Arc Flash-flooded type U.P.S. Battery strings

[ghostbuster7]

These (non-fused) 1.67 volt cells can each produce 14,000 amps on short circuit.
There are over 1000 battery cells in this room that require regular mtce.
What P.P.E. (arc flash)clothing would you recommend.The current standards seem to have missed this area.

 

[dpc]

Yes, there are no recognized standards arc-flash protection for dc systems as far as I am aware.

 

[Zogzog]

First off, I think you are off on your SC calculations, where did you get those numbers?

dpc is right, nothing recognized yet for DC arc flash calcs, until then 70E is recomending using the AC equations.

 

[ghostbuster7]

Zog

According to C&D Tech sheets,and confirmed by their tech rep,the maximum short circuit current is 14,156 amps for model XT2LP-21 battery cell.

Question:

Is the dc voltage high enough on an individual cell to sustain an arc through air? If the answer is no, then maybe this is not an arc flash issue.We may only have to be concerned with higher dc string voltages.

 

[rbulsara]

As dpc said there are no published or accepted standards or requirements dealing with arc flash hazard for DC systems.

It does not mean that the hazard does not exist, but have not been evaluated to the extent of coming up with a standard.

 

[Zogzog]

SO internal battery impedance is 120 microhms per cell? I find that hard to believe.

Are you sure the fault current isnt 14kA for the whole string? How is the system configured?

 

[ghostbuster7]

Zog
As earlier stated, each battery cell is 14 ka. short circuit.Check with your C&D rep for any further info.

Each cell is series connected into a string.

415 volts per string.

5 strings to feed 5 U.P.S. systems.


The client insists on arc flash labels for this battery system.We will have to provide interim AC based arc flash category 4 labelling.This will be very awkward during regular battery mtce. programs.

 

[rbulsara]

I would not put AC based lables on battery system as they will be wrong. Intentionallly placing flase or wrong or misleading lables is a bigger liability and offence than not placing one. If the client insists, he or she needs be informed of the facts that it can not be done as there is no accepted method of analyzing DC arc flash incident energy levels.

 

[Zogzog]

So you have 3,517,808 of fault current from each string?

 

[ScottyUK]

How do you get to that conclusion? Fault current is a function of voltage and impedance (resistance). As the string increases in length the voltage and internal resistance increase proportionally; the fault level remains roughly constant for the bolted short circuit condition as long as the short itself has negligible impedance.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[rcw retired EE]

Zog - 14,000 amps is in the range that I have seen for maximum short circuit current from a large battery bank. Each cell can pass 14 kA through its internal impedance for a very short time. As I understand it, the same current flows through all cells so the bank fault current is the same 14 kA. Connection resistance will knock that value down.


What is not clear is how this can be translated to short circuit calculations for a whole bank or a partial bank. It doesn't seem to make sense that a short across one cell will deliver the same fault current as the entire string. Of course the fault energy will be greater for the higher voltage.

I agree with you that arc flash calculations cannot be done on DC in accordance with any standard.

 

[Zogzog]

"14,000 amps is in the range that I have seen for maximum short circuit current from a large battery bank."

Same here that is why I was asking, Of course I know there isnt 3 Million+ amps of fault current, but I also knwo you wont get 14kA from a single 1.5V cell.

The best way by far is to measure your battery system impedance, fault current is battery system voltage/Z. We did a ton of these tests using an EBITE test set.

Another rule of thumb to use is fault current is 20x 3 hr discharge rate.

 

[rob46]

http://www.arcadvisor.com/faq/dc_direct_current_arc.html

 

[ghostbuster7]

Zog
Enclosed is this battery cel spec. sheets.
Note:the published short circuit current rating on the 2nd page (14,156 amps-max.)

 

[DanEE]

We sell and have installed a number of GNB Absolyte 100A33 battery stacks.. Their short circuit current per the manufacturer data sheets is 11,000 amps. This is not the largest in the GNB Absolyte family, but the largest we like to deal with due to weight (800lbs/tier).

I don't have the data sheets at home, but my best recollection is that the internal resistance of these batteries when new are in the 160 micro ohm/per cell category.

So to the questioning poster on the impedance being in the 120 micro-ohm neighborhood, yes there are batteries in internal impedances in this range..

We use an Alber CellCorder to determine the condition of our customer's batteries which measures applies a load test to each cell and computes the "metallic resistance" in microhms..

Just got back tonight from a site in Charlotte,NC that uses a smaller model of the Absolytes.. 2 stacks paralleled and with about 10 years age on the batteries.. They tested out over a 300 - 550 micro ohm range per cell.

I have not seen any documentation or practice anywhere that interprets NFPA 70 arc flash assessment to cover these DC plants and batteries.

 

[stevenal]

Don't forget about the impedance of the inter-cell connections. String fault current should be less than that of each cell, and I expect it would decay pretty quickly. Until further guidance comes from IEEE, my recommendation is to wear natural fiber clothing under all the PPE required for protection from the acid.

 

[Zogzog]

"Don't forget about the impedance of the inter-cell connections. String fault current should be less than that of each cell, and I expect it would decay pretty quickly"

Right, my point exactly. The 14kA referenced * # of cells in the string isnt what you use for your analysis, it should be treated as a system not individual cells. Those values are for internal cell shorts.

 

[UPSguy]

on the data sheets first page it says that the calcualtions are based on using a string of 60 cells.

 

[peebee]

Zog -- the current for one cell will be about the same for the string, as a string is assembled in series (not parallel). This holds true for rated current as well as fault current. The voltage increases as you add cells to the string, but not the current.

Neglecting the impedance of intercell connections, the 14,000-amp rating for the XT battery is independant of the number of cells in a string.

However, if you have two strings in parallel, then the fault current will rougly double.