant. Countless times throughout our careers we're called upon to come riding in on white horses and shining armor and save the day.

That's after redesigning the doorways to a 36" minimum width in order to accommodate the greatness of the ego that has to pass through it, cleaning the floors to avoid getting a speck of dirt on the penny loafers or wing tips, and having an intern or two polish the pocket protectors until you can see your reflection in the vinyl, and suffering a little grovelling before the idiot manager with a lowly 2 year business degree that approves doing the work.

So why would anyone assume that there is anything that an engineer might not know? Or that an engineer might actually make a mistake or be wrong?

I can't do much about the education situation (except educate the educators when the opportunity arises) but I can do something about the checking and I personally do have my work checked whenever possible even if I have to check it myself.


Top
 Profile Send private message  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 4 posts ] 

All times are UTC - 7 hours


Who is online

Users browsing this forum: No registered users and 4 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
© 2017 Arcflash Forum / Brainfiller, Inc. | P.O. Box 12024 | Scottsdale, AZ 85267 USA | 800-874-8883