I'm doing a 200 amp 3phase 208 service upgrade for a new salon. No prints, no specs. Now I've always had the understanding that it is the EE's responsibility to spec out the AIC rating of the main, then my responsibility to be sure the equipment is to those specs and that it is series rated etc.

Well on this one there is no print, no engineer, so I asked the PoCo what the available fault current was and got the phone equivalent of a blank stare. Then he told me I'd have to figure it out myself. I told him (and it's true) that I have gotten this type of information from them in the past. I also explained that it is their equipment, and he would have to be involved anyway in at least getting me xfmer KVA and impedance etc. etc.

the GC works his day job as a project manager of a large commercial construction firm and is doing this build out on the side (all permitted of course). He has given me a slight attitude of me worrying too much about things that will ultimately hold up the tight schedule of the job, and I am getting little to no cooperation from anyone which will lead to delays. So the GC asked one of his big time design-build guys about the situation and they said they never worry about it unless something unusual catches their eye, and then they'll one of their engineer's run it through. But he says most situations like mine right now don't go much higher than 18,000 amps and a 22K rated main is fine.

So anyway, the whole thing is frustrating me a little. I am wondering what you guys do about this? I've had many people tell me that the first thing you are supposed to do when doing a service is get the available fault current before you start, but from the reaction of the PoCo guy, nobody's doing it. Any and all feedback would be much appreciated.

As a private practice EE I have run into the same difficulty. As a rule you are not going to be able to get the utility to tell you what the available fault current is on the primary side of the transformer for the usual commercial projects. They are too busy to take the time to give you an accurate number on an issue that has little bearing on their operation
In reality however the biggest determinant of fault current delivered to a point is the transformer kva and impedance. The run of cable also contributes to lessening of the fault current delivered to a point.
That leaves being able to determine what the KVA and impedance of the transformer that will be utilized. In the case where you may be utilizing an existing transformer
if you can determine the KVA, you can get an approximation of the impedance from the IEEE BUFF book. The issue you run into is that the actual values can bounce within a
range of values. Worst case is a minimum value impedance.
On new builds, there is a question of what size transformer the utility is going to use on the load. As you have probably noticed, the utilities do not size their units the way we do.
What I have done is call the utility and give them my NEC Code calculated demand and let them tell me based on their experience with that particular occupancy/building what size transformer that they are going to put in. Hopefully you will be able to get a more accurate impedance from them as well and not a guestimate.
Here is a quick and dirty method for determining fault current from a three phase
transformer.

xfmr fla x 100/%impedance of xfmr

Of course this neglects issues such as the impedance of your service conductors and possible motor contribution. But if your run of conductor is relatively short and your motors are small, it probably will
get you an idea of what rating gear you need to be buying.
At this time the AHJ's are sporadically enforcing this provision of the NEC in my part of the world anyway. I've had some tell me that they want to see it, others I believe don't even know what it is. As a rule I do it on all my risers, so I can size the gear right.

What range does the impedance tend to run in for utility-supplied transformers in the 50-100 kVA range?

I din't have any values for transformers that small. I'm sure that there are some Ansi standards that covers those oil filled units 75kva and below. For 112.5kva units the IEEE Buff Book Std. 242 shows a minimum impedance of 1.6% 2% Min. up to 6.2%. I would say that the 2% value is where I would base my calculation if I could not get a more accurate number.

In a pinch, consider using UL Class-T fuses in the service equipment. That will cover a multitude of sins.

Thanks for the info. After looking at the time current curves for the Class T I can see where that would be a good solution to limit fault current. The biggest problem is to recognize when you need to use them. A lot of electricians have an adversion to using fuses on the service entrance

Depending on where this business is, the utility may serve the service with a single 75kVA transformer or just about anything they want. They could also provide a networked transformer bank if it is a commercial area. In many parts of NYC, Con Ed will blanketly state that their is 200kA available.
Assuming a 75kVA xfmr, you would probably be safe with 22kAIC rated equipment. Service laterals will bring that value down and you might be able to get away with 10kAIC.

You cannot install current limiting fuses and assume that they will reduce fault current. Current limiting fuses will help reduce incident energy for arc flash protection, but not reliably reduce fault current due to dynamic impedance produced by downstream circuit breakers. http://ecmweb.com/mag/electric_dont_felled_higher/index.html

edited to add link

[This message has been edited by Ron (edited 07-30-2004).]

JPS1006:
A real good source for fault current calcs (point-to-point) is Bussman Bulletin EPR-1

It's titled "Electrical Plan Review" and is set-up in a workbook fashion. I do not know if it's available on the web, but their site is www.bussman.com.

I have attended a few seminars that Bussman engineers presented thru a State program, and this was a handout at the Short Circ Calc seminar.

Yes, Bussman sells fuses, and they do 'lean' toward fuses, but the info is presented in a VERY unbiased format. IF you can't get a copy from your local rep, I may have a spare copy around, if you're not in a hurry.

BJ: YES......a fused disconnect (when-in-doubt) and under almost all conditions, covers a multitude of sins!

John

Hey guys, thanks for the thoughts.

Hotline,
I have a book "Short Circuit Calculations 'The Easy Way'" Published by EC&M Books. It came with a 3.5" floppy. I've had it for some time and just haven't gotten a chance to open it. Thanks for the Bussman reference.

GamecockEE,
so in your example 112.5kva with 2% impedence,
[112500/(1.73*208)]=312 amps
312*100=31,200

31,200/2=15,600
or
31,200/.02=1,560,000?
I take it 15,600 right?

What do I do if there are 3 cans up on the pole, add the ratings of eack can together? The reason this caught my eye in the first place was that there is only 15' of service drop, so I knew that would play a little less of role in reducing the fault current.

Bjarney & Ron,
I did consider doing exactly that, going to a fused disconnect with class T fuses. I had a real good inspector a couple of years back require (or suggest I can't remember) that because of a 25K AFC on a service change. It was actually that job that really clued me in on this being a part of my job that I should pay closer attention to. Although since then I've learned about series ratings and now I'm not too sure how that factors in.

Besides, I don't even know yet what the AFC is, and I'm afraid that by the time I find out, they are going to be staring at me asking when they can open. I just hope the answer isn't "3 weeks to get the parts, and a day to put it in"

Alright, one last vent. I guess my delemma is this:
My gut has not been thouroghly trained to look up at transformer and say either, "nothing to worry about, looks close better run some numbers, or I know that's over 22K, better run some numbers to see how far over". Since there is no EE on this, I will have to run the numbers anyway just to cover myself, that is if I can get some cooperation. In the mean time, I have the GC and his big timer saying I'm wasting time even worrying about it.

Well I guess that's it, I'll just have to read this book, crunch the numbers myself, and hope for the sake of my own ego that it comes in over 22K, order the 65K main, and tell everybody see told you so and here's your extra for the engineering.

thanks for listening

JPS
15600 amps is correct. I take it that you are utilizing an existing transformer bank The utility might at least come and look at it and tell you what size it is (KVA) You then can get an impedance value from a standard or similar unit. If you have three single phase transformers connected to provide three phase service the impedance of the total bank is the same as a single unit. The KVAs are summed together i.e. three single phase 25 kva transformers = 75 kva. I would assume that the transformer impedances should be the same or close to the same or they would not share the load equally. Maybe some the readers on the utility side of things in this forum could shed some light on their practices when utilizing single phase transformers to get three phase services.

[This message has been edited by GamecockEE (edited 07-30-2004).]

Make every effort to get the numbers from the POCO. They do have them.

We got scr--ed bad once. We made calculations for a 600A 208Y/120 3PH system for an office building fed from a dedicated pad mount transformer. Not taking many chances we installed 42K equipment (225KVA, 625A, 1.5%Z). When the inspector called the POCO for their fault level they said 65K because, during an emergency, they might choose to install a larger transformer.

JBD, That sounds like a number that they pulled out of their rump based on the max withstand found on equipment suppliers catalogs. I really don't see how you could be held responsible for a change made by the utility at a later date. They seem to forget that the equipment goes up in price as the short circuit withstand increases.

thanks again guys.

JBD, I'm going to do that, (try to) get the numbers from the poco. Like I told them, I've gotten them from them before. I think he was just being lazy, or worse, purposefully trying to make things difficult. Oh well.

Just because one engineer did not know what the available fault current was does not mean that another doesn't know. In many cases, you could get an inexperienced engineer (or a lazy one) that doesn't know and doesn't want to ask someone else.

We will give the worst case like 77,000 amperes at the secondary bushings, assuming an infinite bus and a bolted fault. We will not give the available primary fault current because we have a dynamic system. The bank may be fed from one substation today and another one tomorrow.

We will size our transformer bank at around 50% of the actual service size. For instance a 1200 ampere service at 480 volts will have a transformer sized for about 500 kVA of demand.

The reason for the 200 kA in NYC is that several different transformers will feed into the fault at the same time since they use a networked system.

Utility guy from Indianapolis

CharlieE,
Does your utility have access to demand data from metered sites? For instance, if you have a new school of a square footage and certain types of equipment, that you could pull archived data from similar sites and size your transformer based on that information. I know of one in our area who looked at such data to make those decisions (when necessary). Their concern was the initial costs of supplying an oversized transformer and the continuing costs of supplying the magnetizing power to that unit.
I guess where I'm going with this is that if you can nail down the transformer to be used more precisely, it can allow for more
flexibility in what can be done on the customers side of the project.

[This message has been edited by GamecockEE (edited 07-31-2004).]

Yes, we have demand data from the larger, existing customers. However, we do not have that data available for smaller customers. Along with the demand data, we have the power factor data. I teach our engineers to do just that for their load calculations. This works well for chains of all kinds from restaurants to hardware stores, from department stores to sporting goods stores. We will share this data with electrical contractors after we sanitize it (we have to remove the customer's name, account number, address, etc. and then we have to provide the approximated size, type of usage, etc. so the EC will have useful data.

What is interesting about power factor is that we charge for poor power factor (over 85%) and we give a credit for improving power factor over 85%. Most customers do not want to be charged for the poor power factor but are not willing to correct past 85% to get the credit. I fail to see the difference, they are both cash.