Derating Cables Buried In Duct Banks

[stretched2thin]

I'm not exactly an expert on the NEC, but I'm trying to learn. I'm currently working on a project that involves power supply cables (208/120 V - 2000 Amps) being routed through duct banks. The cables are going to have to be derated, due to the fact that they are surrounded by adjacent duct banks with similar cables. There may be as many as 36 to 42 conduits in this duct bank. The NEC 310.15(B) (Annex B) doesn't address anything but very general cases involving derated ampacities. What I'd like to know is what everyone else does for derating the cables. I'm aware that there are several versions of software available that do this. Is this how everyone does it? Or am I missing something and possibly making this more difficult that it is.

 

[dpc]

Actually, from a strictly NEC-viewpoint, I don't believe any derating (due to mutual heating in duct banks) is required for circuits rated 2000V and below. The duct bank derating is only mandatory above 2000V. A few revisions ago, the NEC did attempt to require a similar derating for low voltage circuits, but there was such an outcry from contractors, that the requirement was quickly abandoned.

Of course, this makes little sense based on the physical laws involved. But empirically, there are very few cases of UG conductor failures in LV circuits due to overheating.

To be conservative, you can apply the same derating factors used in the NEC tables for medium-voltage cables. The NEC Tables are based on the Neher-McGrath equations using the thermal characteristic assumptions stated in the tables.

There are numerous software programs available to do this calculation - a manual calculation is a pain, but can be done.

If you are buying enough cable, one of the cable vendors may be willing to do the calculation for you, and provide recommendations.

 

[alehman]

Duct banks with more than two conduits in both dimensions require considerable derating of conductor ampacity. If you are thinking about stacking conduits more than 3 high in the duct bank, the N-M calculations will show the interior ducts are nearly worthless in terms cable ampacity.

The question of whether or not the NEC requires derating seems to be a subject of some debate.

See also
Thread238-44759 Thread238-53067 Thread238-28317 Thread238-25424

 

[jghrist]

No derating is required by the NEC if the loads are calculated in accordance with Article 220. See the FPN on Rule 310.15(B). This is because "not all diversity factors and load factors found in most actual installations are specifically provided for in Article 220." (quoted commentary from NEC Handbook). If the loads are not calculated in accordance with Article 220, then the ampacity should be determined under engineering supervision according to 310.15(C) and Annex B.

It would be prudent for large duct banks to calculate the ampacity under engineering supervision even if it is not strictly required. As alehman points out, the interior ducts of a large duct bank are severely restricted because of heating by surrounding cables.

 

[BJC]

streached2thin.
Some is leaving out an important part of the code. That part being the AHJ who can interpret things the way they see them. The 2002 NEC does not have the proviso that Chapter 9 is for reference only and many AHJs will not accept deviation from it. I know of a couple of cases where cables in duct banks have failed due to over heating. The owners went after the designers claiming faulty design and wanted damages for lost production time and replacement of the ducts/cables. One of the duct banks had been back filled with pea gravel because it was easy to place. The cables failed in hot weather when the water normally around the ducts went away.

I have used the Amp Calc program and it give results that pretty much agree with the values in appendix B of the NEC.
There are configurations in the appendix B that will allow 2000 amps. It's an expensive program.
Table B310.7 in appendix B has a configuration that is good for 2,286 amps if the ducts are in a concrete duct bank with a RHO of 55 ( most concrete is) . If your load factor is less than 100% ( 6 ducts with 750 “MCM cables ) you can use smaller and/or less conductors. You will have to use the chart in Figure B310.1. . A confusing chart but it can help.

The tables do have limits. I have fond that it is almost impossible to take the output of a 2,500KVA transformer and run it through a duct bank. The best option I have found is in cabel tray either above ground or in pre- cast trench.

 

[stretched2thin]

Thanks for all of the input. I've been tempted to go with the AmpCalc program, but wanted to see if maybe I was missing something and making this harder than it seems.

Thanks again.

 

[dpc]

To BJC,

In Annex B of the 2002 NEC, the first sentence says "This annex in not a part of the requirements of this NFPA document but is included for informational purposes only."

Seems pretty clear to me.

I've never had a problem with this issue with any inspectors, but these things can be pretty localized. It seems inspectors in various regions have different hot buttons.

But I agree with jghrist that a prudent engineer should carefully consider if derating is required. You need to do what you think is right, even if not required by NEC.