It is currently Fri Aug 07, 2020 11:06 pm



Post new topic Reply to topic
Author Message
ekstra   ara
 Post subject: Spot-Network Vaults
PostPosted: Wed Apr 20, 2016 8:16 am 

Joined: Tue Jul 17, 2012 12:11 pm
Posts: 2
I have been tasked with performing an arc-flash/short circuit study for a building in downtown Minneapolis. The building in question is fed from a spot-network utility vault with six 1500kva 13.8 to 480/277 transformers (5% impedance) connected to a common secondary bus. From the bus, five services fused at 3000A service the building. The utility fuses are Bussmann KRP-C. None of the service entrance switchgear have main circuit breakers.

Question 1(Short Circuit): Is it permissible to use the peak let-thru of the utility KRP-C fuses to determine if equipment in the facility is properly rated? The peak let-thru is 58kA per Bussmann documentation.

Question 2 (Arc Fault): What value should be used for utility fault contribution? I modeled the network vault using infinite bus at the primary of the utility and got a result of around 200kA. What is the lowest value that makes sense?

Question 3 (Arc Fault): When using a utility contribution of 100kA or greater, the highest incident energy calculated 14 cal/cm2 due to the utility fuse protecting the bus. Is it permissible to use the utility fuse in this manner for labeling the equipment?

Thank you,

Terry


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Spot-Network Vaults
PostPosted: Wed Apr 20, 2016 8:42 am 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 2174
Location: North Carolina
Terry Kalk wrote:
I have been tasked with performing an arc-flash/short circuit study for a building in downtown Minneapolis. The building in question is fed from a spot-network utility vault with six 1500kva 13.8 to 480/277 transformers (5% impedance) connected to a common secondary bus. From the bus, five services fused at 3000A service the building. The utility fuses are Bussmann KRP-C. None of the service entrance switchgear have main circuit breakers.

Question 1(Short Circuit): Is it permissible to use the peak let-thru of the utility KRP-C fuses to determine if equipment in the facility is properly rated? The peak let-thru is 58kA per Bussmann documentation.

Question 2 (Arc Fault): What value should be used for utility fault contribution? I modeled the network vault using infinite bus at the primary of the utility and got a result of around 200kA. What is the lowest value that makes sense?

Question 3 (Arc Fault): When using a utility contribution of 100kA or greater, the highest incident energy calculated 14 cal/cm2 due to the utility fuse protecting the bus. Is it permissible to use the utility fuse in this manner for labeling the equipment?

Thank you,

Terry


Many problems here.

Question 1: For SHORT CIRCUIT purposes, not even close for a couple reasons. First and foremost is because the maximum interrupting capacity of a fuse is always going to be vastly greater than the withstand of virtually any equipment out there. Simply put, there is almost no chance that the equipment will be built heavy enough to do what you propose and that's because it doesn't have to be. Your recommendation will be to re-engineer and re-design the whole system based on fault current which is incorrect. That is why you will find backing fuses in high interrupting capacity circuit breakers for the same purpose. You need to estimate the prospective short circuit current first including all impedances WITHOUT the fuse and then use the fuse let-through curve to determine the current limiting current which will drop it down to around 20-40 kA depending on prospective fault current.

Question 2: Of course it's wrong. As current increases the arc power increases BUT simultaneously the clearing time decreases and does so at a faster rate. Thus you will find that increased short circuit current produces a much lower incident energy. If you use these results, your results will end up underprotecting workers. Contact the utility and get the correct value. It is time consuming but it's just part of the job.

Question 3: See question 2. The "perfect" arc flash model would be to develop a model representing everything from the utility's generators to the prospective fault location. This is of course totally unrealistic but it is quite common to use a few circuit elements such as a transformer and/or overcurrent protection which the utility provides from "across the fence" just in order to establish a realistic value for the utility connection. The alternative is to model the utility as a "utility" (a voltage source and a fixed impedance) which doesn't take into account as many system dynamics.

It has been my experience that at 1500 kVA and 5% impedance you will find yourself close to if not beyond 40 cal/cm^2 on the 480/277 side prior to any available short circuit protection if you model the primary side in a realistic way. Fortunately Class L fuses are very quick and reduce the incident energy down to something reasonable, although quite often they need to be optimized a bit with a good downstream model to predict the maximum actual normal current so that you can use a fuse that is sized as closely as possible without causing nuisance trips. The associated disconnect (unless you do it from the primary side) and fuse enclosure will be near or over 40 cal/cm^2.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Spot-Network Vaults
PostPosted: Wed Apr 20, 2016 9:27 am 

Joined: Tue Jul 17, 2012 12:11 pm
Posts: 2
Thanks Paul.

The utility companies response is that buildings fed from this network shall be designed to withstand 100kA which this building is, fully rated at the service entrance and series rated throughout the distribution. In their response, they state that the fault current will be limited to less than 100kA via the current limiting fuses.

Using the 100kA as a starting point for the short circuit summary, I found that some of the equipment was still 'marginal' (within 1%) of the available fault even with series rating. This prompted me to question the validity of the 100kA value from the utility, hence the reference to the peak let-thru.

As for the arc fault questions, I understand the relationship between the available fault and clearing time which is why I want to lean towards using the lowest realistic utility fault. It is the local utilities policy not to provide the minimum available fault current or associated protective device clearing times, so that avenue is closed.

The goal from the customer's perspective is arc fault mitigation so they can safely perform IR testing at the service entrance gear (they do not currently have IR windows). I am certainly not opposed to labeling the SE gear Dangerous, which will likely be the case since there are so many unknowns on the primary side of the vault. I am only asking, what is the lowest realistic available fault coming from 6 parallel transformers on a common bus?

Thanks,

Terry


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Spot-Network Vaults
PostPosted: Wed Apr 20, 2016 1:58 pm 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 2174
Location: North Carolina
Terry Kalk wrote:
The utility companies response is that buildings fed from this network shall be designed to withstand 100kA which this building is, fully rated at the service entrance and series rated throughout the distribution. In their response, they state that the fault current will be limited to less than 100kA via the current limiting fuses.

Using the 100kA as a starting point for the short circuit summary, I found that some of the equipment was still 'marginal' (within 1%) of the available fault even with series rating. This prompted me to question the validity of the 100kA value from the utility, hence the reference to the peak let-thru.


The utility is giving you the standard utility run around. Typically you have to get access to the relaying or engineering department to get the information and those departments are typically very helpful almost to a fault. But there is another approach starting this year. What you are talking about is distribution equipment, NOT utilization equipment. So 70E isn't even in the picture...it's OSHA 1910.269. The utility is also required to do incident energy analysis of their equipment for their own employees. See 30 CFR 1910.269(l)(8)(ii). Politely ask for THEIR incident energy analysis and just copy/paste it into the report for the customer for those parts of the system. Alternatively if they refuse to produce this then give them a politely worded letter that explains that you cannot determine the incident energy based on the information that they have produced since unrealistic values of current produce unrealistically LOW incident energy values. Request the information again or give them the alternative that you will file a formal complaint with OSHA and/or the state labor board requesting that the utility be investigated (and cited) for violation of OSHA 1910.269(l)(8)(ii) by refusing to produce the requested information for use for protection of your customer's employees. This puts them in the unenviable position of also having to produce documentation to OSHA that they have themselves complied while the OSHA inspector has a PE in their back pocket helping them.

Quote:
The goal from the customer's perspective is arc fault mitigation so they can safely perform IR testing at the service entrance gear (they do not currently have IR windows). I am certainly not opposed to labeling the SE gear Dangerous, which will likely be the case since there are so many unknowns on the primary side of the vault. I am only asking, what is the lowest realistic available fault coming from 6 parallel transformers on a common bus?


Two things here. First off despite the fact that SKM and other software produces default settings that use the signal word "DANGER" for anything over 40 cal/cm^2, this is the incorrect signal word for two reasons. First the number of fatalities due to an arcing fault when an arc flash injury occurs is less than 10%. Second the likelihood of an arcing fault while just walking by or even opening panels is pretty low. The signal word "DANGER" is reserved by ANSI 535 for cases where a life threatening injury is imminent such as removing the guard from a grinder or a saw and sticking your hand into it. The correct signal word is WARNING and use of the word DANGER is an OSHA violation since OSHA mandates use of ANSI 535 for safety signs.

The second problem here is that at the time that the "40 cal/cm^2" standard was set up, the best available PPE was 40 cal/cm^2. Now it goes up above 100 cal/cm^2. Thus 70E stopped at 40 cal/cm^2 at one time. Since that time all kinds of goofy rumors and claims have circulated about the magic 40 cal/cm^2 "barrier". The most recent draft of 70E-2018 has removed it and it is unlikely to be put back in.

The reason that I'm saying this though is that it is simply not true that you can't work on equipment in excess of whatever the PPE that is available. It's just that the types of tasks that can be performed safely is restricted somewhat. As a case in point it would be downright foolish to attempt to rack in a draw out breaker while energized without the necessary PPE because plenty of incidents have happened doing this task. Doing a noncontact inspection (IR scan) is also obviously risk free by itself. The only problem here is opening and closing the doors. There are a number of cases where opening a door can be hazardous:
1. The door is bolted on and not round so it or the fasteners can fall into the bus work while being removed if there is exposed wiring on the other side of the door.
2. There is electrically conductive dust built up on top of the enclosure above the door or is otherwise floating around in the room such as in a machine shop in a cutting area.
3. A whole bunch of tools and parts (fuses) have been stacked up on top of the enclosure that can roll into it when the door is opened.
4. There is a lot of wet mud and/or liquids all over the outside of the enclosure that can run into it when the door is removed.
5. The door is rotted out or the hinges are broken and could fall into exposed parts when the panel is opened.

Notice two things. The first is that this is NOT the normal situation for most electrical equipment. Second notice that all of these potential issues can be discovered externally before the door is opened. Thus an inspection with a written and detailed procedure ahead of time can uncover any of these kinds of hazards. This is particularly true if you have windows provided for inspection of the blades on disconnects. Thus the default tables in 70E may declare this particular task as a risk in the general case (which includes cases 1-5 above) while in the specific case of your equipment the risk may not exist and for most cases does not exist.


Top
 Profile Send private message  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 4 posts ] 

All times are UTC - 7 hours


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
cron
© 2019 Arcflash Forum / Brainfiller, Inc. | P.O. Box 12024 | Scottsdale, AZ 85267 USA | 800-874-8883