Underground duct bank

Tainted

Tainted

Senior Member
Location
New York
Occupation
Engineer (PE)
I am designing (14) 2 inch underground duct bank concrete encased RGS conduits. 7 conduits on top and 7 conduits on the bottom. The voltage is low at 208V

How far apart must these conduits be? Annex B and figure 310.60 suggesting they must be 7.5 inches center to center however I think this is overkill for this voltage

Can these 2 inch conduits be 3 inches apart from each other? This will make them 5 inches center to center. Or are there other factors I'm missing here?
 
The recommended concrete thickness between ducts is 3” in order to assure duct bank resistance. Different construction organization presents more or less the same pattern.
See-for instance Distribution Ductbank - NStar
 
Is this for something that is expected to have a high load factor? If yes, then I think you should either be performing Neher-McGrath to determine duct bank ampacity or following the conduit configurations of Annex B to a tee.

If no then I think you can make the spacing closer. I typically design around available spacing configurations supported by duct bank conduit spacers - something like Carlon Snap-Loc.
 
PD1972 said:
Is this for something that is expected to have a high load factor? If yes, then I think you should either be performing Neher-McGrath to determine duct bank ampacity or following the conduit configurations of Annex B to a tee.

If no then I think you can make the spacing closer. I typically design around available spacing configurations supported by duct bank conduit spacers - something like Carlon Snap-Loc.
Click to expand...
No high load factor here, it's just for 3/0 feeders for apartments
 
PD1972 said:
Is this for something that is expected to have a high load factor? If yes, then I think you should either be performing Neher-McGrath to determine duct bank ampacity or following the conduit configurations of Annex B to a tee.

If no then I think you can make the spacing closer. I typically design around available spacing configurations supported by duct bank conduit spacers - something like Carlon Snap-Loc.
Click to expand...
I think I should still do Neher-McGrath calculations to make sure but doing that by hand is a disaster. Is there a free software for this?
 
Tainted said:
I think I should still do Neher-McGrath calculations to make sure but doing that by hand is a disaster. Is there a free software for this?
Click to expand...
I am unaware of free software that will do these calcs for you. Either slog through the 1957 paper to make your own spreadsheet or pay for it.

Not trying to convince you not to do your due diligence, but if these are feeders for apartments that are sized off NEC 220 I wouldn't worry too much about it as you are going to have a very low load factor. The heating of a duct bank is an issue if the loads have a high enough load factor where the "fluff" inherent in an NEC 220 calculation is not enough to account for the ampacity derating of the conductors due to the heat build up in the duct bank.
 
PD1972 said:
I am unaware of free software that will do these calcs for you. Either slog through the 1957 paper to make your own spreadsheet or pay for it.

Not trying to convince you not to do your due diligence, but if these are feeders for apartments that are sized off NEC 220 I wouldn't worry too much about it as you are going to have a very low load factor. The heating of a duct bank is an issue if the loads have a high enough load factor where the "fluff" inherent in an NEC 220 calculation is not enough to account for the ampacity derating of the conductors due to the heat build up in the duct bank.
Click to expand...
You are right, I think I'm just gonna space it all out 3" evenly and not worry about derating and call it a day.
 
If we take the distance between conduits 5.5"[2.5+3] and according to NEC Annex B concrete thermal resistance 55 ,earth RHO 90, depth of duct bank 30",earth temperature 20oC, conductor temperature 75oC for 3*4/0 copper cables per 2" duct [RMC] we get 131 A ampacity [Neher and McGrath system].
If we take the distance between conduits 7.5" as per NEC, we get 140 A.




 
Julius Right said:
If we take the distance between conduits 5.5"[2.5+3] and according to NEC Annex B concrete thermal resistance 55 ,earth RHO 90, depth of duct bank 30",earth temperature 20oC, conductor temperature 75oC for 3*4/0 copper cables per 2" duct [RMC] we get 131 A ampacity [Neher and McGrath system].
If we take the distance between conduits 7.5" as per NEC, we get 140 A.
Click to expand...
What did you have for the load factor? Apartments will be ~30-50%, which might bring those ampacities closer
 
Julius Right said:
If we take the distance between conduits 5.5"[2.5+3] and according to NEC Annex B concrete thermal resistance 55 ,earth RHO 90, depth of duct bank 30",earth temperature 20oC, conductor temperature 75oC for 3*4/0 copper cables per 2" duct [RMC] we get 131 A ampacity [Neher and McGrath system].
If we take the distance between conduits 7.5" as per NEC, we get 140 A.
Click to expand...
I was under the impression that we can neglect de-rating because it's not really a strenuous load and that there was an outcry long ago which removed code requirement for derating low voltage duct bank systems

Also my system is 2 inch conduits spaced out 5 inches center to center with 3#3/0 wires inside. 7.5 inches center to center is too large I want to make it more compact
 
Last edited:
ron said:
What did you have for the load factor? Apartments will be ~30-50%, which might bring those ampacities closer
Click to expand...
I don't follow how ampacity can depend on load factor, but I'm not particularly familiar with duct bank calculations, can you elaborate?

Per the NEC ampacity is a continuous current rating. So I would think the ampacity of the conductors in the duct bank would be determined by finding the continuous current that all the conductors could carry such the steady state temperature is equal to the conductor insulation rating.

Obviously for other periodic loading profiles, the peak current at which the peak periodic steady state temperature is the maximum allowable will differ from that value. For example, if only 50% of the conductors are carrying current. Or if the conductors are loaded only 50% of the time, and carry no current the other 50% of the time.

But I wouldn't think that ampacity is an appropriate term to refer to those peak allowable currents.

Thanks,
Wayne
 
In my opinion, due mechanical requirement, the minimum concrete thickness between conduit has to be 3". The outside diameter of the 2" conduit it is 2.375" [by the way, the "duct" is a conduit of more than 3" diameter [usually 4" to 6"].
So, the minimum distance centre line-to-centre line it is 5.375”.
Then, in the same conditions except 3*3/0 awg the ampacity for 2*7*3 cables in two rows and 7 columns will be 118 A.
 
Julius Right said:
In my opinion, due mechanical requirement, the minimum concrete thickness between conduit has to be 3". The outside diameter of the 2" conduit it is 2.375" [by the way, the "duct" is a conduit of more than 3" diameter [usually 4" to 6"].
So, the minimum distance centre line-to-centre line it is 5.375”.
Then, in the same conditions except 3*3/0 awg the ampacity for 2*7*3 cables in two rows and 7 columns will be 118 A.
Click to expand...
My take is that what you calculated is the ampacity of conductors due to real-time load and not the NEC load.

For example if my NEC calculation is 200 amps but the real-time actual load is less than 118 amps then it should be fine.

There's no code requirement for sizing ampacity of duct bank on a 208V system like mine plus there is no way that an apartment will have more than 118 amps for a long period of time.

I will use 4/0 just to be safe
 
In my calculation I considered 100% load factor. According to Neher and McGrath you may consider another load factor as you wish.
For 50% load factor the ampacity will be 175 A.
 
Julius Right said:
In my calculation I considered 100% load factor. According to Neher and McGrath you may consider another load factor as you wish.
For 50% load factor the ampacity will be 175 A.
Click to expand...
Can I still use a 200 amp breaker if the actual real-time load will be less than 175 amps?

Reason I say this is because there is no NEC requirement to size the breaker based on the ampacity of conductors using Neher and McGrath calculation on 208V system
 
Tainted said:
How far apart must these conduits be? Annex B and figure 310.60 suggesting they must be 7.5 inches center to center however I think this is overkill for this voltage
Click to expand...
On the one hand, NEC 310.60 only applies above 2000V. (2017) NEC Table 310.15(B)(16) does not place any limits on the location of the conduit containing the conductors, so the table values would apply within your duct bank. So for NEC purposes, at 208V, you get to use the ampacity values in the table, which do not depend on duct spacing.

On the other hand, 2017 NEC 310.14(A)(3) says "No conductor shall be used in such a manner that its operating temperature exceeds that designated for the type of insulated conductor involved. In no case shall conductors be associated together in such a way, with respect to type of circuit, the wiring method employed, or the number of conductors, that the limiting temperature of any conductor is exceeded." It seems reasonable to consider all the conductors within the duct bank as "associated together."

So based on the calculation in post #12, seems like it would be a violation of 310.15(A)(3) to operate the duct bank with all conductors at a continuous current above 118A. Per the definition of ampacity, that would put the conductor ampacity at 118A, not the 200A from Table 310.15(B)(16).

Not sure how to reconcile these separate provisions.

Cheers, Wayne
 
wwhitney said:
On the one hand, NEC 310.60 only applies above 2000V. (2017) NEC Table 310.15(B)(16) does not place any limits on the location of the conduit containing the conductors, so the table values would apply within your duct bank. So for NEC purposes, at 208V, you get to use the ampacity values in the table, which do not depend on duct spacing.

On the other hand, 2017 NEC 310.14(A)(3) says "No conductor shall be used in such a manner that its operating temperature exceeds that designated for the type of insulated conductor involved. In no case shall conductors be associated together in such a way, with respect to type of circuit, the wiring method employed, or the number of conductors, that the limiting temperature of any conductor is exceeded." It seems reasonable to consider all the conductors within the duct bank as "associated together."

So based on the calculation in post #12, seems like it would be a violation of 310.15(A)(3) to operate the duct bank with all conductors at a continuous current above 118A. Per the definition of ampacity, that would put the conductor ampacity at 118A, not the 200A from Table 310.15(B)(16).

Not sure how to reconcile these separate provisions.

Cheers, Wayne
Click to expand...
My interpretation is that "associated together" means conductors associated within the same conduit. If conductors are in separate conduits then they are not really considered "associated together"
 
Tainted said:
My interpretation is that "associated together" means conductors associated within the same conduit. If conductors are in separate conduits then they are not really considered "associated together"
Click to expand...
I'm going to disagree, as there are many sections that discuss the case of conductors within the same conduit, and they never use the phrase "associated together". The phrase "associated together" is only used 3 times in the 2017 NEC; it is an intentional choice made to cover a broader category of interactions than just "within the same conduit."

In a duct bank the conductors in one conduit obviously affect the thermal performance of the conductors in another conduit; that's the reason you have to consider the behavior of the entire duct bank together, rather than just each conduit individually. That situation is clearly under the purview of 310.14(A)(3).

Cheers, Wayne
 
wwhitney said:
I'm going to disagree, as there are many sections that discuss the case of conductors within the same conduit, and they never use the phrase "associated together". The phrase "associated together" is only used 3 times in the 2017 NEC; it is an intentional choice made to cover a broader category of interactions than just "within the same conduit."

In a duct bank the conductors in one conduit obviously affect the thermal performance of the conductors in another conduit; that's the reason you have to consider the behavior of the entire duct bank together, rather than just each conduit individually. That situation is clearly under the purview of 310.14(A)(3).

Cheers, Wayne
Click to expand...
I agree duct bank conductors affect thermal performance to nearby conductors. What I am saying however is if code does not provide guide for 208V systems on how to calculate the ampacity in duct bank, then I can omit the 118A calculation in post #12
 
You must log in or register to reply here.
Top