Fault current reduction for lighting control

Status
Not open for further replies.
hmspe

hmspe

Senior Member
Location
Temple, TX
Occupation
PE
One of the local jurisdictions is very dilligent about having branch circuit lengths long enough so that the available fault current at lighting contactors, time clocks, receptacles, etc., is appropriate for the SCCR of the device. I'm seeing three approaches to the calcs.

A hypothetical system would have RK fuses in the service and 20A branch circuit breakers. Let's assume that the calculated fault current at the service is 65,000A and at the panel is 40,000A, on a 65K/10K series rated system. The first calculation approach uses the 40,000A at the panel as the basis for calculating the branch circuit conductor length required to reduce the fault current. The second calculation approach uses the "let through" of the the RK fuse (around 5000A) as the basis of calculation. The third approach adds a 20A fuse (usually class RK) just ahead of the control device and uses the let-through of that fuse to protect the downstream device, ignoring the fuse(s) and breaker(s) upstream.

I'd appreciate any thoughts on which method(s) are acceptable, and why.

Martin
 
Fault

Fault

the most common method used around here is calculating the fault at the end of the feeder and using extended branch circuit wiring to the conatctor. a 15 or 20 ft run of #12 or #10 wil certainly knock most available faults to within contactor acceptance. I see no prolem using a fuse. I don't have my tables with me, but I was suprised that you have a RK fuse of any size that gave you the 5k downstream. The 20 amp I would understand.
Here most folks prefer the longer wire than introducing an addded fuse.
 
augie47 said:
the most common method ...
Click to expand...

Thanks for the reply. I wanted to hear from others before posting my opinions.

I generally use the "long branch circuit" approach, but on a recent job the 3 phase side required 72' of #12 to drop from 38,150A to under 1000A. This project had both time clocks and contactors, both selected by the contractor. I couldn't find SCCR ratings for the time clocks, and the available SCCR ratings for the contactors depended on the overcurrent device and none matched what was in the field. The contractor wanted to know why I didn't use either the main fuse let-through or a supplimental fuse, and he sent drawings from other engineers showing those approaches. Not a pleasant situation, partly because I don't like having to critique another engineer's work, and partly because I don't have a 30 second explaination for why the method I used is the best method.

My thoughts on the options:

For #1: Manufacturers should publish SCCR ratings for all devices, and the default SCCR rating for devices should be higher than 1000A. Time clocks and lighting contactors typically are mounted next to panelboards. It makes no sense to have to run 72' of conductor to a device 1' away.

For #2: For a typical fuse/breaker/contactor chain, I'm still trying to figure out how the let-through of a fuse can limit the fault at the contactor but, because of "dynamic impedance", the fuse can't limit the fault at the breaker. I'm more than a bit sceptical of the whole "dynamic impedance" thing -- it looks more like a marketing ploy or a laboratory oddity than a real-world condition to me.

For #3: There's still the "dynamic impedance" problem, because there's still a breaker in the chain. If we can't use let-through on a main fuse/breaker combination I don't see how we can use let-through on a breaker/fuse combination.

I may be missing something in my analysis -- it wouldn't be the first time. I've read through several articles on applying low SCCR rated devices, but all of the articles seem to be from breaker manufacturers and are suspect. If anyone has a link to an impartial article, or has any other comments, please post.

Martin
 
Martin,

Do you have a 2005 NECH ? , in 110.10 in the explanatory area I'll type you

a couple of comments from it.

' When the available short circuit current exceeds the SCCR of an electrical

component, the OPD must limit the let-through energy to within the rating of

that electrical component.

Adequate protection can be provided by fuses, molded-case circuit breakers,

and low voltage power circuit breakers, depending on specific circuit and

installation requirements.

Martin, no where in the comments did they mention extending the circuit

conductors by " x" amount of feet to comply. I can see where EC have

problems with this.

It would be much less of a problem if the EE or the plans reveiwer had this

' Detail ' on the working drawings that are used on the job, even if the EC

did not understand 110.10 He/She would install per the prints and the results

could be controlled the first time.

One last thing, if you get a good link on this subject, I would really aprech-

ate a P.M. as I'm still trying to grasp the total concept.

Thanks,
 
I have a theory question here:
For devices such as time clocks and contactors that are used to control loads, why does the dynamic impedance issue associated with a current limiting fuse matter? In the event of a short circuit, these devices will likely simply stay closed and not even try to open.

I understand how a 'current limiting' fuse might cause problems with a downstream breaker, because the downstream breaker might try to open in the fraction of a cycle while the fuse is limiting the current. But I thought that the whole point of current limiting fuses is that they limit the I^2T that the downstream devices see, so doesn't this effectively reduce the impact on devices that stay closed?

But I admit to not really understanding how short circuit ratings are determined and utilized.

-Jon
 
Jon,

It's like having a car tire rated for 100 MPH on a car capable of 200 MPH,

but, the car has a throtlle control to keep it at 100MPH.
 
benaround said:
...

One last thing, if you get a good link on this subject, I would really aprech-
ate a P.M. as I'm still trying to grasp the total concept.

...
Click to expand...

That's pretty much where I am. I haven't found a reference with a clear technical explaination. I've found magazine articles that make sweeping generalizations, and I've found articles from manufacturers that seem to be more marketing than engineering in nature.

As far adding conductor length goes, it seems to be the accepted approch for Phoenix, and it's a sound approach from an engineering standpoint. In my opinion the NECH comments you quoted aren't worth much -- after reading them you really don't more than you did before you read them.

Martin
 
winnie said:
For devices such as time clocks and contactors that are used to control loads, why does the dynamic impedance issue associated with a current limiting fuse matter?
Click to expand...

Good question. I haven't found anything from an accepted authorative source that really explains this. I have to be able to defend my design decisions, and at this point the only approach I can absolutely defend is to treat the system as fully rated, with the fault current at any point limited only by circuit impedances. While that's a safe approach, it's probably overdesign.

Martin
 
First I have to admit partial ignorance to the dynamic impedance issue. Maybe because I have not run into it. Then again I don't usually select the devices and then figure the fault current. Normally I will calc fault current at each panel and then elect the rating for that panel in excess of the available. In your situation, If you have a lighting contactor right next to a main switchboard, why can't you just put a fusable disconnect ahead of the contactor?

Also, jusr reread the IEEE gray book section on short circuit analysis. no mention of dynamic impedance. They pretty much recommend RK1 fuses or longer wire length in your case.
 
The first calculation approach uses the 40,000A at the panel as the basis for calculating the branch circuit conductor length required to reduce the fault current. The second calculation approach uses the "let through" of the the RK fuse (around 5000A) as the basis of calculation. The third approach adds a 20A fuse (usually class RK) just ahead of the control device and uses the let-through of that fuse to protect the downstream device, ignoring the fuse(s) and breaker(s) upstream.
Click to expand...
I would recommend 2 different Classes of fuses. Class J at 20A will let through 1,000 amps RMS or Class CC, with the same. There are Manufacturers with Disconnects to accomidate these fuses.
http://www.cooperbussmann.com/pdf/1062.pdf
http://www.cooperbussmann.com/pdf/1023.pdf
Just my $.02
 
sceepe and david43229,

Thanks for the replies. In regards to sceepe's first comment, I don't work backwards either. I see this when an installation is red tagged and the contractor calls asking what to do to fix the problem.

I appreciate the suggestions on adding fuses; however, what's the documented basis for adding fuses and using let-through? Square D bulletin 1210HO9901, titled "Shedding New Light on Branch Circuit Applications" (see Figure 4) illustrates adding an RK5 fuse before a contactor to raise the SCCR of the contactor, but states that 25' of branch circuit conductor is still required between the branch circuit breaker and the contactor. The system shown is 480V with a 14KAIC panelboard. I don't know if this is a case of a very early publication (1999) that came out before the issue was fully understood, or if it's mostly marketing spin (likely), or if it's entirely accurate. It's still available through the Square D web site. Since it's published by a major manufacturer most people would consider it authoratative.

Most people (including me) have a partial knowledge of the dynamic impedance issue. It didn't exist when I was in college, and the documentation I've seen isn't complete and balanced. I'm still hoping that someone can point us to a link that provides a balanced explaination of SCCR issues and acceptable methods to deal with them.

Martin
 
Dynamic impedance is not a part of short circuit studies. Short circuit calculations are performed using bolted fault currents not arcing currents.

Dynamic impedance is a function of arcing faults. When circuit breaker contacts open they create an arc. This arc adds impedance into the circuit. As the contacts continue to open the arc gets longer adding even more impedance until eventually the arc is extinguished. Arcing currents are normally relatively low compared to the equivalent bolted fault current. The low level of arcing currents are one reason why 480Y/277 circuits may require GF protection and arc-fault breakers where created.

While it is not normally discussed, even fuses should exhibit dynamic impedance due to the internal arcing as the fuse elements melt.
 
jim dungar said:
While it is not normally discussed, even fuses should exhibit dynamic impedance due to the internal arcing as the fuse elements melt.
Click to expand...

I guess that _any_ switch has to exhibit dynamic impedance; because it has to start at close to 0 impedance and somehow get to near infinite impedance...there has to be a transition between the two.

My previous exposure to the term dynamic impedance was in regard to the use of 'current limiting fuses' and SCCR ratings of downstream devices. My (admittedly weak) understanding was that even with a current limiting fuse, since the downstream device would 'see' the initial short circuit current, it could in theory start to operate in response to overcurrent, before the current limiting fuse would limit the current.

It seems that a significant part of Martin's problem is that the contactors probably have not been rated in terms of the short circuit current that they can withstand while closed for the extremely short duration permitted by a current limiting fuse.

Another issue is to ask just what safety goal is achieved by making sure that the SCCR rating for the time clocks and lighting contactors are not exceeded. If the OCPD is suitably rated for the maximum available short circuit current, and the contactor is not being used to _interrupt_ this short circuit current, then a short circuit might damage the contactor without a significant safety risk. Just what is the SCCR rating of the #12 conductor feeding the contactor?

-Jon
 
contactors

contactors

as a side note, kudos to Mike's Forum, agian, in that this is evening being discussed.Hopefully the subject will enlighten some folks. I can recall in my earlier years having my lighting contactors fed 1 ft from a 22 k (or higher) panelboard and no question ever raised. Luckily, as far as I know, the covers never flew off (yet)
 
safe enough ?

safe enough ?

Here is an example of conditions in the Phoenix area, this event took place

in the last two weeks.

This job is a good sized dist. center, GC called in for a final electrical inspec-

tion. Cut to the chase, SES is outdoors goes 550' to an indoor dist, out of

dist. 80' to a 225a 480/277vac panel. Next to the panel is an exhaust fan

control panel ( 30,000a SCCR ) with a fused disconnect in it and 45a current

limiting fuses. EI looks at aic on the c.b. in the panel and it is higher than

the 30,000 in the EFCP, ok, what do we need to do?

WE need to change the conduit to the EFCP to two inch and run it 20'

before hitting the exhaust fan control panel and pull 2/0cu. to feed it. So,

do you think that this is safe enough.
 
augie47 said:
as a side note, kudos to Mike's Forum ...
Click to expand...

I think it's great that this is being discussed, but at the same time I'm a bit disappointed. There have been a number of opinions, but based on what has been posted, there must not be much that's published. There doesn't seem to be a consensus on how to deal with SCCR issues. At this point I still don't have a published reference that I could show an inspector that would support any approach other than calculations like for a fully rated system.

Martin
 
contactors

contactors

I am often presented with the calculations of the AIC based on wire length, etc. from the source to the panel and then to the contactor. I am also presented with fuse "let-thru" charts....I may learn here that I am mistaken in doing so, but I accept either.
 
augie47 said:
I am often presented with the calculations of the AIC based on wire length, etc. from the source to the panel and then to the contactor. I am also presented with fuse "let-thru" charts....I may learn here that I am mistaken in doing so, but I accept either.
Click to expand...

Did you look at the Square D brochure I referenced in post 11? My reading of the brochure is that "let-through" is not acceptable, at least with Square D equipment. Keep in mind that this is an early document, it's obviously from the marketing department, and Square D makes breakers, not fuses. On the other hand, what's there (assuming it hasn't been superceeded) would probably apply to Square D equipment in any case if it's part of the manufacturer's installation instructions.

Martin
 
Status
Not open for further replies.
Top