Neher/McGrath Method for Heat Transfer Calcs. of Ductbanks

Status
Not open for further replies.

Jeff Gifford

New member
I am preparing to do some calculations to verify that a 115kV underground ductbank can pass 70MVA continuously after a new ductbank of 480V at 2000 amps is passed perpindicularly underneath it.

My calculation is to find the distance that the two sets of buried conduits must be kept separate to prevent the existing 115kV ductbank from overheating to the point that it limits the current from the mutual heating of the two sets of duct banks.

The problem was solved apparently by a couple of engineers back in 1957 named Neher and McGraph. The calculations are quite complicated, and several electrical calculation software manufacturers can perform this calculation utilizing the Neher/McGraph method.

The problem is though, that they have only solved the problem (to the best of my knowledge from the research I have done thus far) for a parallel set of duct banks. Intuitively, the parallel methodology will give you a worst case scenario since the mutual heating will take place along the entire lenth that they are in the parallel configuration. For a perpendicular configuration, it will almost be like a point calculation. I need to find the minimum distance the ducts can be place above/below each other, and intuitively the perpendicular calculation will give me this minimum value.

I have contacted two popular s/w providers - EDSA and ETAP, and both admit that there s/w will not do the perpendicular calculation. I also called SKM, another popular s/w for performing coordination / arcflash calcs. etc, and they are still working on being able to do the parallel configuration.

My question is, has anyone done these calcs and is there any software that you know about that you can reccommend to help me solve this problem.

Thanks,

Jeff
 

mayanees

Senior Member
Location
Westminster, MD
Occupation
Electrical Engineer and Master Electrician
Skm

Skm

... no assistance as to the method here, but I have to comment on something in your post.

I love SKM to death, and have been using their software since 1992 - when it was DOS based and crashed every 20 minutes... and they've improved their product 10,000 fold since then. I still use it, as well as Easy Power, and I'm sold on SKM hands down...

... but I gotta say, that every time I've asked them for a new application, they ALWAYS say they're working on it, and it should be out this year.. or next...

Two specific apps that come to mind are: Graphic real-time animation of a power systems & Stability modeling of rotary UPS systems.


They're working on both apps, and they should be out this year... or next. :roll:

JM
 
Last edited:

kingpb

Senior Member
Location
SE USA as far as you can go
Occupation
Engineer, Registered
Intuitively, one passing over the other, perpendicular, should not cause much heat transfer from one to the other, averaged out over the length of the run. To reduce or eliminate the heat transfer hotspot directly, you can install insulating board where they cross over. This is commonly used where steam pipes run across, over, under, or near electrical and other services and surfaces that need to be protected.
 

charlie b

Moderator
Staff member
Location
Lockport, IL
Occupation
Retired Electrical Engineer
Jeff Gifford said:
I am preparing to do some calculations to verify that a 115kV underground ductbank can pass 70MVA continuously after a new ductbank of 480V at 2000 amps is passed perpindicularly underneath it.
If I were presented with that configuration, I would not bother with such a calculation. One ductbank passing perpendicular under another will not have any significant influence. Some things you can pass off by a claim of "engineering judgment." For this one, I would not even pick up a pencil to write the words "engineering judgment" on a sheet of paper.

That said, if you have a real need to prove that this is not a significant influence, I have no analytical method to offer. Neher-McGrath does not cover this situation - this geometry was not addressed in their paper.
 

jim dungar

Moderator
Staff member
Location
Wisconsin
Occupation
PE (Retired) - Power Systems
Would it be possible to model the perpendicular cross over area as a very short parallel run?
 
Status
Not open for further replies.
Top