arc flash

Status
Not open for further replies.
B

bsh

Senior Member
IEEE standard 1584 deals wirh arc flash and states that a 208 volt system fed by a transformer smaller than 125 KVA "generally" is not a problem. I have done a calculation on a system fed by a 75 KVA transformer where the upstream 480 volt breaker will take nearly half a minute to open based on a sustained arc at the 208 volt equipment. I believe that the arc will not sustain itself for this length of time but I am concerned about the energy delivered by the arc until it clears. Does anyone have any experience with arcing faults in 208 volt systems?
 
IEEE says that a 208V arc fed by a trabsformer <125kVA will not self sustain, I have lab tests that prove otherwise. When we do an arc flash study we give the customer an option of doing the analysis below this point or using the tables to determine the HRC below this point.

AS far as your long clearing time goes, use a max of 2 seconds. 2 seconds is the time to react to the arc and move away from it. Of course if the panel is on an area that would be hard to move away from then you should use your actual clearing time.
 
I think the length of time that the cable will actually be intact, e.g. burns through, becomes a factor on such a long clearing time. Once the cable has disconnected itself, the current flow will certainly be interrupted.

I believe the statistics will tell you that the most injuries occur at the 480V level, then medium voltage. 240V and 208V, certainly will create a large spark, bang, and probably require some pants changing, but not cause an accident where unrecoverable burns occur.
 
I was just curious if other assumptions can be made about the clearing time. If an arc flash creates an explosion with enough force to throw a person's body are you able to use less than 2 seconds in the energy calc?
 
Kingpb,
Unfortunatley, arc "blast" i.e. the throwing motion, is not factored into the arc flash (energy release)
Click to expand...
You are correct, however several have done tests "in the box" and published their findings. Out of the box at 22,480 amps available the pressure wave at 18" is 2500 lbs/sq. ft.
Mr Bill,
was just curious if other assumptions can be made about the clearing time
Click to expand...
While this subject has been discussed before on this forum, I assumed that a75KVA transformers is capable of delvering 15,000 + amps fault current, at 2 seconds the Ford Visteon Arc Flash Calculator said level 4.
Just my $.02
 
davidr43229 said:
While this subject has been discussed before on this forum, I assumed that a75KVA transformers is capable of delvering 15,000 + amps fault current, at 2 seconds the Ford Visteon Arc Flash Calculator said level 4.
Click to expand...

For a 208Y/120V 75kVA unit to deliver 15,000 SCA it would have to have an impedance of 1.4% and be fed from an infinite bus. This %Z is an unrealistic value, especially for the common 480-208Y/120V dry-type design. Utilities often use this low of a %Z when designing systems so that their customer's service equipment still has the correct SCCR after a potential utility system upgrade.

Arc flash calculations should never be performed solely on the basis of a utility's design fault current. Unrealistically high fault values will often give unrealistic device clearing times. I wish there was a good way to predict the worst case arc fault incident energy, but for now the only solution appears to be to perform several calculations using different fault levels.
 
Davidr: 15kA of fault current does not translate to 15kA of arc fault current for developing the incident energy level.

Also, a properly sized protective device will de-energize the circuit long before 2 secs (120 cycles). I think it is fundamentally incorrect to assume that the protective devices are going to fail, and not open the circuit quickly. This is getting into contingencies, which certainly can be designed for, but need to be based on an Owners perference and directive.

Then if that's the case, what level of contingency do you want, 1st order, 2nd order, 3rd order? At some point you have to be realistic, and that point is usually financially based.
 
"Ford Visteon Arc Flash Calculator said level 4"

Either that program has problems or it is being used wrong at several plants. I have been in several Ford plants with HRC 5 listed in some substations. I have also seen a 480V main breaker fed from a 3000kVA transformer w/out an inst trip labeled as HRC 1.
 
Kingpb,
The origional poster posted:
where the upstream 480 volt breaker will take nearly half a minute to open based on a sustained arc at the 208 volt equipment
Click to expand...
I was merely following his origional inquiry.
and:
Davidr: 15kA of fault current does not translate to 15kA of arc fault current for developing the incident energy level.
Click to expand...
It does directly equate to the amount of Fault current available at the secondary side of the 75KVA transformer. I mean to say even if it was 12,000 amps, that is the figure you use to calculate.

Jim Dungar,
Even with 12,000 amps available for 2 seconds, what do you calculate at 18"?
I would be curious.
Dave

edited to add at 18"
 
I can force ETAP to give me those same results, however I believe there are some assumptions being made, that should not be made for the analysis.

Each case needs to be modeled with the correct data, this includes cable, protective device curves to get tripping time, actual fault currents (high and low). All of these parameters can have an effect on the outcome.

Using different numbers, and assumptions, I was able to simulate a large swing in calories/cm^2, causing the PPE level to swing from over level 4, to level 0.

I think this is a good case in point were it is vital to use real numbers, and include all pertinent equipment. I have to wonder how many studies are being done inaccurately, thus giving erroneous results.
 
Mr. Bill said:
If an arc flash creates an explosion with enough force to throw a person's body are you able to use less than 2 seconds in the energy calc?
Click to expand...
Seriously? :confused:

"That's okay, I won't get burned because the explosion will throw me clear."

Yikes! :roll:
 
davidr43229 said:
Even with 12,000 amps available for 2 seconds, what do you calculate at 18"
Click to expand...

Dave,
We should get the same answer if we are both using the same IEEE 1584 equation.

My point is that a "design" level fault current should not be the only value examined. Additional calculations should be made using more realistic fault currents. For example, I cannot imagine what conditions would exist where a 480-208Y/120V dry type transformer would have an infinite primary bus and 1.4%Z, so it would not be good engineering to assume it does.
 
I agree with Jim on the assumptions, there are too many variables that can sway it one way or the other and you need to look at each case independently.

With 12kA of fault current on the secondary side of the transformer, no cable, sustained fault for 2 secs; using IEEE 1584 methods, the incident energy I calculated at 18" is 36.34cal/cm^2.

If you add 25' of #2/0 cable, the incident energy is 30.57cal/cm^2

By allowing the breaker to trip according to it's normal trip curve (Cutler Hammer HFD, 125A, rated for 65kA at 480V), with no cable the incident energy is 0.35cal/cm^2.

Next I am interested to see what the outcome will be by using an adjustable trip protective device.
 
Last edited:
DC Arc Flash??

DC Arc Flash??

Could somopne point me in the right direction as far as finding the arc flash protection requirements for a DC system. I work in a plant that has DC cranes that are fed by a rectifier system and trying to get arc flash protection and labels for this equipment. Thanks for your help.
 
There are not currently any DC arc flash calculations available, SKm has a DC S/C module but no DC arc flash. I spoke to a 1584 member recently about this and they said lab testing and developing DC arc flash equations is at the top of the 1584 groups priority list.

I was also told that the current stance is that given the same S/C current and clearing times the Ei produced by a DC arc flash would be less than that of an AC arc so the recommendation is to use the AC equations for now.

Personally I dont agree with this stance based on "playing" with DC and AC hipots in my shop, but then again, my experiments had alot of unknowns so I guess I have to believe the 1584 group until further testing is done on DC arcs.
 
Status
Not open for further replies.
Top