It is currently Wed Aug 05, 2020 6:57 am



Post new topic Reply to topic
Author Message
ekstra   ara
 Post subject: Arc Flash Calculation with Generator
PostPosted: Thu Nov 16, 2017 7:30 pm 
Sparks Level

Joined: Tue Apr 03, 2012 1:28 pm
Posts: 53
Location: Louisville KY
Has any one ever seen modeling of a backup/emergency generator where the arc flash engineering firm used 1000 seconds on the calculation for the gen output to the main gen output breaker? Here is the engineer's reply, based in Singapore:

"SKM Model is based on IEEE 1584 std. The calculation for the generator trip time of 1000s is based on the scenario that the generator internal protection is disabled.

We have generator alternator protection setting for short circuit shutdown at 300% with 9s time delay. Based on Generator Thermal Damage curve, it indicated that generator would only be able to sustain the excitation current for 300% for 10s. Therefore, worst case scenario for calculation of Arc Flash incident energy at the Generator line side, 10s is to be selected."

After this, the arc flash label, that was previously labeled at 7.3 Million calories/cm2, was changed to 600 cal/cm2.

This firm does a lot of calculations for US-based companies in Singapore, Malaysia, Taiwan, etc.

Thoughts?


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Arc Flash Calculation with Generator
PostPosted: Fri Nov 17, 2017 6:30 am 
Plasma Level
User avatar

Joined: Wed May 07, 2008 5:00 pm
Posts: 822
Location: Rutland, VT
I have not seen 1000 secs used for anything. I know that 1000s is the default setting in EasyPower for a cutoff time so maybe SKM uses the same time constraint. What I typically use is 2 sec for the generator as they are usually located outside or in large rooms where someone could get away in the event of a fault.

What size generator is it?

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


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Arc Flash Calculation with Generator
PostPosted: Sat Nov 18, 2017 7:30 am 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 2174
Location: North Carolina
IEEE 1584 specifies a cutoff of 2 seconds unless escape from the work area is impeded. That drops it from 600 cal/cm2 down to 120 cal/cm2...I'm not sure if Oberon makes an arc flash suit that crazy high or not. Nobody else does.

However let's take this a step further and show where the real problem lies to avoid similar issues throughout the report. The professional engineering company is responsible for protecting themselves from liability (law suits). The problem is in the problem statement itself. The professional engineer should state up front that it will be assumed the all protective devices are functioning properly and properly maintained. This is also a requirement in NFPA 70E. If you're not going to maintain your equipment properly in the first place then 70E doesn't apply and neither does IEEE 1584. So don't bother with all those expensive engineering studies because it's a total waste of time and money and a professional engineer that is following the ethical requirements of the PE license should know that and be up front enough to state that and turn away work in that scenario.

Which brings me to the point here. It is on the customer, the plant, to provide the ground rules for the professional engineer. These can and should be mutually agreed upon. As a service provider it is unrealistic to assume that the customer has the knowledge to make an informed decision in most cases so part of the preliminary meetings is to educate the customer in the process of arriving at a mutually agreed upon set of assumptions. In fact I would go so far as to say that a professional engineer should spell them out for the client be sure that the client understands the implications. Then this should be stated again that certain assumptions or "givens" are made in the report and a "don't sue me" clause inserted explaining that deviating from those assumptions means that the report is no longer valid.

That being said some customers are going to be idiots. Particularly operations managers, safety managers, and legal advisors are going to want to set up the assumptions to eliminate any resemblance to liability for themselves and the consequential business decisions are going to be very expensive in terms of making life extremely difficult for maintenance and operations personnel. And the resulting "cheating" that will go on to get around the unrealistic arc flash requirements that ensue all but guarantee that their worst nightmare is that much more likely to come true because plant personnel are basically going to ignore the unrealistic rules that have been created in the interest of protecting liability and not maintaining equipment to minimize "downtime".

The problem here is that some professional engineers take this a step too far and start trying to play god. They believe that they alone are the saviors of humanity and that it is their personal responsibility to protect plant personnel from themselves. They produce engineering studies with no assumptions whatsoever such as assuming that the generator protection is disabled and in so doing come up with all kinds of crazy unrealistic conditions for everyone while doing all the training and other things required. The end result is that credibility for both themselves and the profession is destroyed and millions of cal/cm2 is the garbage result that you get.

Some places where this happens:
1. Generator protective settings in particular. Usually this should simply be part of mitigation particularly if there isn't much protection. Some of the "math" behind it also gets silly/unrealistic when you look at for instance maximum mechanical power input limitations even with "no protective devices".
2. Watch out around main-tie-mains. Assuming all 3 are closed or even that only one main and the tie are closed are guaranteed to produce crazy results.
3. Unrealistic working distances, particularly with medium voltage or utility-style equipment which tends to be odd sizes and shapes.
4. Unrealistic gap distances.
5. Unrealistic utility sources. Often more of a utility generated garbage-in/out problem.
6. Not following the 2 second rule.
7. Unrealistic settings for asymmetrical currents, particularly a problem with ETAP.
8. No cable impedances...although this tends to be the opposite problem (too low of incident energy).
9. Assuming no protective devices or that all protective device settings are at their maximum settings.
10. Using Lee above 15 kV in particular, or really anywhere.
11. Using a calculation below 250-300 V without consideration for arc stability.
12. Crazy "maintenance switch" type scenarios such as assuming protective devices during maintenance operations are set to their lowest setting without consideration for whether or not this will cause nuisance tripping.

I've collected all these over the years particularly from U.S. national engineering firms such as BE&K, Halliburton, etc. This kind of stupidity is rampant in the consulting engineering industry because they think they are playing god and that the customer is a bunch of illiterate red necks who need a professional engineer to save them from every conceivable safety hazard, both real and imaginary. In their defense if you ever do the surveys and equipment walk-throughs and troubleshooting that I've done over time, you will no doubt reach a point where it is very easy to generalize and begin to think of all plant personnel, particularly plant maintenance people, with a very poor view of their capabilities. But as a consummate defender of the lowly maintenance man everywhere, I don't share that view.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Arc Flash Calculation with Generator
PostPosted: Sat Nov 18, 2017 9:53 am 
Plasma Level
User avatar

Joined: Mon Sep 17, 2007 5:00 pm
Posts: 1498
Location: Scottsdale, Arizona
I have seen this and know of a company that used this number.

The history - Liability plain and simple. Years back when arc flash studies were first gaining traction, many were afraid to use the 2 second cut off since they did not want to be the first. I heard all kinds of "excuses" what if the person trips, falls, gets knocked unconscious, has a heart attack etc. etc. Finally, I just stated to someone at this company "Afraid of getting sued?" a sheepish - Yes was the answer - they would not use 2 seconds.

Some time later they began using 1000 seconds. It was pretty much useless as you pointed out and I asked them where it came from. The response was that 1000 seconds was the top of a published time-current curve. I think I must have given them a blank look (deer in the headlights look). They said that although they can not justify 2 seconds from a liability issue, they can justify 1000 seconds since "nothing is published" above that value.

In all fairness, these were the early years and companies were waiting for "someone else" to make the first move regarding interpretations.

That was actually one of the reasons I started the arc flash forum back around 10 years ago and started the Question of the Week. I realized there were many thinking they were the only one and this all helped people realize they were not the only one.

btw, Singapore? Nicest airport in the world! I'm presently in Brussels.

Just curious - how old is the study? i.e. is it older or are people still using the 1000 second approach.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Arc Flash Calculation with Generator
PostPosted: Sat Nov 18, 2017 2:18 pm 
Sparks Level

Joined: Tue Apr 03, 2012 1:28 pm
Posts: 53
Location: Louisville KY
Jim, the study is only 1.5 years ago. The local firm that did the study was quick to change it once questioned and republished it with 2 seconds. It’s still high as Paul stated, but not ridiculous. This room has no escape issues - wide open room with two exits. The problem with the study was that this 1000 seconds appears 6 tines in various scenarios. I must say, I’ve never seen SKM output a 7 million calorie output, but I’ve never put in 1000 seconds! HaHa!


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Arc Flash Calculation with Generator
PostPosted: Sat Nov 18, 2017 2:24 pm 
Sparks Level

Joined: Tue Apr 03, 2012 1:28 pm
Posts: 53
Location: Louisville KY
Paul,

Couldn’t of said it better if I tried! Luckily the staff here does impeccable maintenance and has great record keeping, as I’ve seen in many Asian countries. They do annual C/B current injection testing for reliability reasons. The room in question had very low risk of not being able to escape, this no reason for the ridiculously high time input. As I stated to Jim, I’ve never seen 7 million calories before! The worse I saw before was China, with a calculation of 600 cal, and that was due to the nearest 22kV utility fuse over three long city blocks away with NO local fusing on the input side of the transformer.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Arc Flash Calculation with Generator
PostPosted: Sat Nov 18, 2017 2:26 pm 
Sparks Level

Joined: Tue Apr 03, 2012 1:28 pm
Posts: 53
Location: Louisville KY
Jim, I totally agree on the nicest airport comment. The staff is wonderful, the security agents professional, and the amenities are absolutely the best!

Enjoy Brussels!


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Arc Flash Calculation with Generator
PostPosted: Sat Nov 18, 2017 3:11 pm 
Plasma Level
User avatar

Joined: Mon Sep 17, 2007 5:00 pm
Posts: 1498
Location: Scottsdale, Arizona
1.5 years is pretty recent for using 1000 seconds.

Yes the 1000 seconds is based on the top of the TCC with the justification that nothing is defined beyond that time.

Paul has added lot of great insight from a wealth of obvious experience regarding his comments.

I had a friend that tracked the really large numbers years ago and I believe he had an arc flash boundary of something like 1/2 mile - not sure what the incident energy was but 7 Million cal/cm2?

It does still seem like something might be off. (7,000,000 cal/cm2 / 1000 seconds / 60 cycles/second) = 116.7 cal/cm2 per cycle.

Safe travels!


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Arc Flash Calculation with Generator
PostPosted: Sun Nov 19, 2017 11:07 am 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 2174
Location: North Carolina
BE&K generated a 3500 foot arc flash boundary on a 22 kV enclosed circuit breaker based on using Lee equations because that's what SKM spit out, using all the ties closed with two 100 MVA incoming circuit breakers and 1 54 MW generator although I think the really crazy one only occurs with both incoming utility lines out of service. Never mind that this trips the whole plant out (they need at least one to operate) and that all mains and ties are not all closed simultaneously. I won't even bother to mention the working distance they used which is physically impossible to get that close, nor the silly results for overhead line workers based on I think a 15" working distance when all overhead work is via 10 foot hot stick, or the same kind of 36" working distances when operating ground mounted operators on overhead load break switches and cutouts that are mounted a minimum of 12 feet overhead in the same system. Oh, and the 3500 foot mark probably places you outside of the prefabricated substation's 12 foot width (widest that could be shipped over the highway without turning it into a specially permitted wide load). So any arc flash boundaries beyond say 50 feet long (about the longest you can get onto a long low boy trailer) is probably a crap result.

It didn't help that during the first round they simply ignored cable impedance and did the entire arc flash study based on transformer impedance only, with obviously very wrong results and so everyone had even more disbelief when this updated report came out with really crazy high numbers and obviously incorrect assumptions to the point where across the site many individual plants (several plants on one site) simply kept using the old report.

On the same result/report they also reported that the overhead cabling was underdutied based on short circuit too of course. And there was tons more idiocy where this came from.

As a customer I highly recommend asking for a printout with the incident energy values sorted in descending order. Then review the top ones and look carefully for problems and errors like this. Even if you aren't an "arc flash expert", look at the result and where it is and why it is the way it is. Ask questions. If it looks wrong, it probably is. It is very common to find cases where the first overcurrent protective device downstream of a transformer has crazy high incident energy and in many cases this is simply unavoidable...it's something that needs to be part of your planning when performing maintenance on those pieces of equipment. But elsewhere if the value seems crazy ridiculous, it almost always is or it is pointing to a fat finger error, or it is pointing to a real problem that should be addressed. This is called mitigation, something else that engineering firms don't do as a general rule.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Arc Flash Calculation with Generator
PostPosted: Mon Nov 20, 2017 8:55 am 
Sparks Level

Joined: Tue Apr 03, 2012 1:28 pm
Posts: 53
Location: Louisville KY
It's really sad to see results like this, Paul. I saw a steel mill in Tennessee with a 1.2 MILE arc flash boundary!!! WTH! Who puts this stuff out???

I spent time explaining to the local staff why a 7 milllion calorie value was ridiculous, and that the engineering firm should have caught this on their review PRIOR to releasing it to the customer. Funny enough, the eng firm, after some "feedback" from the audit team, "re-ran" the results in SKM and suddenly these 1000 second values did not re-appear. Magically, they just "disappeared." There was no mention of an error on their part, of course, for liability reasons, I'm sure.

Can you say "Caught with your pants down?"


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Arc Flash Calculation with Generator
PostPosted: Mon Nov 20, 2017 9:03 am 
Sparks Level

Joined: Tue Apr 03, 2012 1:28 pm
Posts: 53
Location: Louisville KY
Jim Phillips (brainfiller) wrote:
1.5 years is pretty recent for using 1000 seconds.

Yes the 1000 seconds is based on the top of the TCC with the justification that nothing is defined beyond that time.

Paul has added lot of great insight from a wealth of obvious experience regarding his comments.

I had a friend that tracked the really large numbers years ago and I believe he had an arc flash boundary of something like 1/2 mile - not sure what the incident energy was but 7 Million cal/cm2?

It does still seem like something might be off. (7,000,000 cal/cm2 / 1000 seconds / 60 cycles/second) = 116.7 cal/cm2 per cycle.

Safe travels!


I agree, Jim, on the new results being questionable at best. I think that the local firm is convinced to NEVER use this engineering firm again. They have proven their own incompetency, in my opinion. They are a Swedish-based firm and their website brags that they do "international arc flash studies for American firms abroad." Hmmm....

Got back home last evening after a 32 hour fly day. These trips to Asia are brutal! Now for time zone and jet lag recovery!!!

Thanks for your input - most appreciated.

Ken


Top
 Profile Send private message  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 11 posts ] 

All times are UTC - 7 hours


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:  
© 2019 Arcflash Forum / Brainfiller, Inc. | P.O. Box 12024 | Scottsdale, AZ 85267 USA | 800-874-8883