conductor derating for temperature

[megloff11x]

Under the various tables 310.XX there is a derating table for temperature that seems to indicate that wire ampacity is specified at 30C/86F. Then there are derating numbers for higher and lower temperatures. And we usually size at 125% in many applications under other articles. Similar tables and deratings appear in NFPA79 as well.

When or do you apply this other temperature derating? And what are we using for temperature, maximum expected or is this supposed to be an annual average?

Most attics can hit 140F. Most plants can hit >100F in the summer, especially in parts south. Most equipment is rated for 40-50C max operating environment.

Where called for, is the 125% rule expected to cover this temperature issue or not?

I may be missing something in the fine print. I prefer to err on the side of thicker, especially if I know it will be a continuous load in a very warm room, but it often gets annoyingly thick, sometimes thicker than the terminals can accept. I won't start a rant on UL listed 20A & 30A rated devices with 14AWG or smaller terminals here.

Matt

 

[jtester]

You will find this to be a "hotly" debated issue. The length of time a conductor is in an ambient temperature higher than 86 degrees is the question.

There are lots of opinions ranging from average the temperature over a year, to use the highest seen in any given time period.

I don't believe the NEC is clear on the subject.

Jim T

 

[winnie]

As far as I know, there is nothing in the NEC itself that specifies this.

As I've stated in the past when this topic has come up, my 'physics opinion' is that the correct temperature to use is a weighted average temperature that would give the same ageing rate as the actual temperature conditions.

Simply using the mean temperature is not a good enough technique, a conductor that spends half its time at 90C and half its time at 50C will age more rapidly than a conductor that is constantly at 70C. An approximation used in the motor design world is that each 10C increase in temperature doubles the ageing rate of the insulation.

I suggest that the correct 'weighted average' to use is the 'log mean exponential', since most of the equations for ageing use an e^T term somewhere in them. However I've not done a detailed analysis of this proposal.

-Jon

 

[charlie b]

There are two circumstances I can recall off the top of my head in which a 125% factor comes into play. In both cases, the factor only influences the amount of current that the conductor is required to be able to handle (i.e., the required ampacity).

But the temperature correction you are describing (and the ?more than 3 current-carrying conductors in one raceway? correction as well) will only influence the amount of current a conductor is permitted to carry (i.e., the available ampacity).

The two concepts do not mix in any way. First you find out how much ampacity you need. Then you go find a conductor that has that ampacity.