Crane Power Rail Arc Flash Calculation

[chelin]

Has anyone calculated or know how to calculate the arc flash levels for crane's power rails (i.e. mini-loaders or warehouse automated cranes).

My concern is that for a 800 ft long power rail, the arc flash level at the feeder connection point is not going to be the same as at the end of the rail due to difference in short circuit levels.

Then; Do I need one sticker for the whole rail? Can I calculate the arc flash using only the distance at the feeder connection? How do I model a rail in a program? (I use SKM)

 

[jghrist]

You need to calculate the incident energy at all possible operating conditions and label according to the worst case. This may be at a high current level feeder connection point or at a long clearing time end of rail point. It might even be somewhere in between if there is a point where the fault current is just lower than a fast trip setting on a breaker.

Unless you have a very good reason not to, I would use one label for the whole rail, showing the worst case.

 

[chelin]

Thanks 'jghrist' it makes sense... now how do I model the rail in a program? should I used a cable with the same ampacity and lenght? or is there any other way to model the rail?

 

[redfurry]

For modeling the rail, the key thing is the impedance. Ampacity may not be as useful.

Your best bet may be to find a conductor with matching material and cross-section.

 

[rbulsara]

This is just a personal opinion, I would treat the whole rail as a bus (equipment) and apply the lable for the worst case as jghrist said.

I presume the entire rail is energized (and de-energized) by a single breaker/device. If so, you cannot differentiate between one point from the other and should use the worst case. Impedance of the rail itself should not matter.



Rafiq Bulsara

http://www.srengineersct.com

 

[dpc]

You could model as bus duct with a few intermediate buses along the way to see what impact the bus impedance will have. Probably not much. This is how we have modeled plug-in feeder bus duct on previous projects.

A bigger decision is whether to model as in air or "in box". Modeling as "in box" will be conservative.

Other issues are adjusting the assumed arc gap to match the bus spacing and deciding on an assumed working distance.

Good luck.

 

[davidbeach]

There will be less fault current at the far end of an 800ft bus than at the source end. You need to at least look at both ends of the bus. If the hazard level/incident energy is roughly similar at both ends, then the worst end is what you should use. If they are significantly different, because one end trips instantaneously and the other doesn't, you should then try to figure out your hazard level/incident energy at the point were it just stops being instantaneous trip. I would not treat it as a single point.

 

[rbulsara]

davidbeach:
That is a good point. The worst case could be at the other end or somewhere in between. Now challange will be to figure out the impedance of the rail!





Rafiq Bulsara

http://www.srengineersct.com

 

[jghrist]

If the arcing current for a fault at the feeder connection point is in the fast trip part of the TCC and the current for a fault at the end of the rail is in the delayed part, then there will be one point along the rail where the current is just below the fast trip setting. You can use that current as well as the two ends to determine the worst case. It doesn't matter where on the rail this occurs.

 

[davidbeach]

Right, and you don't necessarily even have to find that point. You just need to evaluate it by inspection of the TCC.