Annex B and Duct Bank De-ratings

[bonding jumper]

Hello,
I am trying to determine the de-ratings on various ductbanks for a project, the banks are either 2, 8, 10 or 14 ducts per bank, but all are maximum 2 rows across, so 2X1, 4X2, 5X2 and 7X2. For my 8, 10 & 14 banks, can I use figure B-310-2 Detail 3? I know the quantity of ducts is only 6 in the bank shown, but since I have a maximum of 2 rows, each duct is adjacent to a maximum of 5 ducts (including diagional). Thanks, I don't want to have to do a whole Neher McGrath equation for each bank. Thanks.

 

[charlie b]

Re: Annex B and Duct Bank De-ratings

My version of a ?short answer? is that you would not like my version of a ?long answer.?

My opinion is that if you do not exactly match the configurations shown in Details 1 through 10, then you cannot use any of the ampacities in Annex B. Even if you do match one of the details, you still must take care in justifying your choices for the RHO and ?LF? columns.

As to applying the Neher McGrath equation, that is no easy task. The formula given in 310.15(C) is of no value whatsoever. I know of only two ways to handle the situation you are describing. One way is to have a calculation performed ?under engineering supervision.? There are several software packages that can handle this type of calculation. The other way is to disregard the fact that you are dealing with an underground installation, and just take the ampacities from Table 310.16. I have seen this done, and seen it accepted by a group of state-level inspectors. I would offer the caution that this later approach is absolutely invalid, and that it would even be unsafe, unless the conductor ampacity, as taken from the Table, exceeds the required ampacity, as calculated per Article 220. In other words, you need to be able to take some credit for the ultra-conservative nature of the NEC Service Calculation process, if you are going to give up conservatism, by ignoring the mutual-heating effects of underground ducts.

 

[bonding jumper]

Re: Annex B and Duct Bank De-ratings

Thanks for the response Charlie, Do you know of a particular software package that would perform this operation? I work for an engineering firm and we can do the calculations, so basically we are "engineering supervision" but nobody wants to do the calculations by hand. Another firm reviewing our work asked us for our load bank calculations. My co-worker told me to give him an excel sheet that says 4 conduits X 2 rows = 8 conduits. We kept the rows to a maximum of 2 just so we don't have conductors surrounded by other conductors, thus minimizing the negative effects of being in a bank. The only software I have found so far is a program by GE called DSTar, which you have to donate 20,000 to to become a member. Do you know any other programs that are a little more reasonable? Thanks

Note: We have decided, because appendix B is optional we will just go with 310.16. Especially cuz we feel we are safe with limiting ourselves to 2 rows. We will however utilize the tables in 310 for our medium voltage duct banks. Thanks again.

[ December 27, 2004, 04:03 PM: Message edited by: bonding jumper ]

 

[bob]

Re: Annex B and Duct Bank De-ratings

BJ

We have decided, because appendix B is optional we will just go with 310.16. Especially cuz we feel we are safe with limiting ourselves to 2 rows.

Its optional because the NEC is not an engineering document. You have an engineering problem that should be addressed. 310.16 has been used in the past and problems did not occur because of the fat in the caculations of 220.
Apparently problems must have begun to show up prompting the NEC panels to insert Annex B.

Using you logic for the 2 rows, look at detail 3
which shows 2 x 3. Assume the conductor is 500 kcm. Table 316 has 500 cu at 75C rated at 380 amps. Annex B has it at 273 amps Ro = 90.
Thats 107 amps difference, about a 30% reduction.

[ December 27, 2004, 04:43 PM: Message edited by: bob ]

 

[charlie b]

Re: Annex B and Duct Bank De-ratings

Originally posted by bonding jumper: We have decided, because appendix B is optional we will just go with 310.16.

OK, since you put it that precise way, I feel I need to further explain and clarify my view of this topic.

First, the fact that Annex B is ?optional? only means that ?they? can?t make you use it. But this does not give us free reign to do anything we choose. We still have to follow the remainder of the code.

Secondly, take a look at the title of Table 310.16. It includes the phrase, ?Not more than three current-carrying conductors in raceway. . . .? But the installation you have described has up to fourteen times three, or 42 current-carrying conductors in close proximity. The basis of Table 310.16 does not include the possibility that any given set of three current-carrying conductors will be in an environment heated not only by their own internal I2R heat, but also by the I2R heat of 39 other nearby conductors.

Now take a look at the third sentence of B.310.15(B)(2). Suppose you start with a 2x2 array of conduits, keeping one conduit empty. Suppose you picked an ampacity value from Annex B Table B.310.7, using the center column for Detail 2. Finally, suppose that you have two such 2x2 arrays to install. What B.310.15(B)(2) tells us is that you have to keep the two ductbanks five feet apart, in order to make sure the heat from one does not influence the ampacity of the other.

Five feet apart! The installation you are describing has fourteen ducts, any two of which is no more than one foot from another. I cannot see how we could ignore the mutual heating. In my first post, I reported the decision by a group of Inspectors. I did not mention that I did not agree with that decision. But as an engineer, I cannot compel the owner to install more and larger conductors than the Inspector declares were sufficient.

Once again, I offer the caution that in order to use Table 310.16 for underground ducts, the load has to have been calculated in accordance with the very conservative methods of Article 220.

 

[charlie b]

Re: Annex B and Duct Bank De-ratings

Originally posted by bonding jumper: Do you know of a particular software package that would perform this operation?

I know of two, and have used both. One is by far more expensive than the other, but that is because it also performs all types of power systems analysis.

The one that I find easiest to use is called AmpCalc. When I last worked for a company that purchased this software, the cost was around $2,000. You can learn more at the CalcWare web site.

 

[bonding jumper]

Re: Annex B and Duct Bank De-ratings

Ok, I have another question, now that I got a little nervous, I started to perform the calculations and I am trying to understand Load Factor, I saw in a previous post by Charlie B

And just to complicate matters in one more small way, the person signing the calculation would have to justify the use of a 75% load factor. But that is no easy matter. First of all, the term ?Load Factor? is not defined anywhere in the NEC. We must then revert to standard industry definitions: ?Load Factor? is the ratio of average demand divided by peak demand. If you had a facility that was running at 80% power for 16 hours each day, and dropped back to 40% for the other 8 hours, the average would be 67%, the peak would be 80%, and the LF would be 67% / 80%, or 0.83. To get an LF of 75%, you would need a higher peak or a lower average. That sort of manipulation is the reason for ?engineering supervision.?

Now, how does this account for my situation? All our equipment, substations, are sized at 200% capacity. and tied together with tie breakers. Should one feeder to a substation be lost, it can be supported by a tie breaker and its associated substation. So in this situation, we would have our peak current set as emergency situation, or 3000A and average current at 1500A (assuming each substation is 1500A) Our equipment is 4000A and with 4000A feeders, 12sets of 600MCM. Since the most connected load possible is 3000 for 24 hrs, can I assume 75% load factor?

Edit: yes. I had to read Charlie's quote a few times before I completely understood. But my question states exactly what he said, so 75% * 24hrs / 24hrs = 75%. Back to work...

[ December 27, 2004, 05:54 PM: Message edited by: bonding jumper ]

 

[charlie b]

Re: Annex B and Duct Bank De-ratings

Originally posted by bonding jumper: . . . can I assume 75% load factor?

No, I?m afraid that you can?t.

You are talking about a ?peak? value that occurs under one operating condition (all load on one switchboard, with bus tie closed). You are talking about an ?average? value that occurs under a different operating condition (load split between two one switchboards, with bus tie open). This is an ?apples and oranges? situation.

You are also talking about connected load, not a prediction of the load that is likely to be running at any given time. Finally, you are attempting to bring the equipment nameplate rated values into the equation, where they do not belong. The concept of ?load factor? involves actual (or predicted) loading patterns, in terms of how much stuff is likely to be running at what time of day.

If you wish to take advantage of the higher ampacities associated with a 75% load factor (i.e., by using Figure B.310.1 for interpolation), then you have to have one of two things: (1) Actual measurements of load over a significant time, so that reasonable values can be obtained for ?peak? and ?average? load, or (2) A reasonably well-thought out and justifiable prediction of the loading patterns, so that a reasonable ?guess? can be made for ?peak? and ?average? load. If the facility is not yet built, then you only have option (2). But you would have to sell the ?guess? to the AHJ.

 

[nvcape]

Re: Annex B and Duct Bank De-ratings

You can purchase just the duct bank software from etap. You can move the larger loads around in the duct to get the least heating and compare various insulation types. Look at www.etap.com/ugs.htm

 

[charlie b]

Re: Annex B and Duct Bank De-ratings

That is the other software package to which I had alluded (i.e., the pricey one). I did not know you could just buy the ductbank portion.

 

[bonding jumper]

Re: Annex B and Duct Bank De-ratings

Ok this is where I am confused on load factor, we were in the office late last nite discussing this,

If you had a facility that was running at 80% power for 16 hours each day, and dropped back to 40% for the other 8 hours, the average would be 67%, the peak would be 80%, and the LF would be 67% / 80%, or 0.83

if Load Factor is

ratio of average demand divided by peak demand

Then how can we have 80% power for 16hrs a day? Shouldn't that 80% value be taken as 100% or Peak? What is the 80% out of? Is it 80% of total calculated load? 80% of the capacity of the transformer suppling the plant?

The problem I have with Load Factor is our load could be a simple clock, and that clock draws the same current all the time, but if I fed it with 500MCM conductors, that 500MCM conductor would have to be derated to 273 because the clock grants me a load factor of 1.

But if I have a clock and a 50HP pump that runs approx 1hr per day total, served from the same feeder as above, my load factor has decreased to less than .1, and I can actually get a rating of greater than 310.16's 380A, on the 500MCM. So basically annex B is stating that if I add load to my conductors, the KW added is less important than whether the load is constant or not.

Thanks again for your help guys, this has been a very good learning experience so far.

[ December 28, 2004, 10:05 AM: Message edited by: bonding jumper ]

 

[charlie b]

Re: Annex B and Duct Bank De-ratings

Originally posted by bonding jumper: Ok this is where I am confused on load factor. . . .

I appear to have confused you with my choice of terms. When I spoke of 80% load, I meant 80% of the total available load (i.e., the connected load). I?ll try again, using different terms.
</font>

• <font size="2" face="Verdana, Helvetica, sans-serif">Suppose that the total connected load on a facility is 100 amps. That is, if every single light and machine in the facility were turned on at the same time, you would get a reading of 100 amps at the main service.</font>

<font size="2" face="Verdana, Helvetica, sans-serif"></font>

• <font size="2" face="Verdana, Helvetica, sans-serif">Suppose that from 8:00 in the morning to midnight, the measured load (i.e., lights and machines that are actually running) is a constant 80 amps.</font>

<font size="2" face="Verdana, Helvetica, sans-serif"></font>

• <font size="2" face="Verdana, Helvetica, sans-serif">Suppose that from midnight to 8:00 in the morning, the measured load is a constant 40 amps.</font>

<font size="2" face="Verdana, Helvetica, sans-serif"></font>

• <font size="2" face="Verdana, Helvetica, sans-serif">Then the measured peak load for that facility is 80 amps.</font>

<font size="2" face="Verdana, Helvetica, sans-serif"></font>

• <font size="2" face="Verdana, Helvetica, sans-serif">The ?average load? for that facility is calculated by taking 80 amps times 16 hours, adding that to 40 amps times 8 hours, and dividing the sum by 24 hours. That gives us a calculated average load of 66.7 amps.</font>

<font size="2" face="Verdana, Helvetica, sans-serif"></font>

• <font size="2" face="Verdana, Helvetica, sans-serif">Finally, the ?load factor? for that facility is found by dividing the average by the peak. If you divide 66.7 amps by 80 amps, you get a ?load factor? 83.3 %.</font>

<font size="2" face="Verdana, Helvetica, sans-serif">

. . . that 500MCM conductor would have to be derated to 273 because the clock grants me a load factor of 1.

Having a load factor of 1 (i.e., 100%) does not cause us to derate anything. Rather, in some specific cases, being able to justify a load factor of less than one can allow us to ?uprate? the ampacity.

 

[guesseral]

Re: Annex B and Duct Bank De-ratings

Are you guy's talking about underground ductbanks containing conductors rated 2001V and above?

 

[bonding jumper]

Re: Annex B and Duct Bank De-ratings

Nope, this is 480V. But the same principals should apply to both.

[ December 28, 2004, 02:50 PM: Message edited by: bonding jumper ]

 

[bonding jumper]

Re: Annex B and Duct Bank De-ratings

Thanks for the help everybody, I found a company that will perform this calculation for you provided you give them the data required. It is an invaluable sevice they are supplying me, (not going to purchase $1500 software for one duct calculation. And the outcome really shocked me. http://www.midmetal.com.sg/ and then Click on "Need Current Rating?" in the center of the page.

In the above senario, they de-rated my 600MCM's to 274A from the typical 420A. I now have to use 750's instead. I really didn't think it was going to be such a large derating but for this project, I can't leave anything to chance and this would be a very large mistake. If anybody has the software, I would really appreciate if they could authenticate Midland Metal's number. 7x2 ductbank with 4" pvc, 7.5" center to center, and the 2 end ducts are empty spares, so 12 sets of 600 conductors total. The center of the ductbank is 31" below the surface and they used a Rho of 100. Thanks.

[ January 17, 2005, 03:41 PM: Message edited by: bonding jumper ]

 

[wirenut1980]

Re: Annex B and Duct Bank De-ratings

This may be nit-picky, but what if the spares are filled in the future? I do really hate to ask "what if" questions, I swear!

 

[charlie b]

Re: Annex B and Duct Bank De-ratings

It?s a good question.

I have seen two approaches to this situation. My preference is to include the possibility that the spares will someday be filled, in the calculation I perform today. That further reduces my present-day ampacity, but prevents future problems. The other approach is to put on the blinders, and say to yourself that if someone fills the spare conduits, it becomes that person?s responsibility to determine the impact of that future project on the cables I am installing today.

 

[charlie b]

Re: Annex B and Duct Bank De-ratings

Originally posted by wirenut1980: This may be nit-picky. . . .

Not many realize this, but a ?nit? is the larvae stage of certain insects, such as a flea. If you do not ?pick the nits,? then someday the fleas will come back to bite you.

 

[wirenut1980]

Re: Annex B and Duct Bank De-ratings

then someday the fleas will come back to bite you

I got enough of this to worry about with a dog and a cat!

 

[bonding jumper]

Re: Annex B and Duct Bank De-ratings

The spares are just that, spares. Who knows if one set burns up for some reason, and they can't pull them out they may have to use a spare but that wouldn't increase the total number of the duct bank. I don't care about what the spares are used as I just need a response on the reply 5 replys ago. Thanks.

[ January 18, 2005, 01:39 PM: Message edited by: bonding jumper ]