Ampacity Derating

[iwire]

Why would this not be addressed in the code?

Because no one has written a proposal that shows the need for the change and new wording that would easily be understood.

What you are asking for makes sense and would be safe, but how you would word the exception clearly and concisely would be tough.

 

[texie]

Derating

Derating

What they say, you'll have to derate. However, 9 switch legs, I assume it's a 20 or 30A circuit. In which case the derating starts at the wire ampacity, not the overcurrent protection. #12 is good for 30A, so even at 70% you still got 21A which is above the max overcurrent protection you can use. Now if the neutral is sharing, than it doesn't count so you're in the 70% derating (6-9). If that's not the case I believe you have to derate to the 60% (too lazy to check the code right now, but I'm pretty sure it's 60%) which only gives you 18A. It's abit beyond my understanding thought why you have 9 switch legs in a conduit, surely you don't have a 9 gang box? do you?

I'll give you a real world example:
Panelized lighting control such as Clipsal and others it is common to have 1 neutral and 8 switch legs in a home run all fed from the same branch circuit. If you follow the letter of of the derating rules you would have to derate to 50%. In my opinion this serves no purpose and is a terrible waste of copper, labor, etc.

 

[danickstr]

This is the same issue with little generator stab-in boxes off of a load center. They run 10 circuits through a 5 foot conduit, but do not derate, from what I have seen of these installations.

In a purely theoretical defense of the NEC, it could be rewired later to allow other OCPD's to operate the UC's, so that would make the need for derating more practical, especially if the run is not visible from the panel.

 

[Smart $]

I'll give you a real world example:
Panelized lighting control such as Clipsal and others it is common to have 1 neutral and 8 switch legs in a home run all fed from the same branch circuit. If you follow the letter of of the derating rules you would have to derate to 50%. In my opinion this serves no purpose and is a terrible waste of copper, labor, etc.

I don't believe it makes a difference. Let's say you have 20 switch legs supplied by the same 20A ocpd and all is #12 THHN in EMT. Additionally your load when all are energized is at max... 16A continuous. The derating is 50% which takes you from 30A to 15A. Given you have 20 switched circuits, they average 0.8A each. Though in practice they may not be balanced as such, as long as no switched circuit is loaded greater than 15A, no upsizing is necessary.

Here is where you have to remember, the derated ampacity only has to be equal or greater than the connected load. It does not have to be 125% for continuous loading, nor for the load of the entire circuit, and the ocpd does not have to be "derated" if the wire size stays the same.

 

[charlie b]

. . . it would seem logic would tell you that you are spreading the available current over 9 conductors with 9 times the surface area to disipate heat as opposed to 1 conductor.

And the heat released by any one of the conductors causes an increase in the temperature of the air that surrounds the others. That said, it is fairly easy to show that the heat generated by two or more identical wires, each carrying some amount of current, is less than the heat generated by a single wire of the same size, with it carrying the same total current.

Why would this not be addressed in the code?

I think the basic problem is that we are talking about calculating the ampacity of a wire. That means we are determining how much current it can handle. The amount of current a wire can handle should not be based on the amount of current you intend to send through it.

 

[kwired]

Just a thought;

Exception to derating conductor ampacity due to more than 3 CCC in a raceway:
When the total current on all conductors is less than the rated ampacity of any one of them the ampacity of the conductors shall not be required to be derated.

There are too many unintended possibilities that would be allowed with this wording. Wording would need to include that all conductors are on the same branch circuit and only carrying a portion of the total branch circuit load, but if there are other branch circuits in the raceway it will complicate things. You may need more than just clever words to get it into the code - some testing results may be needed also.

 

[Smart $]

There are too many unintended possibilities that would be allowed with this wording. Wording would need to include that all conductors are on the same branch circuit and only carrying a portion of the total branch circuit load, but if there are other branch circuits in the raceway it will complicate things. You may need more than just clever words to get it into the code - some testing results may be needed also.

With proper application of derating, there should be little reason to change or add anything to current Code.

Let's use an example with 60? conductors because derating would affect their sizing before 75? and 90?, say #12 on 20A ocpd.

Two switched hots and a neutral, no derating.

Three to five switched hots and a neutral, 80% derating of 25A leaves us at 20A each... so we are still good without any change in wire size.

Six to eight hots and a neutral, 70% derating of 25A leaves us at 17.5A each... this would limit our non-continuous load on the hots to less than the ocpd rating, but the continuous loading can still be the max of 16A, so most often this will not result in having to upsize.

If we add more switched hots, the limiting factor is likely going to be the derated ampacity of the neutral, for it has to carry the entire load. This is where running a mwbc would be a definite advantage... but how often does one install 9 or more switch legs in one cable or raceway???

 

[Dennis Alwon]

I don't believe it makes a difference. Let's say you have 20 switch legs supplied by the same 20A ocpd and all is #12 THHN in EMT. Additionally your load when all are energized is at max... 16A continuous. The derating is 50% which takes you from 30A to 15A. Given you have 20 switched circuits, they average 0.8A each.

I see your point but what about the branch circuit feed that is in that conduit. If the load is , as you say, 16 amps and we have say 10 conductors- one feed and 9 switch legs all of the same circuit, then we must derate 50%. Now a #12 rated at 30 amps is now only rated 15 amps. Are we not in violation as the code is written?

The question I have is how can the load ever be greater than a single 20 amp circuit. The load from one 16 amp circuit will generate as much heat as the load from 9 switch legs on the same circuit with the same load. Yet we must derate for the 10 conductors but not for the 2 conductor scenario.

I am not sure why an exception cannot be written for all the different scenario.

One being a conduit with unlimited switch legs all from the same circuit needs no derating. Another exception could be 3 way travelers, etc.

 

[Smart $]

I see your point but what about the branch circuit feed that is in that conduit. If the load is , as you say, 16 amps and we have say 10 conductors- one feed and 9 switch legs all of the same circuit, then we must derate 50%. Now a #12 rated at 30 amps is now only rated 15 amps. Are we not in violation as the code is written?

Yes, we are.... I forgot about the single neutral carrying the entire load. But we could run a second neutral. Thus we could go up to 18 switch legs and two neutrals... and as long as our load on each conductor is 15A or less we are good to go.

The question I have is how can the load ever be greater than a single 20 amp circuit. The load from one 16 amp circuit will generate as much heat as the load from 9 switch legs on the same circuit with the same load. Yet we must derate for the 10 conductors but not for the 2 conductor scenario.

Actually the 16A circuit would generate more heat...

I am not sure why an exception cannot be written for all the different scenario.

One being a conduit with unlimited switch legs all from the same circuit needs no derating. Another exception could be 3 way travelers, etc.

I'm sure an exception could be written, and I'm not against such. All I'm saying is that for most "switched" scenarios, wire size does not have to be increased. Several believe that once the derated ampacity goes below the ocpd rating, the wire needs upsized or the ocpd rating needs decreased. That is simply not true. As long as the derated ampacity is equal or greater than the load supplied, no wire size or ocpd rating change is required.

If you look at it from the perspective of distribution, load current [equally divided] on the conductors decreases faster than the derated ampacity.

20A load 100% avg on 1SL no derating
20A load 50% avg on 2SL no derating
20A load 33% avg on 3SL 80% derating
20A load 25% avg on 4SL 80% derating
20A load 20% avg on 5SL 80% derating
20A load 16% avg on 6SL 70% derating
...and so on.

 

[infinity]

Y
As long as the derated ampacity is equal or greater than the load supplied, no wire size or ocpd rating change is required.

So you're saying that a #12 THHN derated to 50% can still be on a 20 amp circuit if the load is limited to 15 amps?

 

[Dennis Alwon]

So you're saying that a #12 THHN derated to 50% can still be on a 20 amp circuit if the load is limited to 15 amps?

Yes because a #12 is rated 30 amps at 90C.

 

[Jim W in Tampa]

Yes because a #12 is rated 30 amps at 90C.

Yes that #12 is 30 amps and at 50% is good for 15 amps. Now tell us where you get the idea that we can still use a 20 amp breaker ? code # -----

 

[Dennis Alwon]

Yes that #12 is 30 amps and at 50% is good for 15 amps. Now tell us where you get the idea that we can still use a 20 amp breaker ? code # -----

I missed what Trevor was saying. You would have to have a 15 amp breaker.

 

[Smart $]

So you're saying that a #12 THHN derated to 50% can still be on a 20 amp circuit if the load is limited to 15 amps?

Yes that #12 is 30 amps and at 50% is good for 15 amps. Now tell us where you get the idea that we can still use a 20 amp breaker ? code # -----

There's a very thin line of interpretation that would allow it. But rather than getting into said interpretation, I believe most AHJ's and code afficionados would not agree with me and deem it a violation... so I will just cede to being wrong.

 

[Jim W in Tampa]

There's a very thin line of interpretation that would allow it. But rather than getting into said interpretation, I believe most AHJ's and code afficionados would not agree with me and deem it a violation... so I will just cede to being wrong.

Not asking you to back down. Just curious if i missed something.

 

[Smart $]

Not asking you to back down. Just curious if i missed something.

Missing something is dependent on your interpretation Code.

Condsider the #12 derated to 15A as tap conductors being protected with a 20A ocpd. This meets the definition of tap conductor in 240.2. We can start the overcurrent protection compliance check at 210.20... otherwise compliant, (B) thereof sends us off to 240.4(E)(3), which sends us to 240.21(A), which says if we meet the requirements of 210.19 we can have ocpd per 210.20.

Under 210.19(A)(4), if you consider Exception No. 1 the rule rather than the exception, we would be in violation. If we just go by the rule and disregard the exception, which is how the Code is supposed to applied, we are compliant. Like I said, a thin line of interpretation, but I say it is legal when literally interpretted. Quite possibly not the intent, but nonetheless, there it is in black and white... and since we started in 210.20 we need not go any further.