Author 
Message 
mnewman

Post subject: Modelling HRC fuses for arc flash analysis Posted: Thu Mar 26, 2015 1:29 am 

Joined: Sun Jul 22, 2007 5:00 pm Posts: 35 Location: Stratham, SW Australia

IEEE15842002 Section 4.6 Step 5 states that if only fuse average melt time is published, add 15% up to 0.03 sec and 10% above 0.03 sec to determine total clearing time.
Where:  Average melt time is equal to prearcing time.  Total clearing time is equal to prearcing time plus arcing time.
Some LV fuse manufacturers publish a +/ 10% current tolerance for their prearcing time curve. The above IEEE15842002 Section 4.6 Step 5 does not take this current tolerance into account. The effect of i2t due to d.c. and a.c. current decrements is also not mentioned in IEEE 15842002.
What is the established practice when modelling HRC fuses for arc flash analysis to IEEE 15842002 ?


Top 


john.a

Post subject: Re: Modelling HRC fuses for arc flash analysis Posted: Sun Mar 29, 2015 11:19 pm 

Joined: Sun Dec 21, 2014 6:44 pm Posts: 4

You correctly note that IEEE15842002 Section 4.6 Step 5 states that if only fuse average melt time is published, add 15% up to 0.03 sec and 10% above 0.03 sec to determine total clearing time.
What I do not understand is this, " Should the +15% and +10% quantities be percentages of time or current?" If they are percentages of time then there is a peculiar phenomenon either side of 0.03 sec. At 0.299 sec melting time the total clearing time is 0.299 * 1.15 = 0.344 sec; At 0.301 sec melting time the total clearing time is 0.299 * 1.15 = 0.331 sec, slower than the above total clearing time. If the +15% and +10% quantities are percentages of current then this phenomenon is eliminated.
Can any readers clarify this quandary?


Top 


Jim Phillips (brainfiller)

Post subject: Re: Modelling HRC fuses for arc flash analysis Posted: Mon Mar 30, 2015 3:55 pm 

Plasma Level 

Joined: Mon Sep 17, 2007 5:00 pm Posts: 1637 Location: Scottsdale, Arizona

Good catch. This is actually something I have been discussing with a few people. What some software companies are doing is using a time of 0.01 seconds for CL fuses in the current limiting region since the time current curve "buries" itself into the horizontal axis and it's difficult to know what the actual time would be.
The discussion was similar to your question, if 0.01 is used, does a margin need added on top of it. I think right now the answer is no in this region. Mostly because a CL fuse is required to begin clearing the fault before the first 1/4 cycle and it must clear the fault before the first half cycle both of which are below 0.01 seconds.
However, in the noncurrent limiting region you use the TCC and you are correct, depending on whether it is by time or by current, you can get completely different results. It would be best to consult the fuse manufacturer. I can't imagine any fuse manufacturer that would not be able to give better guidance. I'm sure this will be sorted out in the next edition of IEEE 1584.


Top 


arcad

Post subject: Re: Modelling HRC fuses for arc flash analysis Posted: Wed Apr 01, 2015 8:39 am 

Sparks Level 

Joined: Wed Jun 10, 2009 5:00 pm Posts: 267 Location: Toronto

john.a wrote: You correctly note that IEEE15842002 Section 4.6 Step 5 states that if only fuse average melt time is published, add 15% up to 0.03 sec and 10% above 0.03 sec to determine total clearing time.
What I do not understand is this, " Should the +15% and +10% quantities be percentages of time or current?" The +15% and +10% quantities are indeed the percentages of time. The IEEE 1584 4.6 Step 5 reads "For fuses, the manufacturer's timecurrent curves may include both melting and clearing time. If so, use the clearing time. If they show only the average melt time, add to that time 15%, up to 0.03 seconds, and 10% above 0.03 seconds to determine total clearing time. If the arcing fault current is above the total clearing time at the bottom of the curve (0.01 seconds), use 0.01 seconds for the time". It is the established practice when modelling HRC fuses for arc flash analysis to IEEE 15842002. Also, the opening time tolerances differ between fuse designs and the percentage of the overload relative to the amp rating. Incrementing the time according to IEEE 1584 guidelines should effectively compensate for the margin of error in published timecurrent curves in most cases. Arc Flash Analytic (AFA V5.0) software program automatically applies the required 15% and 10% time increments per IEEE 1584 when calculating incident energy and arc flash boundary. The program also provides for adding fuses and circuit breaker data including timecurrent characteristics for the devices that are not already featured in the builtin protection device library. This feature allows factoring in the actual timecurrent curve margin of error and/or temperature and altitude derating factors on discretion of the designer.
Last edited by wbd on Wed Apr 01, 2015 9:29 am, edited 1 time in total. 
Deleted link to commercial website for download of software 


Top 


mnewman

Post subject: Re: Modelling HRC fuses for arc flash analysis Posted: Fri Apr 10, 2015 12:14 am 

Joined: Sun Jul 22, 2007 5:00 pm Posts: 35 Location: Stratham, SW Australia

1. Fuse manufacturers have informed that a +/ 10% current tolerance applies to their fuse average melt curve ( perarc curve ).
2. IEEE15842002 Section 4.6 Step 5 states that if only fuse average melt time is published, add 15% up to 0.03 sec and 10% above 0.03 sec to determine total clearing time.
Is the following method correct for modelling fuses used in arcflash analysis to IEEE 15842002 ?
Step 1. Add 10% current to the timecurrent coordinates of the fuse average melt curve.
Step 2. On the curve produced by Step 1. add 15% time up to 0.03 sec, and add 10% time for times above 0.03 sec


Top 


mnewman

Post subject: Re: Modelling HRC fuses for arc flash analysis Posted: Thu Apr 16, 2015 7:30 pm 

Joined: Sun Jul 22, 2007 5:00 pm Posts: 35 Location: Stratham, SW Australia

The 2011 Amendment 2 of IEEE 1584, Clause 4.6 Step 5 states : “If the (fuse) curve has only the average melt time, add 10% plus 0.004 seconds to that time to determine total clearing time. If the total clearing time at the arcing fault current is less than 10 milliseconds, use 0.01 seconds for the time.”
Should the +10% quantity that must be added to the average melt curve be a percentage of time or a percentage of current? I ask this question because European and North American fuse manufacturers have informed that a +/ 10% current tolerance applies to their published prearcing ( melt curves ).


Top 


Jim Phillips (brainfiller)

Post subject: Re: Modelling HRC fuses for arc flash analysis Posted: Mon Apr 27, 2015 12:20 pm 

Plasma Level 

Joined: Mon Sep 17, 2007 5:00 pm Posts: 1637 Location: Scottsdale, Arizona

I have seen a few different interpretation so I checked with a colleague at Mersen  (fuse manufacturer) and here is the response:
The curves are +/ current, usually 10% but there are some variations in the percentage especially at the current extremes (curve time over 2 minutes or close to 0.01 seconds).
Hope this helps.


Top 


john.a

Post subject: Re: Modelling HRC fuses for arc flash analysis Posted: Tue Jun 02, 2015 12:23 am 

Joined: Sun Dec 21, 2014 6:44 pm Posts: 4

Can readers please advise how they apply Clause 4.6 Step 5 of IEEE1584 2nd Amendment of 2011 that specifies, "add +10% plus 0.004 seconds to that time" ( That time being the average melt time)?
Do you add the 10% to the average melt curve in the direction of current or in the direction of time?
Do any readers add 10% in the direction of current to create the maximum melt time curve then apply the following approximation over the entire current range: total clearing time = maximum melt time, neglecting the 0.004 seconds that the 2nd Amendment calls for?
I can confirm Jim Phillip's advice that Mersen, along with Littelfuse, Bussmann, Siba, GE and Comeca advise a tolerance of +/ 10% of the average melt time curve 'in the direction of current'.
The attached curve exhibits the differences between the fuse characteristics obtained as follows: 1. +10% in the direction of current /_ 10% in the direction of current (dark green curve), 2. +10% in the direction of time plus 0.004 seconds/ 10% in the direction of current (red curve) & 3. average melt time curve (lime green curve).
You do not have the required permissions to view the files attached to this post.


Top 



Page 1 of 1

[ 8 posts ] 


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

