It is currently Mon Feb 19, 2018 2:54 am



Post new topic Reply to topic Go to page 1, 2  Next
Author Message
ekstra   ara
 Post subject: Automatic Cat 0 for 480/240V OCD under 100A
PostPosted: Wed Sep 11, 2013 6:49 pm 
Offline
Arc Level

Joined: Thu Jan 10, 2008 8:49 pm
Posts: 483
Location: New England
I have quotes from two different vendors who perform arc flash analysis and both are stating that feeders of 480 or 240V with OCD of 100A or less can be assigned an automatic Category 0 rating. I've asked both for reference but none is produced. But it is also strange that two unrelated companies would be proposing the same concept. Is this founded anywhere in IEEE or NFPA or some other analysis method.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Thu Sep 12, 2013 4:31 pm 
Offline
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 1877
Location: North Carolina
Assuming you are referring to a molded case breaker and most of those at that size are 1-2 cycle devices for instantaneous trips, and it is rare to find more than 10-15x the long term trip setting (tap) so efectively this is almost alwaysan instantaneous trip case, then you need a very high short circuit current to reach 1.2 cal in such a short period of time. I use 50 A as a cutpff. Theoretically it depends on the arcing current and to a lesser extent, voltage, and if the arcing current is low enough, trip times could get long. But then you hit the 2 second cutoff.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Fri Sep 13, 2013 6:13 am 
Offline
Arc Level

Joined: Thu Jan 10, 2008 8:49 pm
Posts: 483
Location: New England
My point is that an 'assumption' at whatever size, should not be made. The primary basis for IE is going to be arcing current and clearing time. Without performing the fault analysis and examining the breaker's trip curve, there is no assurance that the breaker will be operating in its Instantaneous trip range. Fault currents can fall off quickly on smaller gauge long feeders. Arcing current is always less than fault current, and we also have examine the trip time at 85% of arcing current. There is a very high probability that the resulting value will 'NOT' be in the instantaneous range of the trip curve, and consequently IE values can be higher than 1.2 cals. I don't know how this practice got started but I don't agree with it.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Fri Sep 13, 2013 2:22 pm 
Offline
Plasma Level
User avatar

Joined: Wed May 07, 2008 5:00 pm
Posts: 711
Location: Rutland, VT
I agree with haze10. There is currently nothing in any standard or consensus document that would support those statements. I would be wary of that. I will do an analysis done to where I start seeing <1.2 cal/cm^2. Then, if one desires, a sensitivity test can be done on those panels. What I mean by that is taking stock of what the bkr sizes are in that panel and run some scenarios where you have 50, 100, 150 ft conductor runs and see if the <1.2 cal/cm^2 holds up. Then I think you safely say anything fed from that panel is a HRC 0.

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


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Sep 16, 2013 7:34 am 
Offline
Sparks Level

Joined: Fri Apr 15, 2011 7:43 am
Posts: 160
Location: Colorado
We we actually presented with the idea of limiting smaller sized breakers (90A and less) to a category 0. I had one my younger engineers prove this correct or incorrect. The answer was it is correct provided the available fault current is adequate. What does adequate mean? That is dependent on the breaker, some breakers require less available and others more.

So the answer is - unless you KNOW you have adequate fault current you must run the study


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Sep 16, 2013 8:02 am 
Offline

Joined: Wed Dec 02, 2009 4:46 pm
Posts: 29
Location: CT
haze10 wrote:
I have quotes from two different vendors who perform arc flash analysis and both are stating that feeders of 480 or 240V with OCD of 100A or less can be assigned an automatic Category 0 rating. I've asked both for reference but none is produced. But it is also strange that two unrelated companies would be proposing the same concept. Is this founded anywhere in IEEE or NFPA or some other analysis method.

At some fault current point smaller molded case CB, particularly if defined as current limiting, will limit incident energy very well. Even devices as large as 600A. However, it is very dependent on the fault current being sufficient to engage the circuit breaker's instantaneous, probaly the range where mechanical foces are accelerating the mechanism. That is easier the smaller the device. So the comments that state that it depends on arcing current versus curve are correct. The expected arcing current must be enough to engage the device's fast instantaneous operation. If you consult with the manufacturer of the circuit breakers, particularly for recently provided devices they may have data. In tests that I am familiar with, < 1.2 calories with devices up to 600A "if" the prospective arcing current is "high" enough was achieved. Note that enclosure and arcing gap are also a factor. Even if a determination is made that incident energy may be as low as 1.2 cal I certainly would not advise working without some level of face and hand protection and probably more than 1.2 cal rating for clothing/PPE.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Sep 16, 2013 9:37 am 
Offline

Joined: Mon Feb 22, 2010 10:24 am
Posts: 21
Marcelo wrote:
At some fault current point smaller molded case CB, particularly if defined as current limiting, will limit incident energy very well. Even devices as large as 600A. However, it is very dependent on the fault current being sufficient to engage the circuit breaker's instantaneous, probaly the range where mechanical foces are accelerating the mechanism. That is easier the smaller the device. So the comments that state that it depends on arcing current versus curve are correct. The expected arcing current must be enough to engage the device's fast instantaneous operation. If you consult with the manufacturer of the circuit breakers, particularly for recently provided devices they may have data. In tests that I am familiar with, < 1.2 calories with devices up to 600A "if" the prospective arcing current is "high" enough was achieved. Note that enclosure and arcing gap are also a factor. Even if a determination is made that incident energy may be as low as 1.2 cal I certainly would not advise working without some level of face and hand protection and probably more than 1.2 cal rating for clothing/PPE.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Sep 16, 2013 9:40 am 
Offline

Joined: Mon Feb 22, 2010 10:24 am
Posts: 21
This is a dangerous assumption. We always perform the calculations and finding incident energy above 1.2 cal at 30 amps and higher. As stated by others, it is highly dependent on the arcing current. You cannot make an assumpton based on the clearing time alone.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Sep 16, 2013 4:20 pm 
Offline
Arc Level

Joined: Thu Jan 10, 2008 8:49 pm
Posts: 483
Location: New England
I just want to also point out that 'current limiting' breakers are rather rare in industry. Their cost is five times that of conventional thermal/magnetic breakers. While they can serve a purpose, the assumption by vendors that breakers are current-limiting would not be a fair assumption. Note the reverse is true for fuses, for which current-limiting is quite common.

Here is another angle to consider. Art 130 permits recognized analysis methods. NFPA does NOT state that you have to follow IEEE. But we almost all do follow IEEE, and from my experience so does all the popular software. So if the 'method' does not provide an explicit exception, say like IEEE 1584 did when it said analysis was not require on system under 240V and 125KVA - then a vendor can not inject their own exceptions and still claim to be following the method. That's why I asked in my original theme - if anyone knew of some lesser known but scientific method that permitted this exception.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Tue Sep 17, 2013 4:26 am 
Offline
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 1877
Location: North Carolina
There are two ways that you get less than 1.2 cals. First at a given voltage if the arcing current is not sufficient to generate enough heat with a given working distance then the overcurrent device does not matter. Second is if it trips.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Tue Sep 17, 2013 6:47 am 
Offline
Arc Level

Joined: Wed Jun 04, 2008 9:17 am
Posts: 411
Location: Spartanburg, South Carolina
I have tried to establish a general rule-of-thumb based on the length of a circuit. The idea being that for a given breaker size and available fault current, assume NEC minimum size cable, and calculate the length of cable where the IE = 1.2 cal/cm² based on the equation from IEEE Std 1584 Table E.1. Circuit length more than this require analysis.

What I found was that the effort is not worth it. You need to know the fault level at the feeder breaker and the length of the circuit. If you know these, you might as well enter it into the software program (we use SKM) and calculate the IE at the end of the feeder anyway.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Tue Sep 17, 2013 11:38 am 
Offline

Joined: Thu Jul 21, 2011 4:21 pm
Posts: 2
I agree with haze10 and jghrist. No assumptions should be made and they aren't necessary. More importantly, the engineer performing the study should do so in accordance with 70E which does not include any short cuts, rules of thumb, etc. You either perform a complete analysis or you use the HRC tables. By the time you confirm that the installation meets a particular set of criteria (as with the HRC tables), you could have collected the data, modeled the installation, run the analysis and labeled the equipment. If you don't model the installation, are the labels generic?


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Thu Sep 19, 2013 6:53 am 
Offline
Sparks Level
User avatar

Joined: Tue Apr 17, 2012 8:19 am
Posts: 240
Location: Charlotte, NC
haze10 wrote:
I have quotes from two different vendors who perform arc flash analysis and both are stating that feeders of 480 or 240V with OCD of 100A or less can be assigned an automatic Category 0 rating. I've asked both for reference but none is produced. But it is also strange that two unrelated companies would be proposing the same concept. Is this founded anywhere in IEEE or NFPA or some other analysis method.


I have an Arc Flash Evaluation where a 230V bus duct has an Incident Energy of 14 cal/cm^2. I don't know how you could catagorically state anything about a feeder for a certainty without doing an evaluation.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Fri Oct 04, 2013 2:03 pm 
Offline
User avatar

Joined: Sun Jul 13, 2008 5:00 pm
Posts: 24
PCE ATL wrote:
By the time you confirm that the installation meets a particular set of criteria (as with the HRC tables), you could have collected the data, modeled the installation, run the analysis and labeled the equipment.
This is something I have never understood. PCE ATL is absolutely correct. The tables have so many limitations that by the time you find everything you need to verify you are within the limits of the tables, why not just run the calculations and be done with it. And if you use the tables, you have to put the highest HRC on the label which often leads to Cat 4. Really? Perform the calculations.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Sat Oct 05, 2013 3:53 am 
Offline
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 1877
Location: North Carolina
Under 70E-2012, it is helpful to take advantage of the reduced PPE (risk) scenarios where they exist because the text states to do only a hazard analysis and leaves you without options for risk analysis. Not all such strategies are always recommended. Second pray tell how do you use an analysis for over 15 kv? The tables go to 38 kv. 1584 recommends a procedure which grossly overestimates the incident energy. This leaves tables in 70E, tables in NESC, or ArcPro.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Oct 07, 2013 2:18 pm 
Offline
Sparks Level

Joined: Sat Nov 21, 2009 10:23 pm
Posts: 73
Location: Ohio
No disrespect, however, if you were to do 10,000 calculations on 30 amp circuits, you probably would not find one one that exceeded 1.2 cal/sq cm at 18". The exception would be 480-120/208 transformer secondaries and those can also be modeled. It is virtually impossible to see greater than 1.2 cal/sq cm at 18" on small branch circuits. I have investigated over 20 arc flash accidents and have yet to find one on a small branch circuit, most have been service entrance or large feeders.


i have seen plants labeled all the way down to 20A and in my opinion that creates a larger problem then labeling those circuits that need to be labeled. If I see labels every where, eventually I do not fear the label because they are so common. I actually did an investigation that turned into a civil case where over labeling led to an arc flash accident.


rdj wrote:
This is a dangerous assumption. We always perform the calculations and finding incident energy above 1.2 cal at 30 amps and higher. As stated by others, it is highly dependent on the arcing current. You cannot make an assumpton based on the clearing time alone.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Tue Oct 08, 2013 9:35 pm 
Offline
Sparks Level

Joined: Sat Apr 04, 2009 9:50 pm
Posts: 89
Location: San Antonio, TX
I would like to add my thoughts and experience regarding this subject.

After performing many (hundreds) of AFHA's I have found very few instances where a circuit of less than 100A @ 480V result on a AFIE of more than 1.2 Cal/cm2.

I have done an extensive research to corroborate this field observation.

Using SKM I created many scenarios, with short circuit from 5k to 100k, with Square D CB's and Busman fuses from 90A to 30A, with lengths starting at 1 foot to 900 feet (increasing wire sizes according to voltage drop), and the result of this investigation corroborated my field experience.

Only under two situations with circuit breakers (not fuses) I found AFIE larger than 1.2cal/cm2.

First, in circuits longer than 250 feet the arcing fault current is reduced to a value such that the circuit breakers open slow enough to have AFIE larger than 1.2 Cal/cm2. But not much larger! The great majority of branch circuits less than 100A are of less than 250 feet.

Second, very short circuits, less than 15 feet, with available fault currents of more than 35kA will result in AFIE larger than 1.2cal/cm2. But nom much higher than 5 cal/cm2. The great majority of circuits of less than 100A rarely are connected with the parameters mentioned before.

So, there is a valid question that all engineers performing arc flash studies should answer. Can you justify to the client the substantially greater cost (only the survey time will increase about 3 to 4 times more) of including in the survey and calculations circuits of 480V and less than 100A?

If you are concern about missing those few circuits of less than 100A (at 480V) that result in AFIE larger than 100A, I understand your feeling.

Therefore, I developed a procedure that calculates the worst case AFIE that these smaller circuits fed by a panel could deliver, and I print as many labels as these smaller than 100A circuits are present in the panel. I give these labels to the client to affix after giving a training on how to do so. Needles to say that more than 95% of all these labels show the worst-case AFIE of much lower than 1.2 cal/cm2.

And circuits of less than 100A AND less than 15 feet AND connected to service panels (which have a greater possibility of having available faults of 35kA or more) I survey, calculate, and print equipment specific labels.

It is NOT a perfect system. You might point out some cases where this system will fail and report a lower level of AFIE than if you choose a different more "exact" method.

I can do the same with ANY method!

But I believe that this procedure is a reasonable balance between accuracy and cost.

I also believe that nobody will ever find the "real", "exact" AFIE at a specific equipment or location. The calculation system has too many variable, the scope of survey has too many variables, the field data has too many variables, and, more important that all of the preceding, there are as many opinions as engineers regarding the priority and impact of these different variables on the result of the study.

The search for the "exact" or most "accurate" result can drive the cost of the project to an unreasonable figure that will be difficult to justify and still might be questionable that the results are more "accurate" than a simplified method.

If you are wondering why I have included in this discussion circuits of less than 100A @ ONLY 480V is that I have not yet researched the 208V case.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Wed Oct 09, 2013 1:24 pm 
Offline
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 1877
Location: North Carolina
Given the IEEE 1584 normalized calc, lg IE=K+0.662 (I)+0.0966 V+0.000526G+0.5588V*lg I-0.00304G*lg I, we can immediately see that given the same range of currents, lower voltage decreases inicident energy compared to the 480 volt case. Gaps in 208 volt gear are less but not drastically so. Breakers and fuses with the same current should trip at the same or faster rates so your analysis still holds. I should add that operating 480 volt systems with cable lengths as excessive as you are describing also leads to an issue that voltage drop in both normal and in fault conditions becomes excessive and it becomes difficult to trip on ground fault using simple overcurrent (50/51) breakers, and impedance on conduit in particular if used as a ground path becomes unacceptable. This starts pushing into lsg or lsig breakers which are pretty expensive. So there are very practical reasons why very long circuits are problematic from a safety point of view.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Thu Oct 10, 2013 3:15 pm 
Offline
Sparks Level

Joined: Sat Nov 21, 2009 10:23 pm
Posts: 73
Location: Ohio
I agree 100% and that is very close to how I handle studies and the smaller circuits.I believe I am accurate in saying that the the IEEE 1584 group never planned on using the calc's for smaller circuits, they were for panel level only. Instead of worrying about low energy circutis, smaller sources, such as 500 kVA service transformers. concern me, they are hard to mitigate,

To go along with your your premise below, I was working with an electrician on a sub metering panel. The voltage input to the meters was 480V and was fed by a 3P/20A circuit breaker that was no more than 20 ft from the service transformer which was a 2500 kVA. The fault current was in the 45 kA range, the electrician created a arc flash and the 20A breaker did its job and no one was hurt even though we were no more than 24" from the arc flash. There was a blinding flash and a large sound. If if were protected by a fuse we probably would not have seen anything.

i want someone who takes the calc's down to 30A to give me a documentable example of a serious burn on a 30A/480V circuit. I can give you over 20 examples of massive burns on large circuits that have personally investigated. I can also point to two deaths out of those twenty, again, on large circuits. It is a personal opinion, however, I feel it is far more important to spend the time on mitigation techniques, not the 30A circuits.

If we really want to complicate all these issues, read Mike Lang's papers on horizontal bus faults, this makes all past calculations wrong. Lastly, be careful what current you are using in your study, momentary is probably the incorrect assumption if your ocd is operating at greater then 6 cycles.


RECS wrote:
I would like to add my thoughts and experience regarding this subject.

After performing many (hundreds) of AFHA's I have found very few instances where a circuit of less than 100A @ 480V result on a AFIE of more than 1.2 Cal/cm2.

I have done an extensive research to corroborate this field observation.

Using SKM I created many scenarios, with short circuit from 5k to 100k, with Square D CB's and Busman fuses from 90A to 30A, with lengths starting at 1 foot to 900 feet (increasing wire sizes according to voltage drop), and the result of this investigation corroborated my field experience.

Only under two situations with circuit breakers (not fuses) I found AFIE larger than 1.2cal/cm2.

First, in circuits longer than 250 feet the arcing fault current is reduced to a value such that the circuit breakers open slow enough to have AFIE larger than 1.2 Cal/cm2. But not much larger! The great majority of branch circuits less than 100A are of less than 250 feet.

Second, very short circuits, less than 15 feet, with available fault currents of more than 35kA will result in AFIE larger than 1.2cal/cm2. But nom much higher than 5 cal/cm2. The great majority of circuits of less than 100A rarely are connected with the parameters mentioned before.

So, there is a valid question that all engineers performing arc flash studies should answer. Can you justify to the client the substantially greater cost (only the survey time will increase about 3 to 4 times more) of including in the survey and calculations circuits of 480V and less than 100A?

If you are concern about missing those few circuits of less than 100A (at 480V) that result in AFIE larger than 100A, I understand your feeling.

Therefore, I developed a procedure that calculates the worst case AFIE that these smaller circuits fed by a panel could deliver, and I print as many labels as these smaller than 100A circuits are present in the panel. I give these labels to the client to affix after giving a training on how to do so. Needles to say that more than 95% of all these labels show the worst-case AFIE of much lower than 1.2 cal/cm2.

And circuits of less than 100A AND less than 15 feet AND connected to service panels (which have a greater possibility of having available faults of 35kA or more) I survey, calculate, and print equipment specific labels.

It is NOT a perfect system. You might point out some cases where this system will fail and report a lower level of AFIE than if you choose a different more "exact" method.

I can do the same with ANY method!

But I believe that this procedure is a reasonable balance between accuracy and cost.

I also believe that nobody will ever find the "real", "exact" AFIE at a specific equipment or location. The calculation system has too many variable, the scope of survey has too many variables, the field data has too many variables, and, more important that all of the preceding, there are as many opinions as engineers regarding the priority and impact of these different variables on the result of the study.

The search for the "exact" or most "accurate" result can drive the cost of the project to an unreasonable figure that will be difficult to justify and still might be questionable that the results are more "accurate" than a simplified method.

If you are wondering why I have included in this discussion circuits of less than 100A @ ONLY 480V is that I have not yet researched the 208V case.


rdj said: [url='http://arcflashforum.brainfiller.com/goto/post?id=13868#post-13868']↑[/url]
[INDENT=1]This is a dangerous assumption. We always perform the calculations and finding incident energy above 1.2 cal at 30 amps and higher. As stated by others, it is highly dependent on the arcing current. You cannot make an assumpton based on the clearing time alone.[/INDENT]
- See more at: [url='http://arcflashforum.brainfiller.com/threads/2978/?utm_source=Brainfiller+Arc+Flash+Members&utm_campaign=1580b724c8-Arc+Flash+Forum+February+27%2C+2012&utm_medium=email&utm_term=0_4ef962b869-1580b724c8-20746977#sthash.ipqELCra.dpuf']http://arcflashforum.brainfiller.com/threads/2978/?utm_source=Brainfiller Arc Flash Members&utm_campaign=1580b724c8-Arc Flash Forum February 27, 2012&utm_medium=email&utm_term=0_4ef962b869-1580b724c8-20746977#sthash.ipqELCra.dpuf[/url]
RECS wrote:
I would like to add my thoughts and experience regarding this subject.

After performing many (hundreds) of AFHA's I have found very few instances where a circuit of less than 100A @ 480V result on a AFIE of more than 1.2 Cal/cm2.

I have done an extensive research to corroborate this field observation.

Using SKM I created many scenarios, with short circuit from 5k to 100k, with Square D CB's and Busman fuses from 90A to 30A, with lengths starting at 1 foot to 900 feet (increasing wire sizes according to voltage drop), and the result of this investigation corroborated my field experience.

Only under two situations with circuit breakers (not fuses) I found AFIE larger than 1.2cal/cm2.

First, in circuits longer than 250 feet the arcing fault current is reduced to a value such that the circuit breakers open slow enough to have AFIE larger than 1.2 Cal/cm2. But not much larger! The great majority of branch circuits less than 100A are of less than 250 feet.

Second, very short circuits, less than 15 feet, with available fault currents of more than 35kA will result in AFIE larger than 1.2cal/cm2. But nom much higher than 5 cal/cm2. The great majority of circuits of less than 100A rarely are connected with the parameters mentioned before.

So, there is a valid question that all engineers performing arc flash studies should answer. Can you justify to the client the substantially greater cost (only the survey time will increase about 3 to 4 times more) of including in the survey and calculations circuits of 480V and less than 100A?

If you are concern about missing those few circuits of less than 100A (at 480V) that result in AFIE larger than 100A, I understand your feeling.

Therefore, I developed a procedure that calculates the worst case AFIE that these smaller circuits fed by a panel could deliver, and I print as many labels as these smaller than 100A circuits are present in the panel. I give these labels to the client to affix after giving a training on how to do so. Needles to say that more than 95% of all these labels show the worst-case AFIE of much lower than 1.2 cal/cm2.

And circuits of less than 100A AND less than 15 feet AND connected to service panels (which have a greater possibility of having available faults of 35kA or more) I survey, calculate, and print equipment specific labels.

It is NOT a perfect system. You might point out some cases where this system will fail and report a lower level of AFIE than if you choose a different more "exact" method.

I can do the same with ANY method!

But I believe that this procedure is a reasonable balance between accuracy and cost.

I also believe that nobody will ever find the "real", "exact" AFIE at a specific equipment or location. The calculation system has too many variable, the scope of survey has too many variables, the field data has too many variables, and, more important that all of the preceding, there are as many opinions as engineers regarding the priority and impact of these different variables on the result of the study.

The search for the "exact" or most "accurate" result can drive the cost of the project to an unreasonable figure that will be difficult to justify and still might be questionable that the results are more "accurate" than a simplified method.

If you are wondering why I have included in this discussion circuits of less than 100A @ ONLY 480V is that I have not yet researched the 208V case.

RECS wrote:
I would like to add my thoughts and experience regarding this subject.

After performing many (hundreds) of AFHA's I have found very few instances where a circuit of less than 100A @ 480V result on a AFIE of more than 1.2 Cal/cm2.

I have done an extensive research to corroborate this field observation.

Using SKM I created many scenarios, with short circuit from 5k to 100k, with Square D CB's and Busman fuses from 90A to 30A, with lengths starting at 1 foot to 900 feet (increasing wire sizes according to voltage drop), and the result of this investigation corroborated my field experience.

Only under two situations with circuit breakers (not fuses) I found AFIE larger than 1.2cal/cm2.

First, in circuits longer than 250 feet the arcing fault current is reduced to a value such that the circuit breakers open slow enough to have AFIE larger than 1.2 Cal/cm2. But not much larger! The great majority of branch circuits less than 100A are of less than 250 feet.

Second, very short circuits, less than 15 feet, with available fault currents of more than 35kA will result in AFIE larger than 1.2cal/cm2. But nom much higher than 5 cal/cm2. The great majority of circuits of less than 100A rarely are connected with the parameters mentioned before.

So, there is a valid question that all engineers performing arc flash studies should answer. Can you justify to the client the substantially greater cost (only the survey time will increase about 3 to 4 times more) of including in the survey and calculations circuits of 480V and less than 100A?

If you are concern about missing those few circuits of less than 100A (at 480V) that result in AFIE larger than 100A, I understand your feeling.

Therefore, I developed a procedure that calculates the worst case AFIE that these smaller circuits fed by a panel could deliver, and I print as many labels as these smaller than 100A circuits are present in the panel. I give these labels to the client to affix after giving a training on how to do so. Needles to say that more than 95% of all these labels show the worst-case AFIE of much lower than 1.2 cal/cm2.

And circuits of less than 100A AND less than 15 feet AND connected to service panels (which have a greater possibility of having available faults of 35kA or more) I survey, calculate, and print equipment specific labels.

It is NOT a perfect system. You might point out some cases where this system will fail and report a lower level of AFIE than if you choose a different more "exact" method.

I can do the same with ANY method!

But I believe that this procedure is a reasonable balance between accuracy and cost.

I also believe that nobody will ever find the "real", "exact" AFIE at a specific equipment or location. The calculation system has too many variable, the scope of survey has too many variables, the field data has too many variables, and, more important that all of the preceding, there are as many opinions as engineers regarding the priority and impact of these different variables on the result of the study.

The search for the "exact" or most "accurate" result can drive the cost of the project to an unreasonable figure that will be difficult to justify and still might be questionable that the results are more "accurate" than a simplified method.

If you are wondering why I have included in this discussion circuits of less than 100A @ ONLY 480V is that I have not yet researched the 208V case.


rdj said: [url='http://arcflashforum.brainfiller.com/goto/post?id=13868#post-13868']↑[/url]
[INDENT=1]This is a dangerous assumption. We always perform the calculations and finding incident energy above 1.2 cal at 30 amps and higher. As stated by others, it is highly dependent on the arcing current. You cannot make an assumpton based on the clearing time alone.[/INDENT]
- See more at: [url='http://arcflashforum.brainfiller.com/threads/2978/?utm_source=Brainfiller+Arc+Flash+Members&utm_campaign=1580b724c8-Arc+Flash+Forum+February+27%2C+2012&utm_medium=email&utm_term=0_4ef962b869-1580b724c8-20746977#sthash.ipqELCra.dpuf']http://arcflashforum.brainfiller.com/threads/2978/?utm_source=Brainfiller Arc Flash Members&utm_campaign=1580b724c8-Arc Flash Forum February 27, 2012&utm_medium=email&utm_term=0_4ef962b869-1580b724c8-20746977#sthash.ipqELCra.dpuf[/url]
Flash wrote:
No disrespect, however, if you were to do 10,000 calculations on 30 amp circuits, you probably would not find one one that exceeded 1.2 cal/sq cm at 18". The exception would be 480-120/208 transformer secondaries and those can also be modeled. It is virtually impossible to see greater than 1.2 cal/sq cm at 18" on small branch circuits. I have investigated over 20 arc flash accidents and have yet to find one on a small branch circuit, most have been service entrance or large feeders.

i have seen plants labeled all the way down to 20A and in my opinion that creates a larger problem then labeling those circuits that need to be labeled. If I see labels every where, eventually I do not fear the label because they are so common. I actually did an investigation that turned into a civil case where over labeling led to an arc flash accident.


PaulEngr wrote:
Given the IEEE 1584 normalized calc, lg IE=K+0.662 (I)+0.0966 V+0.000526G+0.5588V*lg I-0.00304G*lg I, we can immediately see that given the same range of currents, lower voltage decreases inicident energy compared to the 480 volt case. Gaps in 208 volt gear are less but not drastically so. Breakers and fuses with the same current should trip at the same or faster rates so your analysis still holds. I should add that operating 480 volt systems with cable lengths as excessive as you are describing also leads to an issue that voltage drop in both normal and in fault conditions becomes excessive and it becomes difficult to trip on ground fault using simple overcurrent (50/51) breakers, and impedance on conduit in particular if used as a ground path becomes unacceptable. This starts pushing into lsg or lsig breakers which are pretty expensive. So there are very practical reasons why very long circuits are problematic from a safety point of view.


rdj said: [url='http://arcflashforum.brainfiller.com/goto/post?id=13868#post-13868']↑[/url]
[INDENT=1]This is a dangerous assumption. We always perform the calculations and finding incident energy above 1.2 cal at 30 amps and higher. As stated by others, it is highly dependent on the arcing current. You cannot make an assumpton based on the clearing time alone.[/INDENT]
- See more at: [url='http://arcflashforum.brainfiller.com/threads/2978/?utm_source=Brainfiller+Arc+Flash+Members&utm_campaign=1580b724c8-Arc+Flash+Forum+February+27%2C+2012&utm_medium=email&utm_term=0_4ef962b869-1580b724c8-20746977#sthash.ipqELCra.dpuf']http://arcflashforum.brainfiller.com/threads/2978/?utm_source=Brainfiller Arc Flash Members&utm_campaign=1580b724c8-Arc Flash Forum February 27, 2012&utm_medium=email&utm_term=0_4ef962b869-1580b724c8-20746977#sthash.ipqELCra.dpuf[/url]


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Oct 14, 2013 7:48 am 
Offline
Sparks Level

Joined: Thu Oct 18, 2007 7:10 pm
Posts: 250
Location: NW USA
Very good discussion (with no 100% clear conclusion though many valid observations)

I would only add that many molded case circuit breakers seem to clear instantaneous based on frame size and not breaker rating, so a small breaker rating might be misleading.


Top
 Profile Send private message  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 23 posts ]  Go to page 1, 2  Next

All times are UTC - 7 hours


Who is online

Users browsing this forum: No registered users and 1 guest


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