Coordination and arc flash studies estimated values

[deltawhy]

Hello everyone,
I have been doing coordination and arc flash studies for low voltage commercial and industrial distribution systems. So far, I have had no problems in acquiring the necessary information to do the studies (grid MVA, X/R, line voltage, R0/1/2, X0/1/2). However, I have come across a project where the utility will not provide the information I require no matter how many ways I ask. Has anyone experiences this, and if so, how do you go about circumventing this? Would it be appropriate to estimate values based on info I may find in the buff book? This is the only option I can think of.

Thanks

 

[eeprom]

Usually the utility will give you the info. Obviously not always. If you are doing low voltage, then you must have a transformer on site. Could you assume a infinite bus upstream of the transformer and use the short circuit through the impedance of the transformer? That may be conservative, but I think it is fair to assume that the service transformer is most likely to be the greatest impedance in the system.

EE

 

[deltawhy]

Thanks for your response EE. I have assumed an infinite bus ahead of the utility transformer. The problem is that I cannot acquire the X/R value at the connection point. The reason being the utility grid in that area "changes" so often that it negates giving out the info. Thus I am left with the choice of guessing a conservative X/R value. The problem is, a conservative guess will increase my asymm fault levels and give completely wrong values for arc flash. Any thoughts?

Thanks again

 

[rcw retired EE]

My understanding of arc flash calculations is that we should calculate the maximum and the minimum fault levels. Maximum gives the worst case arc flash for maximum current. Minimum fault level calculations give arc flash level associated with the longer clearing times of lower fault currents.

I haven't checked the numbers, but my gut feeling is your error in estimating X/R will be in the same range as the error in estimating only the maximum fault current.

 

[dpc]

The X/R ratio will be dominated by the service transformer, so I think you can approximate that. There are some tables that estimate the distribution feeder X/R based on the approximate distance from the substation. If your only concern are the arc-flash levels on the secondary of the transformer, I think these X/R approximations will be close enough. Since the arc-flash calcs are based on the three-phase symmetrical bolted fault current (assuming IEEE 1584 equations), the X/R ratio is not of prime importance for arc-flash calculations anyways. It is more critical for determining adequacy of circuit breaker interrupting ratings.

I would be more concerned about getting the actual impedance of the utility service transformer.

 

[healyx]

I agree with dpc - the feed transformer usually makes the inclusion of the upstream network almost negligible. For max fault level assume infinite bus upstream. I'm from the IEC world so I'm not sure how this translates but for a 1000kVA transformer it would be reasonable to assume an X/R ratio of about 3-3.5:1. If it is bigger increase X/R, if it is smaller decrease X/R. For the network, assuming 250MVA with a 10:1 X/R ratio is reasonable for the minimum fault level. You will want to get the impedance % of the transformer - it will have a large effect on the calculation.

 

[deltawhy]

Excellent, thanks a lot for the info.