Coiling Cable to Reduce Short Circuit

[CaptainKu]

Hey all. I have a existing panel that arrived on site that is only rated 10KA SCCR. This is a 75HP air compressor. I was planning on running a 1/0 cable roughly 60feet to the panel but the short circuit at the end of the cable would still be 14KA. I have already trimmed down the calc as much as possible including using utility provided AFC instead of infinite bus. My thought was to extend the circuit to reduce the current but I would need to add about 100 feet. I am looking at the best way to install this cable and have a spare 24" tray I could install to loop the cable in ~4 times to get 100 feet added. I know looping in a circle could create additional inductance but would 4 loops of 1/0 cable in a 12' tray really cause a issue? Is there a better way to do this?

 

[ron]

We have extra length added to reduce available short circuit current via running the raceway / circuit around the room as many times as needed, which avoids any potential magnetic field issues. You can compare that cost / effort to installing an isolation 480-480V transformer to add impedance.
Also, I've had luck pushing back on the mechanical equipment vendor to come up with some tested assembly, or essentially a series rating for their equipment.
Do NOT fall into blatantly adding current limiting fuses, because that is not NEC compliant.

 

[charlie b]

I have seen this done on larger distribution panels. Not really a good idea, but sometimes you have to whatever is doable.

The more loops, the higher the inductance. The larger the diameter of the loops, the smaller the inductance. Without doing the math, and I have no desire to do the math, I would have to believe that your 4 loops will not cause an issue by way of adding inductive reactance. However, it may create an issue with voltage drop.

You could add a 480-480 isolation transformer and use primary-side taps to raise the secondary voltage. But then, voltage may get too high during low loading periods. Also, this wastes energy, in case that is a consideration.

I agree that "current-limiting" fuses are not the answer. Their name sounds encouraging, but they don't work the way their name suggests.

The best way to avoid this situation is putting minimum SCCR requirements in the specs. I gather that you are past that point already. I agree with Ron; push back on the equipment supplier.

 

[petersonra]

How did you come up with 1/0 conductors? Any chance you can use #1 instead? You might just make it without coils.

 

[winnie]

Loops add inductance to individual conductors. But if you have a complete set of circuit conductors all following the same path, the net current in the bundle should be zero, and you don't see any additional circuit inductance.

The loops _may_ add inductance to fault conditions, if the fault current doesn't follow the same looping path.

 

[electrofelon]

How did you come up with 1/0 conductors? Any chance you can use #1 instead? You might just make it without coils.

Good catch. Although we don't know the voltage, assuming it is 480, you would need 120A conductors and #1 would do it. Another idea ( if the #1 doesn't get you there) is to use a #2 conductor at its 90 degree rating. You would need to splice #1 to the #2 to avoid the 75 termination limitation.

 

[CaptainKu]

Thanks all! After reviewing the cable sizing again I believe I can get it to a #1 with no issues but still would need to add about 40 feet, much better. I tend to agree with current limiting fuse answer discussed here but did want to clarify something. The electrical contractor is trying to argue you can just use CL fuses with out excess cable. Before I respond back that this would only work if I went through the panel and confirmed series ratings of each 480V component it would not be compliant which would likely cost more in engineering time then just looping 40 foot in the tray. Could you make the argument that since there is no dynamic protection in the circuit (only motor contactors with no instantaneous protection) that the RMS current let through curve of the fuse could prove that CL fuses could protect the panel without additional cable?

 

[David Castor]

The extra cable will add resistance and inductance - the resulting higher impedance will increase your voltage drop, so you might want to check that.

 

[jim dungar]

Could you make the argument that since there is no dynamic protection in the circuit (only motor contactors with no instantaneous protection) that the RMS current let through curve of the fuse could prove that CL fuses could protect the panel without additional cable?

You could under your stamp and insurance.
I never did it under my PE license.

It takes a lot of current for some fuses to enter into their current limiting region and we don't know the withstand times of many components. Fuses are great for stuff like conductors, but I feel more comfortable with tested combinations.

 

[topgone]

Thanks all! After reviewing the cable sizing again I believe I can get it to a #1 with no issues but still would need to add about 40 feet, much better. I tend to agree with current limiting fuse answer discussed here but did want to clarify something. The electrical contractor is trying to argue you can just use CL fuses with out excess cable. Before I respond back that this would only work if I went through the panel and confirmed series ratings of each 480V component it would not be compliant which would likely cost more in engineering time then just looping 40 foot in the tray. Could you make the argument that since there is no dynamic protection in the circuit (only motor contactors with no instantaneous protection) that the RMS current let through curve of the fuse could prove that CL fuses could protect the panel without additional cable?

It's done every time. Just route back the cables a little on the trays (if on tray) to get the desired length. It will be very ugly though if the cables are overhead as you have to coil those extra length.
The other option is to use air core reactors but you have to endure the inherent voltage drop. If you can, please post the details of your install. Maybe posters can guide you with numbers.

 

[Jraef]

I once had to reduce a system AFC from 44kA to below 42kA because a vendor F’d up when ordering a 1200A MCC for an e-House setup at a gravel pit. We trenched around the building and installed two extra loops. The copper was expensive, but not as expensive as the alternatives. We had looked into current limiting reactors but they were super expensive for something this big.