It is currently Wed Sep 20, 2017 1:01 pm



Post new topic Reply to topic
Author Message
ekstra   ara
 Post subject: DC Arc Flash and Major Storage Batteries
PostPosted: Thu Dec 09, 2010 3:12 pm 
Offline

Joined: Thu Dec 09, 2010 2:30 pm
Posts: 2
We are in the process of designing and building a Hybrid power system consisting of Photovoltaics, diesel generators and a large (480 volt; 4800 AH) battery bank. Fault current for this battery is certainly high enough to warrant arc fault safety considerations, but I don't seem to be able to find appropriate literature on clacs for batteries. Can anybody point me in the right direction?


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Fri Dec 10, 2010 9:49 am 
Offline
Sparks Level
User avatar

Joined: Tue Oct 26, 2010 5:00 pm
Posts: 112
Location: Maple Valley, WA.
Unfortunately, the IEEE 1584 Standard performed experimentation and derived AF equations for only AC systems (50 & 60 Hz). We are hoping that they will be including this for DC systems in the future.

There may be others which have done experimentation, but currently I am not aware of any way to calculate the AF energy and boundary for DC systems.

_________________
Robert Fuhr, P.E.; P.Eng.
PowerStudies


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Fri Dec 10, 2010 11:22 am 
Offline

Joined: Thu Dec 09, 2010 2:30 pm
Posts: 2
SO, Are DC systems, and specifically batteries exempt from the requirements of codes and regualtions regarding arc flash? In so far as the available short circuit current is enormous with a battery of this size , it is clear that the safety hazards for personnel working on batteries are significant.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Fri Dec 10, 2010 4:42 pm 
Offline
Sparks Level
User avatar

Joined: Tue Oct 26, 2010 5:00 pm
Posts: 112
Location: Maple Valley, WA.
[quote="Calicojack"]SO, Are DC systems, and specifically batteries exempt from the requirements of codes and regualtions regarding arc flash? QUOTE]

No, I would not say that they are exempt from code requirements and regulations regarding AF. There just is not a tool in the tool box for us to calculate it. Everybody knows that there is a hazard but there is no way to know how much of a hazard is it.

Personally, I do not go into the large battery rooms without wearing a HRC 4 Blast Suit. I use to go in wearing a cotton t-shirt, but no longer.

_________________
Robert Fuhr, P.E.; P.Eng.
PowerStudies


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Fri Dec 10, 2010 10:28 pm 
Offline
Arc Level

Joined: Wed Jun 04, 2008 9:17 am
Posts: 409
Location: Spartanburg, South Carolina
There are methods available to calculate incident energy for dc faults. They are not covered by industry standards, however.

See IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 46, NO. 5, SEPTEMBER/OCTOBER 2010
DC-Arc Models and Incident-Energy Calculations
Ravel F. Ammerman, Senior Member, IEEE, Tammy Gammon, Senior Member, IEEE,
Pankaj K. Sen, Senior Member, IEEE, and John P. Nelson, Fellow, IEEE


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Sun Dec 12, 2010 7:32 am 
Offline
Plasma Level
User avatar

Joined: Mon Sep 17, 2007 5:00 pm
Posts: 1202
Location: Scottsdale, Arizona
DC Arc Flash Calculations

jghrist wrote:
There are methods available to calculate incident energy for dc faults. They are not covered by industry standards, however.

See IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 46, NO. 5, SEPTEMBER/OCTOBER 2010
DC-Arc Models and Incident-Energy Calculations
Ravel F. Ammerman, Senior Member, IEEE, Tammy Gammon, Senior Member, IEEE,
Pankaj K. Sen, Senior Member, IEEE, and John P. Nelson, Fellow, IEEE


DC Arc Flash Calculations
The referenced paper is the best source at the moment. It was presented in 2009 at the PCIC conference and contains a review of decades of DC arc flash research and theory. What is interesting is that most of the existing research and theory points towards the same conclusion. In addition, recent DC testing as part of the IEEE / NFPA collaborative effort also confirms much of what is already known.

With all of this information already available, when I heard “we have nothing on DC arc flash” for the zillionth time, I realized this is an incorrect statement. I began to sift through it all earlier this year and put together some worksheets, problems and included it in my arc flash calculation training class.

Some questioned why I pushed forward with it since none of this is an official standard. Because the AC arc flash calculations began the same way. There were papers and theory before the 2002 IEEE 1584 and companies that were ahead of the arc flash curve used what was available at the time.

After announcing that I put together the DC arc flash worksheets / problems / examples towards the first part of this year, I was immediately contacted by a couple of the major arc flash software companies. Since then, I have been involved with DC arc flash both domestically and overseas, with prototype projects, and with large traction power systems.

What is amazing is DC arc flash is really not that difficult. So, here are the basic steps based on some of the equations in the referenced paper.

1) Calculate bolted DC current
The first step like with any analysis is you need to develop a circuit model. This includes the DC voltage (VDC), and DC equivalent resistance (RDC) of the system. With this information VDC / RDC = ISC bolted where IDC bolted is the bolted short circuit current.

2) Use iteration to calculate DC arcing resistance and Iarc
Determine the DC arcing resistance is bit more cumbersome. The paper has an equation to calculate the DC arcing resistance but it is dependent on knowing the arcing short circuit current. This creates a dilemma since you can not calculate the arcing short circuit current without the arcing resistance and you can’t solve for the arcing resistance without the arcing short circuit current. This requires an iterative solution. As a first approximation / iteration, begin by assuming the arcing current is 50% of the bolted current and then solve for the DC resistance.

3) Determine the final answer through iteration
Once the first approximation of the DC arc resistance has been made, substitute it into the circuit and re-calculate for the arcing short circuit current. VDC / (RDC + Rarc). This produces a new Iarc. Re-calculate for a new RDC by using the new Iarc and continue until the answers converge (i.e. stop changing) This may take many iterations.

4) Calculate the power in the arc
Once the Rarc and Iarc are known, the power in the arc is I^2R where I = Iarc and R = Rarc.

5) Calculate the energy in the arc
The energy in the arc is Power x time so it is simply I^2R x Rarc x time in seconds. This produces watt-seconds which is the same as joules.

6) Calculate the incident energy
Next there are two equations depending on whether the arc is in a box or air. Spherical equations are used for air and alternate equations are used for the box.

The calculations are not at all difficult so I pushed what was available out onto “center stage” this year in hopes of moving this forward instead of “there is nothing available for DC”. I understand the 2012 NFPA 70E will include some info on approach limits and HRC's for DC arc flash as well.

One thing that is very different than AC calculations is many DC battery systems do not have protection. What do you use for the arc flash duration? You can plot graphs of incident energy vs. time and make a judgment on how much exposure you can have vs. the PPE that you would like to wear.

I hope this helps everyone understand DC arc flash calculations a bit better. Contact me if you have any questions.

_________________
Jim Phillips, P.E.
Brainfiller.com


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Dec 13, 2010 9:53 am 
Offline
Arc Level

Joined: Tue Jan 13, 2009 5:00 pm
Posts: 414
Jim,

Care to share that equation you refereed to in bullet 2?

Do you know if the arcadvisor calculator uses the same approach?

The maximum power transfer into the arc will occur when the arc resistance is equal to the source resistance, and the voltage drop across the arc is half the open circuit voltage. Seems like one could avoid the iteration in this manner. Would the results obtained this way be overly conservative?


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Dec 13, 2010 6:18 pm 
Offline
Plasma Level
User avatar

Joined: Mon Sep 17, 2007 5:00 pm
Posts: 1202
Location: Scottsdale, Arizona
stevenal wrote:
Jim,

Care to share that equation you refereed to in bullet 2?

Do you know if the arcadvisor calculator uses the same approach?

The maximum power transfer into the arc will occur when the arc resistance is equal to the source resistance, and the voltage drop across the arc is half the open circuit voltage. Seems like one could avoid the iteration in this manner. Would the results obtained this way be overly conservative?


The equation is:

Rarc = [20 + (0.534 * gap)] / (Iarc^0.88)

Rarc = ohms
gap = millimeters
Iarc = amps

I don't know what arcadvisor uses. I know a few of the major S/W vendors have been exploring this - as mentioned a few of them contacted me right away(one within and hour of us announcing what we did)

Yes the max power transfer uses 0.5 and this was used in a theoretical model published a few years ago. I found that with the above model, Iarc tends to be greater than 0.5 I bolted. I dont' know if this occurs for every case but for high current traction systems it has been my experience.

The paper is worth obtaining from IEEE.

_________________
Jim Phillips, P.E.
Brainfiller.com


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Tue Dec 21, 2010 6:12 pm 
Offline
Arc Level

Joined: Tue Jan 13, 2009 5:00 pm
Posts: 414
So I'm finding a maximum IE when the gap is zero (assuming negative gaps are not allowed).

Sound right?

Arcadvisor has very different results.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Thu Dec 23, 2010 8:44 am 
Offline
Plasma Level
User avatar

Joined: Mon Sep 17, 2007 5:00 pm
Posts: 1202
Location: Scottsdale, Arizona
stevenal wrote:
So I'm finding a maximum IE when the gap is zero (assuming negative gaps are not allowed).

Sound right?


In real world conditions, if the gap is zero then Rarc will be zero. This would result in Iarc^2 * Rarc becoming zero as well. i.e. no power in the arc and no energy from the arc.

However, mathematically, the equation has some limits. The arc resistance becomes quite non-linear at currents below 1,000 amps. The arc resistance is still not perfectly linear above 1000 amps but it is considered linear enough for a solution. The gaps for the equation range from 5 to 500 mm.

It's not perfect but AC arc flash calculations began in a similar way and continue to evolve. I'm trying to do my part to get the info out there so everyone can help move DC arc flash along. Good to see there is so much interest. :)

_________________
Jim Phillips, P.E.
Brainfiller.com


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Thu Dec 23, 2010 1:38 pm 
Offline
Sparks Level

Joined: Tue Feb 24, 2009 7:05 am
Posts: 252
stevenal wrote:
So I'm finding a maximum IE when the gap is zero (assuming negative gaps are not allowed).

Sound right?

Arcadvisor has very different results.


Same as AC, you can't know the maximum IE while looking only at the arc current or arc power, you need to also factor in the time-current curve of the OCPD which will stop the arc (not much applicable for battery cells and such though).

For solving only for Iarc, any iterative solver will do the job (Excel's Goal Seek or Solver functions will help you there, unless you like to put Newton-Raphson's method together by yourself). That'll give you quite easily Iarc, Rarc and Parc, but for the arc energy you need the proper TCC to get t from Iarc.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Thu Dec 23, 2010 4:32 pm 
Offline
Arc Level

Joined: Tue Jan 13, 2009 5:00 pm
Posts: 414
I simply checked enable iterative calculation in Excel, and all works fine. I used the 1584 2s to run away time and an 18" distance, and see that by adjusting the gap downward the IE always increases.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Thu Sep 22, 2011 6:08 am 
Offline
User avatar

Joined: Mon Sep 19, 2011 10:28 am
Posts: 8
Location: LLynchburg, VA
Additional DC equations reference

After several years of empirical studies in Toronto in which the arcing energy was measured from DC faults, there are some basic and conservative IE equations derived in the paper "Arc Flash Calculations for Exposures to DC Systems", authored by Daniel R. Doan of Dupont that was published in IEEE Transactions on Industry Applications, Vol. 46, No. 6, November/December 2010. This work is cited as a reference in the newly released NFPA 70E 2012 in the Informative Annex D.8.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Tue Oct 18, 2011 7:01 am 
Offline

Joined: Wed Sep 28, 2011 12:22 pm
Posts: 1
Hi

I have troubles figuring what the "gap" could be in the system I'm working on.
I have no troubles using my PPE but as other "newbies" will work on it with no knowledge of batteries, I'd like to have more information to convince them.

this is composed of several 48V batteries wired in series to voltages up to 400-500V.

the mechanical design of all parts has been made respecting clearance and creepage for more than 1000V


Image
is the gap:
  • A the distance at the breaker input
  • B the distance between terminals in the 2 modules I connect
  • C the distance between terminals


if A, the connection to the breaker is made first, without the battery connected
if B, gap gets from infinite to 0 when I make the connection
if C, it is several inches, gap is huge, no arc here...

the main risk identified is shorting 2 poles with a tool or a loose cable.
shorting current is somewhere in the 4kA in the "optimal" conditions but I would not define this as an arc current : it happens even on a single module and projects melted metal particles due to the poor contact on the short location and the high current. Anyway this only asks for FR cloting, face shield...

I can't find a gap in the design that would allow an arc to appear and be sustained.

if somebody could enlighten me on this one, I would appreciate.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Wed Oct 19, 2011 2:17 pm 
Offline

Joined: Tue Aug 14, 2007 6:25 pm
Posts: 7
In the absence or inability to come up with better data or calculations, this migt just be a good case to simply default to the new NFPA 70E DC tables for PPE on DC systems.


Top
 Profile Send private message  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 15 posts ] 

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