Cable tray sizing

[Julius Right]

In my opinion you need only 2.15*1.024” clearance between group of cables in square configuration
as per 392.80 Ampacity of Conductors. (A) Ampacity of Cables, Rated 2000 Volts or Less, in Cable Trays.
(2) Single-Conductor Cables. (d) Where single conductors are installed in a triangular or square configuration in uncovered cable trays.
In this case the ampacity will be as per 310.15(B)(20)-for 75oC rated copper conductor 600 mcm 553 A, for aluminum 440 A.
Nevertheless you have to check art.310.15 Ampacities for Conductors Rated 0–2000 Volts. (B) Tables. (5) Neutral Conductor.

 

[fandi]

Sounds like you are making progress... :thumbsup:

Note there is no specific "gap" required, but you will have to keep each set together (conductors immediately adjacent to each other).

MC cable has a wider diameter than THHN in single conductor... so that'll not save you anything on width.

Multiconductor MC may save you width, but you then have the handling issues... 4C/600kcmil MC is no picnic getting in tray.

Per Southwire Simpull THHN, #600kcmil OD = 1024 mils or 1.024 inch. #250 kcmil OD = 694 mils or 0.694 inch.
There will be 5 sets of THHN single conductor cables:
Set 1:
- Bottom: (3) #600kcmil (3 hot)
- Top: (1) #600 kcmil (neutral) and (1) #250kcmil EGC.
Set 2 thru Set 5: same as Set 1 except there's no EGC.

And I have last confusion about which of the following method I should use before finalizing the calc here:
- Method 1: Add all the area (in square inches) of all cables in tray: 20 x #600kcmil area + 1 x #250kcmil area = 16.838 square inches.
Per Table 392.22(B)(1) Column 1, the maximum allowable fill area of 16 inch wide tray is 17.5 square inches so 16 inch wide tray is good.
- Method 2: Per 392.80(A)(2)(d), add all the bottom diameters of each set (3 x 1.024" in this case) and add minimum 2.15 x OD for the gap between each sets (it seems NEC doesn't require the gap though):
3 x 5 x 1.024" +2.15 x 4 x 1.024" = 24.16" so 24" wide tray can be used.

On a side note, per section 392.80(A)(2)(a), the ampacities for 600kcmil or larger single conductor cable in uncovered cable tray shall not exceed 75% of the allowable ampacities in Table 310.15(B)(17), it's looks like 600kcmil is overkill: #600kcmil single conductor has the ampacity of 690A at 75 degree C, 75% of 690A = 517.5A > 400A, maybe I should use less sets.

 

[Smart $]

...it's looks like 600kcmil is overkill... maybe I should use less sets.

As the old cliche goes, "back to the drawing board"... :rant:



...but don't forget that your maximum circuit ampacity is per 110.14(C) limitations... and that always use Table 310.15(B)(16)... unless you use the remote "transition" method* to increase tray ampacity.

* see pg 4 here: http://www2.schneider-electric.com/...7/en_US/Wire Terminations 0110DB9901R2-02.pdf

Note cable tray is the only wiring method I know that doesn't require an enclosure for splices (i.e. transition terminations).

 

[fandi]

As the old cliche goes, "back to the drawing board"... :rant:



...but don't forget that your maximum circuit ampacity is per 110.14(C) limitations... and that always use Table 310.15(B)(16)... unless you use the remote "transition" method* to increase tray ampacity.

* see pg 4 here: http://www2.schneider-electric.com/...7/en_US/Wire Terminations 0110DB9901R2-02.pdf

Note cable tray is the only wiring method I know that doesn't require an enclosure for splices (i.e. transition terminations).

When I treat 2000A connection as (5) sets of 400A, I used table 310.15(B)(16) to decide I need to use 600kcmil which is good up to 420A at 75 degree C.
Now, this is cable tray so table 310.15(B)(17) shall be used for single conductor in free air. So the 600kcmil ampacity is now 690A at 75 degree C.
That's a big difference.

 

[Smart $]

When I treat 2000A connection as (5) sets of 400A, I used table 310.15(B)(16) to decide I need to use 600kcmil which is good up to 420A at 75 degree C.
Now, this is cable tray so table 310.15(B)(17) shall be used for single conductor in free air. So the 600kcmil ampacity is now 690A at 75 degree C.
That's a big difference.

Yes, but your circuit ampacity still cannot exceed 110.14(C) limitations, which uses Table 310.15(B)(16) always. If you went with say 5 sets of 500kcmil your circuit ampacity cannot exceed 380A x 5... and that would not be protected per 240.4(C)... so you have to go with 5 x 600kcmil... at least out of the box and back in the box at the other end.

In between, you can use 392/Table 310.15(B)(17) values as long as you make a qualified "transition" away from the box at each end.

 

[fandi]

Yes, but your circuit ampacity still cannot exceed 110.14(C) limitations, which uses Table 310.15(B)(16) always. If you went with say 5 sets of 500kcmil your circuit ampacity cannot exceed 380A x 5... and that would not be protected per 240.4(C)... so you have to go with 5 x 600kcmil... at least out of the box and back in the box at the other end.

In between, you can use 392/Table 310.15(B)(17) values as long as you make a qualified "transition" away from the box at each end.

With the conductor rated at 90 degree C and the 2000A circuit breaker terminations at two ends rated at 75 degree C, I don't see any problem with 110.14(C) limitations.
But I agree with you that the connection between the circuit breakers at two ends with the cable tray is in the box and not free air.

 

[Smart $]

With the conductor rated at 90 degree C and the 2000A circuit breaker terminations at two ends rated at 75 degree C, I don't see any problem with 110.14(C) limitations.

There isn't any using 600kcmil, but you can't go any smaller in the boxes. You can go smaller in the tray if you make a proper transition. That's all I'm trying to say.

 

[Julius Right]

I think so will be better:

 

[dema]

Ground size

Ground size

It was mentioned, but I'm not sure it was comprehended - the required ground size for multiple runs is greater than for single runs. When you have two sets of 500MCM, you are required to have a 1/0 ground in each run - yes two 1/0 grounds for twice the current instead of one #3. 3-600MCM's require a 3/0 in each run. See 250.122 and the explanation in the handbook.

 

[Smart $]

It was mentioned, but I'm not sure it was comprehended - the required ground size for multiple runs is greater than for single runs. When you have two sets of 500MCM, you are required to have a 1/0 ground in each run - yes two 1/0 grounds for twice the current instead of one #3. 3-600MCM's require a 3/0 in each run. See 250.122 and the explanation in the handbook.

Not true.

250.122 in part says:

Where conductors are installed in parallel in the same raceway,
cable, or cable tray as permitted in 310.10(H), a single
equipment grounding conductor shall be permitted. Equipment
grounding conductors installed in cable tray shall meet
the minimum requirements of 392.10(B)(1)(c).

A single 250kcmil Cu EGC is all that is required.

 

[Carultch]

It was mentioned, but I'm not sure it was comprehended - the required ground size for multiple runs is greater than for single runs. When you have two sets of 500MCM, you are required to have a 1/0 ground in each run - yes two 1/0 grounds for twice the current instead of one #3. 3-600MCM's require a 3/0 in each run. See 250.122 and the explanation in the handbook.

That is only true when you route them in separate raceways, or separate wiring structures in general.

You need to have 1 full size EGC per raceway, or per wiring structure. If you parallel the sets in separate wiring structures, each wiring structure needs a full size EGC. If you parallel the sets in the same wiring structure, one full size EGC will suffice for all sets within that wiring structure. If you parallel the sets two per structure in 3 structures for a total of 6 sets, then you need a total of 3 EGCs.

Is there a better NEC term I should be using instead of "wiring structure"?

 

[Smart $]

That is only true when you route them in separate raceways, or separate wiring structures in general.

You need to have 1 full size EGC per raceway, or per wiring structure. If you parallel the sets in separate wiring structures, each wiring structure needs a full size EGC. If you parallel the sets in the same wiring structure, one full size EGC will suffice for all sets within that wiring structure. If you parallel the sets two per structure in 3 structures for a total of 6 sets, then you need a total of 3 EGCs.

Is there a better NEC term I should be using instead of "wiring structure"?

The NEC calls them Chapter 3 Wiring Methods in the general sense... but a coined term is required in this case case IMO, because with the cable tray method, you can have single conductors, multiconductor cables, and conduit raceways all in the the same tray. The latter two being a wiring method within a wiring method.