It is currently Fri Aug 14, 2020 1:24 am



Post new topic Reply to topic
Author Message
ekstra   ara
 Post subject: Utility Fault Data on Primary
PostPosted: Fri Dec 07, 2012 11:16 am 
Sparks Level

Joined: Fri Apr 15, 2011 7:43 am
Posts: 177
Location: Colorado
It is not unusual to get utility fault data on the secondary side of the transformer. Many times this is based on infinite bus. AND it flat wrong especially when considering arc flash but it the best we can get. This causes the first disconnecting means to be dangerous! If we can get the fuse that protects the utility transformer we can, at the very least, get a reasonable arc flash category for the first disconnecting means (based on utility info). The problem is we cannot enter infinite bus into our software.

It is reasonable to estimate the high side utility system as 1000 greater to get nearly the same secondary fault current? Example - We have 12.74kV/480, 500KVA, 5% utility XFMR and are given secondary fault current as 12kA (infinite bus). We are also given the utility fuse at 12.47. Is it reasonable to say the utility source is 500MVA at 12.47 so we can model the transformer and high side fuse?

My thinking is 1Vpu/(Zsys+Zxfmr) changes only slightly compared to 1Vpu/Zxfmr

It is hard to do arc flash without the utilities on board!

Toughs/Comments?


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Sat Dec 08, 2012 8:43 am 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 2174
Location: North Carolina
If it is fused then it is likely that the maximum fuse size is a 200E fuse. SMU-20 style fuse packages are available from multiple vendors. I Anything larger is rare. For 15kv and higher voltage classes only one vendor I know of makes larger fuses, and their lead times are high for these (standard quote is 8 weeks for fuse or holders). I could be wrong but a set of fuses and holders and hardware above 200E is costing me about $12,000 while a recloser with everyone option runs $20,000 installed prices, never mind the differences in operating costs and the degree of protection you get with the recloser even if you don't recloser.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Dec 10, 2012 9:45 am 
Arc Level

Joined: Wed Jun 04, 2008 9:17 am
Posts: 428
Location: Spartanburg, South Carolina
500 MVA is over 20 kA at 12.47 kV. This is higher than what you would normally find on a 12.47 kV system and could be safely used as a maximum fault current. For a minimum, I'd use 2 kA if I couldn't get the actual value.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Dec 10, 2012 12:40 pm 

Joined: Thu Aug 30, 2012 8:04 am
Posts: 2
Regarding the advice from jghrist, 2kA as a total fault current on the primary side of a utility transformer may be reasonable for a rural system, but is too low for a suburban or urban system or industrial park. I'm a utility engineer. Our utility has a fault current range of 2kA to 18 kA at 13.8 kV. This depends on many factors, such as the substation transformer size and impedance; whether the substation operates with a closed main-tie-main breaker arrangement or open tie; whether the service transformer is physically near the substation or distant from it; and so on. In response to engrick, the effect of the primary side impedance is overwhelmed by the service transformer impedance. These two issues taken together might lead you to select 10 kA as your primary side fault current in lieu of better info from the utility. Service transformer impedance is very important to get correct data. My utility may have a range of 2.5% to 5% depending on size and configuration and transformer efficiency. I place the highest importannce on the size/rating/type of primary fuse(s) protecting the utility transformer. Most have high voltage protection fuses only, but some have secondary voltage fuses or overcurrent protection of some sort. This overcurrent protection data can be incorporated into your analysis for a better estimate of the duty on the first disconnecting means.

Based on a range of studies I've performed, guesses are going to be misleading. I recommend you work with the utility to get the service transformer impedance and the primary & secondary overcurrent protection data. Ask what is the utility's range of primary fault current so that you can do a high to low sensitivity check on primary source impedance - but I'm guessing that primary source system impedance will have only a small effect.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Dec 10, 2012 1:55 pm 
User avatar

Joined: Sat Feb 07, 2009 9:24 am
Posts: 32
Location: Swanton, Vermont
Neil
As we've read here before and many of us have experienced, utility information is all over the place. Some are difficult to even get a reply from and often the information available is incomplete...at best.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Tue Dec 11, 2012 8:50 am 
Sparks Level

Joined: Fri Apr 15, 2011 7:43 am
Posts: 177
Location: Colorado
If we can get the transformer XFMR impedance is there a "safe" way to estimate the primary fault current so that the secondary fault current is close - enough. The fact that the secondary is based on infinite system impedance is wrong


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Tue Dec 11, 2012 5:42 pm 

Joined: Thu Aug 30, 2012 8:04 am
Posts: 2
To engrick: A "safe" estimate is relative isn't it? It depends whether your arc flash hazard is more severe at a higher fault current or a lower fault current. With or without actual utility data, I recommend calculating the arc flash hazard at a minimum of two levels of fault current. If you accept the statement that the transformer impedance is likely to be much greater than the system source impedance, then you can use the transformer impedance instead of the infinite system impedance to examine the secondary fault current. This will yield a higher calculated fault current than actual. If you wish to check arc flash hazards including some source impedance, estimate the total system impedance to be a percentage of the service transformer impedance. This will yield a smaller calculated fault current value, possibly closer to actual, but unknown. (You could just as easily take a fraction of the higher calculated fault current. Either method is still an estimate.)

Here is a real example from my utility system. Suburban feeder to a 1000 kVA 480 transformer. (Feel free to check my math... Per Unit calculations can be confusing.) Disclaimer: This of course is only one example. Please don't extrapolate the use of this data into your situation without comparing the system parameters. Use good judgement.

substation transformer impedance Z = 9.2% at 161/13.8 kV and 15 MVA and X/R = 34.
9300 ft. feeder cable Z = 0.41 + j 0.37 ohms
1000 kVA service transformer impedance Z = 4% at 13.8 kV /480 V and 1000 kVA and X/R = 20

Convert these various impedances to a common per unit base at 100 MVA.
0.0180 + j 0.6131 pu = sub transformer
0.2150 + j 0.7614 pu = feeder cable

0.2330 + j 1.3745 pu = total primary source impedance prior to service transformer
(3 ph. fault current at 13.8 kV primary side of service transformer = 3000 amps)

0.2000 + j 3.9950 pu = service transformer

0.4326 + j 5.2339 pu = total impedance from system through service transformer
(3 ph. fault current at 480 V secondary side of service transformer - 22,900 amps)

system Z = 1.39 pu
transf Z = 4.0 pu
ratio = 1.39 / 4 = 35%
net X/R to service = 12

Had we only used the service transformer impedance, then the calculated secondary fault current is 30,070 amps.

How does 22,900 compare to 30,000 amps when examining the clearing time of the overcurrent device? How does this translate into the calculated arc flash hazard?

I realize that one feels "out in the cold" without utility data. But remember, the utility feeder can change... without your knowledge. You will at least know if the service transformer is changed out since it is on your property. So by focusing on the service transformer impedance and then doing a sensitivity check at lesser fault currents, you can build a picture of the range of hazards. It obviously requires judgement.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Thu Dec 13, 2012 3:21 pm 
Sparks Level

Joined: Fri Apr 15, 2011 7:43 am
Posts: 177
Location: Colorado
Thank you Neil. This is the same approach I was looking at using with a few more refinements. I think you nail most everything I was concerned with. We are working on several hundred plants that are located on small distribution systems that do not have system information. I am in Denver and the local utility only provide infinite bus for thier distribution system.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Mon Dec 17, 2012 10:13 am 

Joined: Wed Aug 25, 2010 7:15 am
Posts: 24
Location: St. Paul, MN
For 480V services, a number of utilities will provide the primary fuse, transformer KVA, impedance, and voltages, and the secondary fault values and sometimes X/R. I prefer to have the utility contribution on the line side of the primary fuse.

This quickest way I have found to calculate the primary utility is to let SKM do the per unit calculations for me. I put a temporary utility on the secondary of the transformer then I apply a datablock to the one-line that shows the R and X pu values for the components. For pos seq values you just need to subtract the transformer values from the temporary utility values and enter them in the primary utility component.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Sat Jan 19, 2013 11:46 am 

Joined: Tue Mar 02, 2010 1:50 pm
Posts: 25
Earlier on in my arc flash work I would run scenarios with high and low impedance values for the distribution transformer to get the range of available fault duty as has been mentioned before. One scenario would assume a low fault current from the utility and a high impedance on the distribution transformer and visa versa. It worked for being very conservative.

Since I work for the utility I was able to get more sofisticated and calculate the fault current at the substation bus (often for different transmission configurations), then include the model of the feeder and lateral in my system model in SKM. Then I felt more confident in my results.

One of the interesting things that is facing me right now is the generating plant I'm modeling can be islanded with one other plant. If I am to account for the possible range of available fault currents I have to account for the lowest, which is the fault contribution of a 2 MVA machine during an islanded event. This makes the range of fault currents available at the 115kV level 14 kA to 30 A, which I think will play havoc with my hazard analysis results.

Casey


Top
 Profile Send private message  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 10 posts ] 

All times are UTC - 7 hours


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
© 2019 Arcflash Forum / Brainfiller, Inc. | P.O. Box 12024 | Scottsdale, AZ 85267 USA | 800-874-8883