Derating -- diversity?

Status
Not open for further replies.
rabtrfld

rabtrfld

Member
Location
Milwaukee, Wisconsin
Suppose you have a bank of light switches, say 3... one branch circuit, L & N, enters the box through conduit. Coming out through one conduit, there is one neutral and three travelers -- same wire gauge.

The outgoing conduit would run at the same temperature as, or cooler than, the input. Any time the current is split between travelers the parallel resistance per foot is less and the temperature rise is less.

And yet, for derating purposes, the input has 2 CCC's and is rated 100% but the output has 4 CCC's and is derated to 70%.

This seems silly; is there a way around it?

BTW, my old Book uses the term "diversity" in the derating notes, which are confusing, but "diversity" is not in the index, glossary, or table of contents. It seems to mean that not all wires are energized at once. That wouldn't apply to this case.
 
I'm a little confused how can you have 3 switches and only three travelers? Or did you mean three switch legs? And for 4 CCC's the derating would be 80%.
 
rabtrfld said:
And yet, for derating purposes, the input has 2 CCC's and is rated 100% but the output has 4 CCC's and is derated to 70%.
Click to expand...

I don't see anything in 310.15 that allows us to consider a fixed lighting load. I derate as per the rating of the conductor from T310.16.
 
You're stuck with 80% deration, even though for conductors #14-10 it won't change a thing.
 
The actual installation has so many wires the adjustment factor is 50%, and it hurts! The tube is plenty big -- 20% fill.

If the customer wants to break the lighting circuit into several banks, then:

1) When they are all on, the current is shared between several wires, so the conduit runs cooler than if they were all in one bank.

2) Some of them can be turned off at times, again the tube runs cooler.

PITA that some other, unrelated branch circuit gets derated just because it shares a tube with these extra wires which are actually improving the situation, not harming it.
 
Another way to look at it.... Lets say your lighting load is 15 amps and divided equally on the switches. This means that each switch leg is only carrying 5 amps. IMHO derating becomes a moot point.
 
Lcdrwalker said:
Another way to look at it.... Lets say your lighting load is 15 amps and divided equally on the switches. This means that each switch leg is only carrying 5 amps. IMHO derating becomes a moot point.
Click to expand...


I agree but the NEC doesn't give you the option to ignore derating.
 
Lcdrwalker said:
Another way to look at it.... Lets say your lighting load is 15 amps and divided equally on the switches. This means that each switch leg is only carrying 5 amps. IMHO derating becomes a moot point.
Click to expand...

Once again, the real problem is other ckts. There are some #8s that need to carry 30 amps, and some #12s that serve 20-amp receptacles. At 70% they are both OK; at 50% there is a problem. The #8 rated 27.5 amps can squeak by with a 30-amp breaker because that is the nearest trade size.

And even if the load is 5 amps, the wire is still connected to a 15-amp breaker, and that determines the circuit rating, no?
I don't see anything in 310.15 that allows us to consider a fixed lighting load. I derate as per the rating of the conductor from T310.16.
Click to expand...
Maybe we should have it inspected during the winter when we can apply the ambient temperature allowance!
 
parallel wires

parallel wires

Somewhere in the Book is a formula for derating parallel wires. For instance, two parallel wires do not double the ampacity, but somewhat less (I think it may be 1.414). This makes sense as it is notoriously difficult to split current equally in low resistances.

Maybe in the case of multiple switch legs they could be rated as parallel conductors? Then they could be counted as one CCC as far as derating other wires in the same tube?

Note that if each switch leg went out through a separate conduit, any leg might carry the full branch circuit current, and must be counted as a CCC within its own tube. Only if some switch legs share a tube could they be counted as one.

I think the code is inflexible because of the difficulty of validating and inspecting such a design. If the inspector sees ten wires in a tube she can not be expected to know how they are routed and connected throughout the system. Imagine peering in a crowded pull box and trying to determine if a set of parallel wires are staying together or going separate ways.
 
rabtrfld said:
The #8 rated 27.5 amps can squeak by with a 30-amp breaker because that is the nearest trade size.
Click to expand...
That would be true if, and only if, the load is below 27.5 amps. We are allowed to use the next higher breaker size; we are not allowed to use a wire with an ampacity that is lower than the load. But you said that they must carry a load of 30. So I am not clear on your situation.
 
rabtrfld said:
The outgoing conduit would run at the same temperature as, or cooler than, the input.
Click to expand...
Here again I am confused by what you are trying to describe. First of all, the temperature of a conduit almost never comes into play. There are derating factors that we must apply, if the ambient temperature (meaning specifically that we presume that there is, at the moment, no electricity in the building at all, and therefore no heat input from any wires) is too high. But we do not derate on the basis of what happens to the temperature inside a wire when one or more wires are carrying current.
rabtrfld said:
Any time the current is split between travelers the parallel resistance per foot is less and the temperature rise is less.
Click to expand...
You do not have any wires in parallel, or perhaps I continue to misunderstand the application. I think you are saying that two wires enter a box, are connected to a set of switches, and the conduit(s) leaving the box are bringing power from the switches to a bunch of lights. If that is what you mean, then none of the wires are in parallel, because they are serving different loads.
 
If you don't understand the problem, you are trying too hard. I intended to introduce a thought problem of principle, not a specific case. Kind of like, "Johnny has five green socks and three blue ones. How many socks must he pick to guarantee a matched pair." Most people would understand what is intended without wondering how dark it is or whether the socks are wool or cotton.

Consider Table 310-16 note 10(a) "A neutral conductor which carries only the unbalanced current from other conductors, as in the case of normally balanced circuits of three or more conductors, shall not be counted when applying the provisions of note 8."

I hope everyone reading this understands why this is, without having to explain it. The point of this topic is to say, that a similar principle applies in other situations, where unfortunately the Code does not provide a similar exception.

I apologize for not using emoticons. I learned to write on a mechanical typewriter that used lower-case ell for the number 1. Sometimes we had to indicate humor using only the letters A through Z and the space bar.
 
the temperature of a conduit almost never comes into play
Click to expand...
Where I said "temperature of the conduit" I should have said, "temperature of the wires in the conduit," or better, "temperature of the insulation of the wires in the conduit," or better, "maximum value of the weighted time integral of the temperature at any point in the insulation of the wires in the conduit," but no matter how I say it somebody will probably find fault. I was trying to keep it simple.

If you put your left and right hands on the two tubes you will feel a difference. Of course the Code does not consider this. That is my point. A relevant physical phenomenon exists which is not recognized by the code.

But we do not derate on the basis of what happens to the temperature inside a wire when one or more wires are carrying current.
Click to expand...
We derate because of what happens to the temperature inside [the insulation of] a wire when one or more wires are carrying current. We derate on the basis of simpler considerations for practical reasons.
 
rabtrfld said:
Where I said "temperature of the conduit" I should have said, "temperature of the wires in the conduit," . . . . A relevant physical phenomenon exists which is not recognized by the code.
Click to expand...
That phenomenon is recognized.
(1) If the actual ambient temperature experienced in the field is below the value you use for derating, and
(2) If the actual amount of current flowing in the conductor is below the value you calculated for the conductor's ampacity, then
(3) The temperature rise experienced by the conductor (i.e., due to I2R heating), when added to the ambient temperature, will give you a net temperature that is within the conductor's insulation rating.
 
Last edited:
I believe that there is a bit of circular discussion going on here.

When conductors carry current they self heat.

If the conductors get too hot, then their insulation will degrade faster than is acceptable.

The rules for conductor ampacity and conductor protection are designed to prevent conductors from overheating.

rabtrfld has described two conductor arrangements, call them 'SUPPLY' and 'LOAD'. The code rules for ampacity and protection require 'derating' in the 'LOAD' arrangement relative to the 'SUPPLY' arrangement. However the 'LOAD' arrangement self heating will always be less than or equal to the self heating of the 'SUPPLY' arrangement.

To restate his example, the 'SUPPLY' arrangement consists of two conductors, grounded and ungrounded, in a raceway, feeding a set of switches. The 'LOAD' arrangement consists of a larger number of conductors, one grounded and several switched ungrounded, all in the same raceway fed by these switches. Because all the current flowing on the 'LOAD' switched ungrounded conductors must have originally come from the 'SUPPLY' ungrounded conductor, the total or net current flowing on this _set_ of conductors must be equal to the current flowing on the single supply conductor.

Because this same total current is now distributed over a larger copper cross section, the conductor self heating will be reduced. Since self heating is reduced, one would think that conductor ampacity would be _increased_

Because there are now more 'current carrying conductors', the derating rules require that conductor ampacity be reduced.

-Jon
 
Status
Not open for further replies.
Top