DC Bussing Derating

[Johnshealey]

I have an application for DC bussing at 12VDC, 1,000A. A previous installation used three cables 350kcmil THHN. Assuming 90? insulation, we have 1,050A ampacity at 86?F, free air.

There are six cables running 200-300' in Sched. 40 PVC. Derating for 95?C ambient (94%) and 4-6 cables in a conduit (80%), I get a total ampacity of 790A. If using 600kcmil, with derating, the total ampacity would be 1,072A.

Am I right? Do we need 600kcmil? Do the derating requirments apply to DC bussing?

 

[winnie]

The derating tables apply to any NEC installation. Doesn't matter if it is 12V or 480V. Since this is an underground install, you might try going with the 'engineering supervision' approach, where the actual ampacity is calculated from the physical characteristics of the conductor and surroundings.

However you have a much bigger issue, one that totally blows away the derating issue: voltage drop.

Voltage drop is _not_ an NEC requirement, so you could choose to ignore it. But this would probably be a very bad design choice. Assuming 3 parallel 600kcmil copper conductors, and a distance of 250 feet, I calculate a circuit resistance of 0.005 ohms. Push 1000A through this resistance, and you have a voltage drop of about 5V. Almost negligible on a 480V system, but terrible on a 12V system.

-Jon

 

[zbang]

My first thought was voltage drop, too. Is there a specific reason for running DC that distance? Seems like with that much copper involved, an AC feed and transformer/rectifier would be cheaper & easier.

 

[Johnshealey]

Thanks for comments on DC Bussing

Thanks for comments on DC Bussing

I probably should have mentioned that it's an electroplating application. We're adding new DC rectifiers and cabling to an existing system. The process parameters are not known, but it's reasonable to assume they're actually pushing about 500A (50% capacity). The current would be controlled at the rectifier and the voltage would rise to meet the total load (limited to 12V). The 12V x 1,000A = 12kW rectifiers have been working for many years with three 350MCM cables, so line losses are not a big concern. My concern is that no derating of conductors was done by the original system designers and that the new conductors should be sized for full-scale operation (1,000A), even if it's never used in practice.

This isn't an underground application. The conductors run in 4" PVC conduit up the wall of a rectifier room, across the ceiling, and down to distant plating stations. There is one 4" conduit per plating station with six 350MCM cables in each. Any new installation would be subjected to local inspection; the fact that it was done improperly in the past wouldn't help our case much.

I made a mistake in the original post: its 95?F ambient, not 95?C!!!

 

[dereckbc]

I design very large DC plants for the telecom sector, and we never ever use NEC tables 310.xx for any cable sizing. It is all designed based on maximum allowed voltage drop vs the operatong voltage. So here is what the telecom sector uses for maximum voltage drop and max load current from the battery terminals to the load equipment terminals.

48 = 2
24 = 1
12 = 0.5

With 12 volts, at 1000 amps, and 250 feet, you are talking some huge cable quantity that exceeds any NEC minimum requirement.