310.10(H) - Conductors in Parallel

[kwired]

If you chart the size vs ampacity, in sizes greater than 1/0, its no longer proportional. The ampacity of 500 mcm is not double that of 250.
So with larger services you get more ampacity per $$ by running in parallel.

Good stuff.
“Engineering supervision” is allowed in Exception 2....so why can’t it be the same for new installations?
Exception 1 allows control cables to be in parallel for voltage drop to instruments. Go figure....

This one area of the Handbook that could use some enhanced explanation.





Sent from my iPhone using Tapatalk

If a conductor is parallel to another conductor with control wiring it generally isn't for the purpose of attaining a higher overall ampacity conductor out of multiple sub elements like it is for power conductors.

Handbook is useful but commentary portions is not official NFPA interpretation of the code either. It does mention this right at the front of the book. Any such enhanced explanation is from opinion and reasearch done by those that write such explanation and not an official NFPA explantion. If you can find when such language is adopted and find the ROP's you should find out why the CMP made the decision they made at that time to put in the language they put in.

 

[infinity]

Years ago, it was common to see paralleled 10-3 NM supplying residential range circuits.

How many years ago? The #1/0 minimum is in the 1965 NEC (oldest edition I currently have access to) and I would guess that it was in there well before that as well.

310-10. Conductors in Multiple. Conductors in sizes 1/0 and largermay be run in multiple provided the arrangement is such as to assure equal division of total current among all conductors involved. All ofthe multiple conductors shall be of the same length, of the same conductormaterial, circular-mil area, same insulation type and terminatedin the same manner. Where run in separate raceways or cables, theraceways or cables shall have the same physical characteristics.

 

[electrofelon]

Parallel small conductors exist All Over America as result of older wiring practices... Ever put an afci breaker in an older residence and it trips immediately?
Chances are neutrals of different circuits are tied together ie paralleled. If they are on different legs, then the installer has unintentionally created a multiwire branch circuit with two neutrals versus one.

As for why you cannot run two sets of 14 / 3 to a dryer receptacle, many breakers do not accept/aren't listed for 2 wires, and I seriously doubt that a 14 30 receptacle's terminals are rated for two wires.

I do not know why the cut off limit is 1 / 0, though it had to be somewhere.

Its kinda surprising now that I think about it, but I dont think I have ever seen small paralleled conductors, except "quasi- paralleled" conductors where it was most likely unintentional like a double fed circuit or neuters tied together somewhere.

 

[LarryFine]

How many years ago? The #1/0 minimum is in the 1965 NEC (oldest edition I currently have access to) and I would guess that it was in there well before that as well.

Not sure when the wiring was done, but likely before 1965. The ones I recall were the old silver-and-tar NM without GEC. They were run so tightly in parallel that they shared staples. The MN could be used for major appliances as long as it had an insulated neutral and originated from the main panel.

 

[Dale001289]

[QUOTE=Carultch;1917415]One reason I suspect, is that the relative manufacturing tolerance is less in larger sizes, which means due to this factor, there is less differences in path resistance and less non-uniform current division among the paths.

For instance, suppose you have a manufacturing tolerance of +/- 1 kcmil. This isn't necessarily realistic, but it is a number we can use to make my point mathematically. On a #10 wire, this is about +/- 10%. On a 1/0 wire, this is +/- 1%. So the relative errors due to manufacturing variability in proportion to total size are less for larger sizes.[/QUOTE]

Southwire cables, are manufactured to a tolerance of +/-2%regardless of kCMIL.

 

[Carultch]

[QUOTE=Carultch;1917415]One reason I suspect, is that the relative manufacturing tolerance is less in larger sizes, which means due to this factor, there is less differences in path resistance and less non-uniform current division among the paths.

For instance, suppose you have a manufacturing tolerance of +/- 1 kcmil. This isn't necessarily realistic, but it is a number we can use to make my point mathematically. On a #10 wire, this is about +/- 10%. On a 1/0 wire, this is +/- 1%. So the relative errors due to manufacturing variability in proportion to total size are less for larger sizes.

Southwire cables, are manufactured to a tolerance of +/-2%regardless of kCMIL. [/QUOTE]

As mentioned, it was just a reason I suspected. Termination manufacturing tolerances might be another reason, if they have additive tolerances as opposed to percentage tolerances.

This article states that the #1/0 and larger rule, is to limit the use of parallel conductors to situations where they are truly needed. Of course, it is subjective to determine where it really is "truly needed", and why it is needed to limit their use in the first place. Because a 300A circuit can be done in a single set of 350's or parallel 1/0's. Some breakers force you to inevitably use parallel cables, like 400A devices with maximum 500 kcmil terminals.
https://iaeimagazine.org/magazine/2013/03/16/parallel-conductors-revisited/

 

[Dale001289]

Southwire cables, are manufactured to a tolerance of +/-2%regardless of kCMIL.

As mentioned, it was just a reason I suspected. Termination manufacturing tolerances might be another reason, if they have additive tolerances as opposed to percentage tolerances.

This article states that the #1/0 and larger rule, is to limit the use of parallel conductors to situations where they are truly needed. Of course, it is subjective to determine where it really is "truly needed", and why it is needed to limit their use in the first place. Because a 300A circuit can be done in a single set of 350's or parallel 1/0's. Some breakers force you to inevitably use parallel cables, like 400A devices with maximum 500 kcmil terminals.
https://iaeimagazine.org/magazine/2013/03/16/parallel-conductors-revisited/[/QUOTE]

Thanks for sharing.


Sent from my iPhone using Tapatalk

 

[roger]

Not sure when the wiring was done, but likely before 1965. The ones I recall were the old silver-and-tar NM without GEC. They were run so tightly in parallel that they shared staples. The MN could be used for major appliances as long as it had an insulated neutral and originated from the main panel.

Here's a good one I found

Roger

 

[kwired]

Here's a good one I found

Roger

Is that a parallel supply or leads to additional outlets? I've seen leads to additional outlets several times on such receptacles, don't know why you would want to fight terminating them in such tight space though.

 

[JFletcher]

That Tombstone receptacle cannot be compliant, there's no way you can put three sets of Romex through that clamp... And it appears that half the clamp is missing anyway and somebody probably broke the plastic in the bottom to physically get the wires in there.

I seriously doubt that is a parallel installation, and it would have been so much easier just to mount a junction box a foot or so away from that Tombstone and make all the splices there.

Edited to add... where I would consider parallel conductors smaller than 1/0 acceptable would be a ring circuit for low-voltage wiring, not so much paralleling wire, but feeding the lights in a ring to reduce voltage drop.

 

[AtTheKeyboard]

Turn the world into a circus?

Turn the world into a circus?

I don't know what the past generations of codemakers were thinking, but I agree with the decision to limit the size at which paralleling begins - well... ...to avoid a circus.

I could easily see Contractors changing to just a few sizes of wire spools if parallel #12's could be used. Parallel runs filling up exactly 40% of the conduit, etc. It would be so hard to tell what was going on after a panel schedule was lost, etc. - just a forest of wires, in different conduits.

I wonder if this clutter-avoidance is part of it, or if there is an actual technical reason. For me, I'm glad I can generally track how loads are fed by the conductors that feed them - especially in old installations. I can only imagine how messy panel interiors would be without some restriction.

 

[kwired]

I don't know what the past generations of codemakers were thinking, but I agree with the decision to limit the size at which paralleling begins - well... ...to avoid a circus.

I could easily see Contractors changing to just a few sizes of wire spools if parallel #12's could be used. Parallel runs filling up exactly 40% of the conduit, etc. It would be so hard to tell what was going on after a panel schedule was lost, etc. - just a forest of wires, in different conduits.

I wonder if this clutter-avoidance is part of it, or if there is an actual technical reason. For me, I'm glad I can generally track how loads are fed by the conductors that feed them - especially in old installations. I can only imagine how messy panel interiors would be without some restriction.

I figured it out if there would be no such rules and if 240.4(D) were not an issue that if you needed to run a 200 amp 3 phase feeder with same size neutral and used the raceway for EGC, if you want to use 12 AWG you would need minimum of 8 conductors per phase, but then comes ampacity adjustments. After ampacity adjustments if you still want to stay with 12 AWG THHN/THWN conductors you would need 20 conductors per phase and with same size neutral that is 80 #12AWG in one raceway - would need to be in 2 inch raceway. EGC should still need to be just one 6 AWG copper.

20 times per foot on conductor cost for 12 AWG vs 1 time per foot for 3/0AWG - looked at online price of one of my suppliers the 12 AWG method is about 18 cents per foot times 4 (ABC&N) = 72 cents per foot less for the feeder. Can't imagine it would save on labor at all and possibly even be more labor in many instances to manage all those individual conductors while installing them.

Aluminum feeder with single conductor per phase should be significantly less.

 

[ggunn]

For one thing, the larger the conductor the less difference in resistance the same difference in conductor length makes.

 

[AtTheKeyboard]

Regarding costing - I was generally referring to the clutter potential, not intending to generate a cost-based concept.