None Linear Reactance of Wire 3

[Mbrooke]

What is behind the none linear drop in conductor reactance? Is it the stranding of manufacturing variants between gauges? Or something else?


For example 8 gauge and 1 gauge wire goes up in reactance, and #6 while having a lower reactance than #8 is still higher than #10

 

[waross]

I am going to guess a typo.
Have you been able to verify this information from a second source?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Mbrooke]

Have not been able to verify. Its from NFPA 70, so I'm less inclined to assume a typo.

 

[waross]

Do you have any earlier editions to check?
Solid to stranded?
Thicker insulation?
Interesting that those inconsistencies are not present in the next column for ferrous conduit.
However, looking further down the table, there are inconsistencies in the ferrous column that are not present in the no-ferrous column.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Mbrooke]

The tables look the same but I'll have to do more digging.

Nothing about this table seems to make sense 100%.

 

[DM61850]

I would guess different assumptions were made on the conduit and/or it being used as power cable. The inductance is going to change with the conduit and if you are running power cable and the three phases are in the same run, the inductance will be lower the closer they are to each other. You aren't going to serve three phase anything with small gauge wire.

 

[Mbrooke]

That table is based on 3 phase for all wire sizes, at least it claims so:

 

[bacon4life]

Table 9 of NEC/NFPA70 shows the same values for both the 2005 and 2020 editions.

In addition, the effective Z at 0.85 pf does not seem to be calculated from the formula in footnote 2. For example, #8cu in AL conduit should be 2.185 instead of 2.26.

 

[Mbrooke]

I've noticed that too. Makes me wonder.

 

[7anoter4]

NEC Table 9 notes:
1. These values are based on the following constants: UL-Type RHH wires with Class B stranding, in cradled configuration.
If we shall use the formula which is indicated by most of manufacturer [Okonite,Prismian,Kerite]
x=2.π.f.(0.1404.log10(S/r)+0.0153)/10^3 and considering S=insulated[RHH]cable diameter*1.15 we shall get the following results for 10,8,6 awg:
0.1353,0.1423,0.1332 [ohm/km] and multiplying by 1.2 we get 0.1623,0.1708,0.160
That means 0.1-4% difference with respect the NEC.

 

[cuky2000]

Here is another approach determining that the NEC Table 9 has a reactance deviation max. under 6%.
For most of the LV calculations, the NEC table 9 is acceptable data considering that the resistance of the conductors is the dominant factor to determine the total impedance of the conductor.

See below for more detail information on this matter.

---------

 

[Mbrooke]

Star given! Validated somewhat by this, but you are better:

https://www.mikeholt.com/img/product/pdf/1038225525sample.pdf

Is the table correct enough to use for larger sizes? And in general is the formula Ze= R x PF + XL x sin of {arccos PF} accurate enough? Is it worse case or give or take close enough?

 

[cuky2000]

As we said earlier, the data on Table 9 for reactance, AC resistance, and effective Impedance at 85% PF, is accurate enough for most practical applications.
As a general comment, we should remember that the AC resistance is larger than DC resistance because of the skin effect. DC data it is not recommended to be used for accurate calculations in AC systems.

The reactance, on the other hand, depends upon the cable dimension (metallic conductor, Insulation & jacket) and the conductor arrangement in the raceway due to the natural random lay. This can only can be estimated by assumption.

It should be noted that the 600 V cables listed in table 9 there is a combination of 7 insulation thickness (15-70 mils) and 6 jackets thickness (4-9 mils) that definitely will impact the lineality of cable reactance data. That's why I do suspect the values were determined based on an actual test of typical rather than calculation.

In summary, I do believe that for practical purposes, the data in the NEC table 9 is accurate enough for most AC LV calculations

 

[wroggent]

See proposal 6-251 on page 640.

https://www.nfpa.org/Assets/files/AboutTheCodes/70/NEC-TCRA-1986-Articles 516 to Chapter 9.pdf

Maybe NFPA still has the calculations somewhere? I couldn't find anything.

If you're curious page 390 here

https://www.nfpa.org/Assets/files/AboutTheCodes/70/NEC-TCRA-1983-Articles 551 to Chapter 9.pdf

is the 1984 version of the table. I believe prior to 1984 there was just a table with DC resistance values and another table with factors for multiplying to account for skin effect.