Arc flash study

Status
Not open for further replies.
ggunn said:
How do you perform those calculations, and do you do them for every commercial PV installation? Do you generate those numbers for the entire electrical system or just for the PV installation?
Click to expand...
The right way to do it is to do the entire system, including the PV system, and run scenarios ranging from no Zero to 100% PV. If you're adding a 400 kW PV system, that'll contribute somewhere around 15 calories based on the numbers for a 400 kVA transformer, so that becomes significant. To Joethemechanic's point about maintenance folks not being trained from an arc flash standpoint, I say that varies greatly from plant to plant. I have experience with P&G, and I can say that they're programs are first rate and that all of their employees have PPE and wear it as appropriate. But the fact of the matter is that injuries from arc flash are few and far between, and 95% of electrical fatalities are from shock or contact with electricity. I think 70E goes to the extent it does to de-energize systems because that rare arc-flash injury can be so horrific. Take the Steve Lenz injury that was featured in the 2012 NFPA 70 (NEC) Handbook. He was hospitalized and had 5 skin peels before being released and all he was doing was installing a power monitor on the side of an 800-amp piece of gear, but a self-tapping screw pierced the insulation of a parallel 500 kCM conductor. It's those types of incidents that have driven 70E to be so seemingly paranoid. But as long as you wear PPE for the application, in accordance with table 130.5(C) Estimate of the Likelihood of Occurrence of an Arc Flash Incident, then you're covered.
 
mayanees said:
The right way to do it is to do the entire system, including the PV system, and run scenarios ranging from no Zero to 100% PV. If you're adding a 400 kW PV system, that'll contribute somewhere around 15 calories based on the numbers for a 400 kVA transformer, so that becomes significant. To Joethemechanic's point about maintenance folks not being trained from an arc flash standpoint, I say that varies greatly from plant to plant. I have experience with P&G, and I can say that they're programs are first rate and that all of their employees have PPE and wear it as appropriate. But the fact of the matter is that injuries from arc flash are few and far between, and 95% of electrical fatalities are from shock or contact with electricity. I think 70E goes to the extent it does to de-energize systems because that rare arc-flash injury can be so horrific. Take the Steve Lenz injury that was featured in the 2012 NFPA 70 (NEC) Handbook. He was hospitalized and had 5 skin peels before being released and all he was doing was installing a power monitor on the side of an 800-amp piece of gear, but a self-tapping screw pierced the insulation of a parallel 500 kCM conductor. It's those types of incidents that have driven 70E to be so seemingly paranoid. But as long as you wear PPE for the application, in accordance with table 130.5(C) Estimate of the Likelihood of Occurrence of an Arc Flash Incident, then you're covered.
Click to expand...
A PV inverter is typically not capable of providing more than about 110% of full load current, similar to the contribution from a VFD. That is nowhere near what a 400kVA transformer would contribute.
 
Related to this, they are also asking for fault current calculations. In the past I have taken the conservative approach of ensuring that kAIC and SCCR ratings of the PV equipment meet or exceed the AFC of the utility transformer, which is also admittedly the lazy approach, but I want to understand the calculation before I just hand it over to an on line calculator. In one paper it says "Find the conductor constant (C) from the table" but it doesn't have a table in it or say what table. In another there is a table that it just calls "Table 4" but it doesn't say where it comes from. A footnote says that C is equal to the reciprocal of the impedance per foot of the conductor, but using the Effective Z numbers from Table 9 the numbers don't match up. I assume that C means conductivity but there is no table in the NEC that looks like their "Table 4".

Should I just enter the numbers into an on line calculator and not worry my pretty little head about what is going on behind the curtain?
 
ggunn said:
Should I just enter the numbers into an on line calculator and not worry my pretty little head about what is going on behind the curtain?
Click to expand...
Pretty much don't sweat the details.
In particular that program you are asking about has been vetted through more than 50 years of use. It along, with many others, takes some industry acceptable shortcuts, but it is easy to use for service entrance labeling.

In the past I would use a spreadsheet based method that came from hand based sheets from back in the late 60's. It takes into account the X/R ratio which I kind of like.
 
jim dungar said:
A PV inverter is typically not capable of providing more than about 110% of full load current, similar to the contribution from a VFD. That is nowhere near what a 400kVA transformer would contribute.
Click to expand...
A 400 kVA transformer at unity power factor is equal to 400 kW. What am I missing? Is the power factor low for a PV system?
 
I'm wondering, will a grid tied inverter without any storage really contribute to the available fault current? How fast do they shut down? Do they shut down for under/over voltage and frequency? Is that a parameter you can set?
 
ggunn said:
Can you tell me how you would go about performing an AF study on, say, a 400kW AC PV system connected via a supply side interconnection on the line side lugs of a main service disconnect?
Click to expand...
We model the AC system in SKM or etap based on the design drawings and shop drawings for the actual equipment, which would be all the new equipment from the POI to the inverter AC side. We also put in the existing service SWBD where the POI is located. The existing service SWBD gets an arc fault sticker but in this scenario, it's the only existing equipment that would. We run a short circuit study and then an arc fault study. The arc fault module provides a PDF of the arc fault labels that we provide the client, the client has them printed on label material and applies the labels to the equipment.
If we do a DC study we model the DC system from the strings to the inverter and get a DC arc fault label for the inverter and for the DC combiner if a separate enclosure.
We use SKM and etap because our power studies also include the short circuit study, OCPD coordination, and equipment AIC/SCCR evaluation.
 
The problem with using any software is garbage in - garbage out. They are not magic and if we don't provide good input then the arc fault label may be dangerously inaccurate. It can be difficult to get the utility contribution from the utility and the OCPD part numbers from the client. Pressured by time people are tempted to use "typical" values that actually have no connection to the specific project.
 
Status
Not open for further replies.
Top