AC and DC conductors

[ggunn]

300.2(C) allows AC and DC conductors in the same raceway as long as they all have insulation ratings as high as the highest circuit voltage. 690.31(C) allows DC conductors in the same enclosure enclosure as inverter output conductors when they are separated by a barrier or partition. Does the partition obviate the need to have the insulation all at least rated as high as the maximum voltage, i.e., can there be 1000VDC circuits in with inverter output circuits in 600V THWN-2 if there is a barrier?

 

[wwhitney]

Quick correction on the references, for 2020 NEC they are 300.3(C) and 690.31(B).

Not sure as to your question, but on first reading I would say "no." That's purely from comparing 300.3(C) and 690.31(B). They both use the same language of "same equipment wiring enclosure, cable, or raceway," while 690.31(B) has the "unless" clause about a barrier or partition, but 300.3(C) doesn't. So I don't see any allowance in 300.3(C) for using a barrier or partition to satisfy that section, and the way 690.31(B) is written shows that such an explicit allowance would be necessary.

On the other hand, I would say that a sufficiently beefy partition could be used to turn one enclosure into two enclosures. E.g. one that is twice the thickness of the enclosure itself, duplicating the material thicknesses you would have if two separate enclosures were touching.

Cheers, Wayne

 

[ggunn]

Thanks or the reference correction. Put more succinctly: Can we put 1000V DC conductors in a gutter with 480/277 inverter output THWN-2 conductors if there is a partition or barrier per 690.31(B)?

 

[wwhitney]

I'd say no, the proposed configuration would comply with 690.31(B) but still violate 300.3(C).

Chapter 6 can modify Chapter 3, so if the two sections were on identical topics, such as voltage only, then it would be fine. But 300.3(C) is about voltage, and the word voltage doesn't appear in 690.31(B) proper. So the 690.31(B) allowance wouldn't give you any relief from the voltage requirements in 300.3(C).

Cheers, Wayne

 

[ggunn]

I'd say no, the proposed configuration would comply with 690.31(B) but still violate 300.3(C).

Chapter 6 can modify Chapter 3, so if the two sections were on identical topics, such as voltage only, then it would be fine. But 300.3(C) is about voltage, and the word voltage doesn't appear in 690.31(B) proper. So the 690.31(B) allowance wouldn't give you any relief from the voltage requirements in 300.3(C).

Cheers, Wayne

OK, but 300.3(C) doesn't mention a partition. Common sense says (yeah, yeah, I know...) that the reason 300.4(C) is worded that way is that it assumes that the conductors can come in contact with one another. If there is a barrier, they cannot, which is the point of the barrier.

 

[wwhitney]

OK, but 300.3(C) doesn't mention a partition.

That's right, it doesn't have the "unless there's a partition language." So to me that means a partition is immaterial.

Oddly enough, the presence of the "unless" language in 690.31(B) means that the lack of the language in 300.3(C) is more definitive..

Cheers, Wayne

 

[ggunn]

That's right, it doesn't have the "unless there's a partition language." So to me that means a partition is immaterial.

Oddly enough, the presence of the "unless" language in 690.31(B) means that the lack of the language in 300.3(C) is more definitive..

Cheers, Wayne

We have a test case coming up in a territory controlled by an AHJ who are sticklers for the letter of the law, so I will let you know how it turns out. Meanwhile, I'd like to hear from the rest of you guys.

 

[wwhitney]

You can't just use two different "gutters"? (Likely wireways)

Cheers, Wayne

 

[jaggedben]

Thanks or the reference correction. Put more succinctly: Can we put 1000V DC conductors in a gutter with 480/277 inverter output THWN-2 conductors if there is a partition or barrier per 690.31(B)?

Can you use 1000V conductors for the AC in that location? (Is that less expensive than separate enclosures?)

 

[ggunn]

Can you use 1000V conductors for the AC in that location? (Is that less expensive than separate enclosures?)

Possibly; I don't know, but that merely dodges the point without addressing it. I hope that the writers of the extensively amended content in 690.31(B) in the 2020 NEC took into account that most PV systems connected to 480/277V services these days have DC voltages that require 1000V conductors.

 

[pv_n00b]

The purpose of the partition is to provide physical separation of the conductors in the different circuits, not electrical isolation. So all the conductor insulation still needs the rating of the highest-rated conductor.

 

[wwhitney]

Possibly; I don't know, but that merely dodges the point without addressing it. I hope that the writers of the extensively amended content in 690.31(B) in the 2020 NEC took into account that most PV systems connected to 480/277V services these days have DC voltages that require 1000V conductors.

690.31(B) is silent on voltage rating. There's zero reason to think that 300.3(C) doesn't still apply.

Cheers, Wayne

 

[don_resqcapt19]

In general, I treat a partition in a wireway as two separate raceways.

 

[ggunn]

690.31(B) is silent on voltage rating. There's zero reason to think that 300.3(C) doesn't still apply.

I realize that that is your opinion.

 

[jaggedben]

Possibly; I don't know, but that merely dodges the point without addressing it....

Was it not addressed yet? I agree with Wayne. In your scenario you can put the circuits in the same raceway only if there's a partition and all conductors are rated 1000V or more. And I think there's no real debate about it.

 

[wwhitney]

the same raceway

The remaining debate is alluded to by Don's post. When does a partition make one wireway into two wireways?

I would argue certainly yes in the case the partition fully divides the interior volume into two two non-communicating spaces, and the partition has a thickness of at least twice the enclosure wall thickness. As then it duplicates the spatial arrangement of material that you'd get by just putting two wireways touching each other.

Other cases aren't so clear to me.

Cheers, Wayne

 

[jaggedben]

The partition does not have to be made of the same material as the enclosure. (It's actually unclear in 690 what it is and isn't allowed to be made of. Is cardboard or paper okay?) If it's not, I don't think you have multiple raceways.

So to consider the compartments separate raceways, I would add to your criteria that the partition fully divide the enclosure, that it also be the same material. Probably also that it be fixed in place?

Raceways such as a wireway generally have listings or at least Chapter 3 articles and definitions, and not meeting the above criteria isn't going to change how many raceways there are. I haven't seen any wireways with factory partitions. I've seen some perfunctory partitions in equipment that don't create two enclosures. So, practically speaking, I don't see that the 'it becomes two raceways' argument will have real-world utility when it comes to installation. I think my post #15 pretty much sums it up.

 

[don_resqcapt19]

The remaining debate is alluded to by Don's post. When does a partition make one wireway into two wireways?

I would argue certainly yes in the case the partition fully divides the interior volume into two two non-communicating spaces, and the partition has a thickness of at least twice the enclosure wall thickness. As then it duplicates the spatial arrangement of material that you'd get by just putting two wireways touching each other.

Other cases aren't so clear to me.

Cheers, Wayne

The factory supplied partitions for any wireway that I have ever seen are the same thickness as the wireway itself and are intended to separate the wireway into two separate parts for exactly this purpose. That is the purpose of having high voltage conductors on one side and low voltage conductors on the other side with the conductor insulation being only suitable for their respective circuit voltages.

 

[ggunn]

My interpretation is that the barrier requirement in 690.31(B) replaces/overrules the voltage restrictions in 300.4(C), backed up by a common sense consideration that if the wires cannot come into contact the relative voltage and insulation ratings are irrelevant, along with the fact that 690.31(B) does not refer to 300.4(C) and say that it applies. Of course, it's neither my interpretation nor anyone else's but that of the AHJ that is important, but that is how I am going to present my case if the AHJ challenges our installation.

 

[wwhitney]

My interpretation is that the barrier requirement in 690.31(B) replaces/overrules the voltage restrictions in 300.4(C), backed up by a common sense consideration that if the wires cannot come into contact the relative voltage and insulation ratings are irrelevant, along with the fact that 690.31(B) does not refer to 300.4(C) and say that it applies.

That's a bad argument, as the default is that 300.3(C) still applies. That is well established by 90.3, which says Chapter 6 "may supplement or modify" Chapter 3. 690.31(B) would need to explicitly modify 300.3(C) to provide any relief.

Just go with Don's argument that "partitions in a wireway are a standard way to comply with 300.3(C) and are sold for that purpose." Assuming that is the type of partition you are going to use.

Cheers, Wayne