Derating conductors of one circuit

mhulbert · Sep 8, 2005

I have a question on derating...I have a Lutron GrafikEye that is feeding 6 LV Lighting transformers. They are located about 40' apart. Lutron recomends/prefers that a seperate neutral be run back to the GE unit for each load. The Grafik Eye is fed off one 20A circuit. If I have 12 conductors in EMT (6 hots and 6 neutrals), all technically current carrying, I will have to use #10. Now, this makes no sense to me because it is impossible for the total current in this pipe to be above 20A, since all of these wires orginate off of one circuit.

Is there any wording in the code that would allow me to not derate as we are usually required to do with seperate circuits? If it makes it any easier to understand, imagine a 20A circuit fed to 6 toggles, with all of the switch legs going out of in the same pipe, along with neutrals. What do you think?

Mike

infinity · Sep 8, 2005

Re: Derating conductors of one circuit

I agree with your logic but I'm unsure if it's supported by the NEC. It would seem that 12 current carrying conductors fed from one 20 amp circuit in a properly sized conduit would not produce any more heat than 3 CCC's but I don't recall an exception that would allow you to avoid the derating.

iwire · Sep 8, 2005

Re: Derating conductors of one circuit

Mike the short answer is your stuck with 10 AWG.

As it stands now there is no exception in the NEC to give you some relief.

I have limited experience with the graphic eye controller but if I remember even using 12 AWG on them was a pain. Hope you installed the largest box possible.

iwire · Sep 8, 2005

Re: Derating conductors of one circuit

I have to type faster.

I agree with infinity.

bphgravity · Sep 8, 2005

Re: Derating conductors of one circuit

I think most of us agree that a conductor's insulation will not suddenly fail if the temperature exceeds it's rating. (remember the gas usage anology Charlie was so kind to post) Based on that, I think most would also have reason to believe that if a conductor and its terminations were to exceed their temperature rating on occasion over it's life usage, the ultimate outcome may be a few months maybe years off its expected life, which is near forever to start with. Now that is really assuming worse case scenerio.

Let's take for example 3 receptacles in a single box. Each receptacle will serve a computer. Each is fed from an individual 20A branch circuit. Each circuit shares the same raceway. This makes 6 current-carrying conductors, however even with all three computers on at the same time, the load on each circuit is not even half of its rating. On sme occasions, one or more computers may not even be on resulting in even less load on the circuits.

When all is said and done, what protection is provided by making adjustments to the conductor ampacities when it is absolutely obvious the temperature ratings of the conductors will never be exceeded. Even if the cleaning persons come in and plug in 3 vacuum cleaners for 30 minutes, what hazard is really waiting to happen?

This is only a thought experiment. I understand the code has no exceptions or sliding scale to meet each and every possible installation, but I just wonder if the ampacity values and all the adjustmens an/or correction factors tend a little to the conservative side.

iwire · Sep 8, 2005

Re: Derating conductors of one circuit

Originally posted by bphgravity:
I just wonder if the ampacity values and all the adjustmens an/or correction factors tend a little to the conservative side.

I agree and would go so far to say very conservative in many applications.

I think that is the nature of a code that covers so many different types of occupancies and installations. They must figure on the worst case situation which may happen in only a very small number of installations.

charlie b · Sep 8, 2005

Re: Derating conductors of one circuit

I don't understand why there is a problem here. Let me start with this: What is the load on each of the LV lighting transformers?

With 12 conductors, each #12, and let me assume THHN, the ampacity of each conductor is 15 amps. You are not telling us that each of the six LV transformers has a current higher than 15 amps, are you? The total current coming in to the GrafikEye is under 20 amps, and I will even assume it is under 16 amps, since lighting is generally considered "continuous." Therefore each of the 6 circuits that branch off from the GrafikEye can have a current no higher than about 3 amps. The ampacity of 15 is certainly high enough for that. I suspect you could even use #14, unless the manufacturer says you can't.

What am I missing?

iwire · Sep 8, 2005

Re: Derating conductors of one circuit

Charlie that makes sense but it does not meet the requirements of the NEC.

The OCP in this case can not be higher than the derated rating of the conductor regardless of the connected load on any particular conductor.

I could forget about derating almost entirely if we could do what you suggest. Many times the prints we have show about 8 to 10 amps of load on each lighting circuit but I still have to base the conductor size on the OCP rating of 20 amps.

mhulbert · Sep 8, 2005

Re: Derating conductors of one circuit

Charlie,
Each zone on the unit is limited to 800W (about 6.6 A), and the total for the unit is 1920W. You are right, none of these wires will ever see near 20A, however, it seems the NEC does not allow for a situation like this, and I am still going to have to use #10.

Seems like an exception should be put in, something like "If a group of wires orgininates from one OCPD, then only 2 of these wires shall be counted as current carrying"

Mike

charlie b · Sep 8, 2005

Re: Derating conductors of one circuit

Originally posted by mhulbert: . . . the total for the unit is 1920W.

That's where the problem lies. Suppose, just for discussion purposes, that the total had been 1440 watts (12 amps total, or 2 amps per circuit). We will call this one "continuous," again for the purposes of discussion.

The ampacity of the #12's is 15 amps, as I said before. You could protect this circuit with a 15 amp breaker. The total load is 12 amps, but since it is continuous you need to limit the load to 80% of the breaker's rating. 80% of 15 is 12. Therefore, this would have been acceptable.

That is why my first question was "what is the load?" The fact that your load is beyond 80% of the 15 amp breaker forces you to go to a 20 amp breaker. The breaker selection, in turn, forces you to go with #10 wires. You need conductors for which the derated ampacity is at or above 20 amps. At 50% derating, a #10 is good for 20 amps.

Originally posted by iwire: The OCP in this case can not be higher than the derated rating of the conductor regardless of the connected load on any particular conductor.

That had been my intended point, though I did not say so in my first thread. In most of what we do, the design process begins with, "What is the load?"

mhulbert · Sep 9, 2005

Re: Derating conductors of one circuit

Charlie,
Not sure I follow you. The total load going through the conduit will be no larger than 1920W, but each "zone" of this unit is limited to 800W. Obviously, with 6 zones, we can't load each one up all the way. It seems like I "effectively" have 6.66A breakers on each wire, since the unit will overheat and shut off if we go above the 800W.

Looks like I'll be using #10, even though it will be overkill, the rules are rules!

hardworkingstiff · Sep 9, 2005

Re: Derating conductors of one circuit

Sounds to me that you may be able to use Annex B, Table B.310.11 to justify using #12 wire (that's the 2002 code).

iwire · Sep 9, 2005

Re: Derating conductors of one circuit

Originally posted by hardworkingstiff:
Sounds to me that you may be able to use Annex B, Table B.310.11 to justify using #12 wire (that's the 2002 code).

Sure, if your an engineer.

charlie b · Sep 9, 2005

Re: Derating conductors of one circuit

Originally posted by hardworkingstiff: Sounds to me that you may be able to use Annex B, Table B.310.11

Not an option. All of the six lighting transformers will be running at the same time, so all 12 wires have current at the same time. So you aren't going to get a load diversity os 50%

charlie b · Sep 9, 2005

Re: Derating conductors of one circuit

Originally posted by mhulbert: Charlie, Not sure I follow you.

It's probably me who is not following you. I can't draw and post pictures, so let me describe what I think you are trying to build. Then you can tell me where I have it wrong.

You have one breaker (for which the trip rating is to be determined) supplying power to one branch circuit (for which the conductor size is to be determined).

This branch circuit supplies power to a "black box," with the brand name "GrafikEye" on the outside. I have no idea what this box is, nor what it does, nor what is inside it.

From this box you have twelve wires going out (for which the size is to be determined). They all start off in the same conduit, and will branch off to separate loads somewhere downstream.

One pair of wires (from this set of 12) supplies power to a LV transformer. Another pair supplies another LV transformer. In all, six LV transformers are supplied by the 12 wires.

There is no separate overcurrent protection for the 12 wires. So they must be protected by the branch circuit breaker mentioned above.

Therefore, the branch circuit wires to the "black box" and the 12 wires leaving that box must all be the same size, so that the one breaker can protect them all.

The load on any one set of wires (i.e., any one LV transformer) is limited to 800 watts. If the load goes higher, something will burn up (I don't know if there is any internal protection against this happening).

The total load is limited (again, I don't know by what means) to 1920 watts. That is equivalent to 16 amps.

Now we start the design process.

The ampacity of the 12 outgoing wires needs to be high enough to handle their load. That is 6.6 amps, as you have said.

But the ampacity also has to be high enough to be protected by the selected breaker.

The load seen by the branch circuit breaker is 16 amps, and is continuous. That requires us to use a breaker that is at least 125% of that load. A 20 amp breaker will do this job.

Therefore, the conductors must have an ampacity of 20 amps or higher.

The two branch circuit conductors are in a conduit by themselves, and therefore have no derating factor. You could use a #12 for this application, if there were no other consideration.

But there is a 50% derating for the set of 12 outgoing conductors. In order to get a 20 amp ampacity, you need to use #10.

Is this the right configuration and the right design process? Do I have the "black box" all wrong? Are the LV transformers inside the "black box"? Are the 12 wires going someplace different than I have described?

mhulbert · Sep 9, 2005

Re: Derating conductors of one circuit

Charlie,
You got it right on. Thank you for laying it out that way, as I now have a better idea of the sequence in which to determine the ampacity of our conductors. Too often, I use memory, etc in calculating ampacity of wires, or go through all the derating, etc in the wrong order.

As for the "black box", it is really just 6, 800W dimmers in a 4 gang unit. The 1920W total is probably a limit of the heat dissipation of the Grafik Eye, as they used to have one that was fed by 2 15A circuits. I should have posted this at the beginning of this post, here's a link to the PDF cutsheet on said black box. Scroll to page 6, for a wiring diagram: http://www.lutron.com/grafikeye/specs/grx3000.pdf

Again, thanks for the assistance!

Derating conductors of one circuit

mhulbert

Senior Member

infinity

Moderator

iwire

Moderator

iwire

Moderator

bphgravity

Senior Member

iwire

Moderator

charlie b

Moderator

iwire

Moderator

mhulbert

Senior Member

charlie b

Moderator

mhulbert

Senior Member

hardworkingstiff

Senior Member

iwire

Moderator

charlie b

Moderator

charlie b

Moderator

mhulbert

Senior Member