It is currently Wed Jul 08, 2020 3:46 am



Post new topic Reply to topic
Author Message
ekstra   ara
 Post subject: Re-close: How does it effect the calculation
PostPosted: Fri Aug 15, 2014 9:41 am 

Joined: Thu Jun 05, 2014 5:52 pm
Posts: 19
Hi there,

I am calculating an Arc flash and struggle with a 1 time re-close. Will it double the thermal incident energy not by double the time but total heat is the two add up? In fact, both will come out the same number but double the time doesn't make muck sense since the waiting time to re-close is omitted. On the other hand, double the total energy doesn't sense since heat from the first Arc will decay before the second one active as well as the possibility of the second have the same Arc tip location, sustainable level, serve level is very low due the the concussion pressure of the first one.

Have any thought?
Thanks
==============================


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Fri Aug 15, 2014 4:10 pm 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 2174
Location: North Carolina
I think you aren't really looking at this the right way.

There is a calculation out there suggesting a minimum reclose time which generally indicates a minimum time between reclosing attempts of about 0.2 seconds or more. This may be different in an industrial application but most utilities use at least 2-3 seconds between reclosing attempts. The fastest vacuum interrupter reclosers out there call for a minimum open/close time of about 0.08-0.09 seconds which is a far cry from the old oil filled reclosers that would explode if you tried to reclose any fsater than about 8-10 minutes.

Thus most users of reclosers would not even fall within the "2 second" recommended timing from IEEE 1584. So the answer is that you would ignore subsequent reclosings. Otherwise you would be adding events until you get to the 2 second recommendation.

Second, subsequent reclosing attempts are usually very different from the first incident so simply "doubling" is not going to work. One approach ("fuse saver") is to have a very fast trip curve initially and one the second attempt, which is faster than the fuse curves, and then a longer curve on subsequent events to attempt to trip a fuse. The opposite is also true...a slow initial trip to avoid nuisance tripping due to transient disturbances (lightning), followed by much shorter almost instanteous tripping to minimize stress on the equipment while closing onto a fault.

Third, the subsequent "pulses" also trigger sectionalizers if those are available. And if they are, the recloser needs to wait for a sufficient amount of time to allow the sectionalizer to operate. These devices aren't very fast since they don't need to be so again a 1-2 second time delay between reclosing attempts would be considered very fast.

Fourth, I'm assuming here that we are talking about an outdoor, overhead environment. Reclosers don't generally make sense in indoor environments primarily because the concept relies on the idea that we have "self-healing" insulation (air) which is not generally the case in an indoor environment. Arcs on overhead lines are rapidly propelled in most cases away from the initiation point of the arc and away from the power source. So the assumption of a nice, stable arc similar to conditions in the IEEE test does not exist. It is unlikely that the arc will even stay stationary long enough to cause an arc flash injury where reclosers are in use.

Finally, remember that the whole concept of a recloser is to take advantage of the fact that at utility voltage levels once an arc initiates, it tends to continue to conduct through air which is heated to the point that it becomes a conductor. Absent thermal energy from an arc, the air rapidly returns to a nonconductive state. Since whatever initiated the arc (wildelife, vegetation, lightning, wind, etc.) is usually either destroyed or transient, over 90-95% of the time there is no chance of a restrike. This has nothing to do with any mysterious "concussive force". If the arc clears after the first event then there is simply no more arc flash in the first place (happens 90-95% of the time). If however the line is still clearly shorted in some way, it is far more likely that this second event is because of a bolted/dead short scenario such as physical damage where a line is in contact with a grounded object or another phase in which case this is a "bolted" fault and there is no arc at all. The only way to sustain an arc during a reclosing attempt is if the conditions are nearly the same as the first event. We'd have to have two conductors in close proximity that were not either magnetically or thermally seperated during the reclosing attempt. Frankly I'm a bit dubious as to the circumstances under which this could actually occur in practice. I'm not saying that it's impossible...just that I don't think that there's a high probability.

So my recommendations boil down to this:
1. Consider the likelihood that the reclosing attempt even matters. Consider:
A. Is it outside the "2 second rule", as it is with most utilities?
B. Given the fact that 90-95% of the time there is no second arc (recloser closes and stays closed) and that almost all of the less likely cases (5-10%) involve "bolted fault" conditions, a second arcing fault seems extremely unlikely. Given that ESFI gives the likelihood of an arc flash injury from OSHA data as about 0.1 incidents per year per 10,000 workers, that arc flash fatalities comprise less than 10% of all serious arc flash injury cases, and we're talking about some small fraction of less than 10% of all faults in the first place, we're rapidly approaching likelihoods of less 1 in a million workers per year, a number that is generally an acceptable lower cutoff with most safety-oriented agencies for fatalities. In other words, this would be an extremely rare case on par with the likelihood of being hit by a meteorite, something we don't normally protect against either.
2. If we pass "test #1", consider each event as a separate incident energy and then add them. This is necessary because the trip curves for each subsequent reclosing attempt are typically programmed differently. If we are waiting minimal time between reclosing attempts, the arc is known to form at a rate of nanoseconds to a few milliseconds depending on conditions so for all attempts and purposes, it is instantaneous and adding incident energies is a valid approach.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Mon Aug 18, 2014 4:26 pm 
Sparks Level
User avatar

Joined: Sun Jan 04, 2009 5:00 pm
Posts: 284
Location: Louisville, KY
I agree with Paul on most points. The likelyhood of a recloser event happening is pretty rare. Whatever caused the fault would be to be in place 2 seconds later. It should NOT be doubled as each hit is basically different.

We did a "recloser simulation" at the lab last month at 4 cal/cm2. The result with an 8 cal fabric was NO BURN. This will be in our paper at the 2015 IEEE-ESW in Louisville, KY Jan 26-29, 2015.

BUT it is also unlikely that the same place in a fabric will be hit three times because of the 2 second delay (if that is the recloser setting).

I differ with Paul in that I don't think the 2 second rule from IEEE 1584 will always apply since that is cutting off the arc at 2 seconds. There was no intention of using this with reclosers in the committee intent as I understood our discussions.

The clothing fabric we tested for the utility did quite well as long as it was 1/2 it's arc rating when exposed at it's arc rating, the second exposure could only take about 5-6 cal without breakopen or burn being recorded.

Hope this helps,

Hugh Hoagland
e-Hazard.com


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Tue Aug 19, 2014 11:01 am 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 2174
Location: North Carolina
The justification behind the "2 second rule" is rather tenuous. It refers to either the idea that the victim is physically propelled from the area or that the victim is capable of escaping (fleeing) on their own and this is sufficient time to allow human reactions to kick in. Whether or not the argument is tenuous is a valid consideration but then we go down the slippery slope of arguing what an appropriate cutoff would be. This argument has been made countless times and so far all that I'm hearing is that although everyone can agree that the "2 second rule" is more or less an assumption and not based on any sort of scientific evidence, no good alterantives have been recommended.

So I just applied the same logic. If the victim can't escape, then you would ignore the 2 second rule anyways. If escape is possible/likely, then apply it. The same, though tenuous, arguments supporting a 2 second rule for a continuous arcing condition would apply for a discontinuous (reclosing) case. In fact at low (<250 V) voltages, several measurements have shown the arc to actually stop and then reignite later so it happens even within the parameters given by IEEE 1584.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Tue Aug 19, 2014 8:37 pm 

Joined: Thu Jun 05, 2014 5:52 pm
Posts: 19
Thanks Paul,

I totally agree with you and that was in my initial report. However, it was rejected and I am instructed double the time resulting double the energy.
==================================


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Wed Aug 20, 2014 9:17 am 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 2174
Location: North Carolina
Just wait until next month/year. OSHA is requiring a risk assessment for generatoin/transmission/distribution equipment. 70E is changing to a new task table that breaks out the risk assessment portion, and the terminology is changing from requiring a "hazard analysis" to a "risk assessment".

Although this is sort of implicitly stated in 70E-2012 definition of an "arc flash" and vaguely referred to in various informational notes, it doesn't really come right out and say to do a risk assessment.

Once this occurs, there will be a new emphasis placed on what this new wording means and I suspect we'll see the light bulb come on with a lot of folks suddenly realizing that electrical equipment is not inherently dangerous and should not be treated as such. It is not even designed to support work on it as if it is inherently dangerous. Hence the reason that we are all running around worrying about PPE as a "solution".

There are efforts to move forward on making equipment safer to work on such as the new IEEE 1683 standard or solidly insulated switchgear or SF-6/vacuum switchgear that is plug welded shut but for the most part this equipment doesn't seem to be going mainstream very quickly. The idea especially with the latter switchgear designs that there are no user serviceable parts whatsoever (run to failure) seems to scare a lot of folks off even though these designs have been operating successfully in underground vaults, even while submerged 99% of the time, with little or no failures to speak of.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Thu Aug 21, 2014 8:34 pm 

Joined: Thu Jun 05, 2014 5:52 pm
Posts: 19
Please correct me if I am wrong; I thought that Jan. 1st 2015 is the date that Utility has to fully compliance with OSHA 269 including risk assessment.

Yes, I agree that pushing for PPE as a solution which in turn favoring "most conservative value" to protect not only the employee but employer again regulation fine and law suite. I respect that philosophy but more interested in Arc flash mitigating solution and providing a clear working environment where the risk is clearly defined and working boundary (protection) is marked.

Most manufactures are already produced Arc flash graded equipment but it comes with higher cost and bigger foot print or thicker material. I lean to engineer control solutions such as time delay, remote control. And yes, the new switchgear design focus on modular and solid dielectric which in turn requires less maintenance but not no maintenance unless one willing to pay for NASA/Military specification. Traditionally, Distribution and Residential units tend to run-to-failure not because their equipment do not require service but economic sense. Nowadays, regulation fine and law suite lead to proactive maintenance. Distribution Underground Network is a totally difference animal compare to the rest of the power system. The first underground network was built back in early 1920 but Automation was already a main function of the system. Most of the Network system was double contingencies in design which provides the most reliable circuit in power delivery system but since it so reliable problems was unseen by the publish as well as utility most of the time. In other word, failures occur frequently in underground system but unseen by most of people except the Network Operation. Recently, there are more underground events open to the publish such as Vault fire, Man-Hole fire, Man-hose explosion due to aging infrastructure and aging work-force. Solar power - Distributed Generator will be the most challenger to the Network system.

Arc flash in the Network system (480V) is the most dangerous due to high fault current, low voltage within a highly obsoleted technology electrical system. Whether an Arc flash would occur in a 216V secondary grid network is a grey area. EPRI seems to believe that Arc flash less likely occurs at that voltage level but the grid is provided by many 500KVA to 3500KVA transformers. In a spot Vault 480V, most the time the secondary collective bus was provide by 3 parallel transformers. Available fault current per transformer between 20kA to 35kA in general. Power can flow from either direction and the Network protector is only acting on revert power. Up stream protection device is the substation breaker which won't see the secondary fault until the fire reach into the transformer core and coil. Overload and over current protection is mainly depend on network protector's fuse and cable limiter. The only way to reduce fault current is switching out one transformer out of service while working on the second one and having the third one continue providing power to the customer. If a risk assessment is required then should all three operation modes (n, n-1, n-2) be studied? How many label will be need for each network protector?

By the way, EPA requires all utilities (both T&D) tracking and report SF6 gas annually. Also, there is no good way to test in service vacuum bottle. I believe that solid dielectric switches will take the main stream.
====================================


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Mon Aug 25, 2014 11:29 am 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 2174
Location: North Carolina
7604 wrote:
Please correct me if I am wrong; I thought that Jan. 1st 2015 is the date that Utility has to fully compliance with OSHA 269 including risk assessment.


There are 3 dates. The rule changes themselves are effective "immediately" (in 2014). The arc flash studies were due in April. Some other changes are due in January.

Quote:
Yes, I agree that pushing for PPE as a solution which in turn favoring "most conservative value" to protect not only the employee but employer again regulation fine and law suite. I respect that philosophy but more interested in Arc flash mitigating solution and providing a clear working environment where the risk is clearly defined and working boundary (protection) is marked.


So in other words first you try to engineer out the hazard, second to try to minimize exposure by not working energized if not needed, and third using PPE as a last resort. That's what the new 70E is actually stating. This is a big change from simply doing a study and wearing PPE and then only doing the engineering when the PPE is inconvenient, effectively inverting the steps.

Quote:
Most manufactures are already produced Arc flash graded equipment but it comes with higher cost and bigger foot print or thicker material.


This is only partly true. The GIS/VCB stuff is actually smaller since air as an insulator takes more space. Higher cost is not necessarily true either. Old oil/air style reclosers and switches run close to $35K+ for 35 kV class gear. Kyle and G&W are $25K for vacuum/solidly insulated, and Elastimold and Tavrida are under $20K. So the prices are not necessarily corresponding either. In terms of industrial metal-clad style switchgear though my experience has been that it is 10-50% higher if you buy "arc resistant".

I have a problem with arc resistant gear though. Switching is one of the safest functions. Maintenance activities inherently carry much higher risks. Arc resistant gear only helps protect against arc flash during switching operations. It does nothing for maintenance activities. So in my opinion, this is a waste of money since we are doing something about the lowest exposure.

Quote:
And yes, the new switchgear design focus on modular and solid dielectric which in turn requires less maintenance but not no maintenance unless one willing to pay for NASA/Military specification.


There are two versions of "no maintenance". First, we have repairs, regreasing, and cleaning. Second we have periodic functional testing to verify that the equipment is still working. Since it spends most of its operating life static, the only way to tell if it works is to test it. All of the former work is rendered impossible as a result of "zero mainteance" equipment. This stuff is definitely not exotic at all. A lot ofunderground gear is "maintenance free" design. It still has to be tested if there is any effort to ensure that it works, but if it is not working, it just has to be replaced as a single piece. There is no way for instance to replace a failed bearing or a bad linkage.


Quote:
Arc flash in the Network system (480V) is the most dangerous due to high fault current, low voltage within a highly obsoleted technology electrical system. Whether an Arc flash would occur in a 216V secondary grid network is a grey area. EPRI seems to believe that Arc flash less likely occurs at that voltage level but the grid is provided by many 500KVA to 3500KVA transformers.


There is no question. It has happened. Two electricians in Georgia in 2009 were killed and OSHA investigated it in a construction (240/120) panel. Lab tests (some are published on Mersen's site) have proven that it can happen. The question is more one of magnitude. In this regard, IEEE 1584 is not helpful and neither is Lee. EPRI is doing testing on the subject, which is the best approach baring large improvements in the theory.

Quote:
The only way to reduce fault current is switching out one transformer out of service while working on the second one and having the third one continue providing power to the customer. If a risk assessment is required then should all three operation modes (n, n-1, n-2) be studied? How many label will be need for each network protector?


This is an area we all struggle with. I have about 50 portable substations that move every 6 months in a mining system. They are between 500 kVA and 10,000 kVA, easily enough to kill someone. As they move, the arc flash ratings can go up and down. Do I do multiple stickers depending on distance or replace stickers on every move.

Quote:
By the way, EPA requires all utilities (both T&D) tracking and report SF6 gas annually. Also, there is no good way to test in service vacuum bottle. I believe that solid dielectric switches will take the main stream.


You can test the vacuum. It still has to come out of service, same as any other breaker at least needs the mechanism functionally tested (will it actually open/close). Don't hold your hopes up on solid dielectric though. We have two different kinds of insulation to deal with. The insulatioin that protects non-arcing parts can be anything (solid, air, oil, SF6, and theoretically vacuum). The insulation that interrupts a circuit in the contacts must be the type that is regenerative. So it must be air, oil, SF6, or vacuum. The only way to have a true solid state switch is with a solid state circuit breaker. These exist and have amazing switching times (microseconds) but have efficiency issues and are extremely expensive. Effectively this is the same as using a thyristor drive to produce power, using it as a "feed through". Since solid insulation cannot be used for switching contacts, we will have to continue to use some kind of fluid (air/oil/water/SF6) to insulate switching contacts.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Wed Aug 27, 2014 9:49 pm 

Joined: Thu Jun 05, 2014 5:52 pm
Posts: 19
Paul,

I believe you can buy an Arc resistant switchgear cost less than another non-Arc resistance one which has similar functions but not a like kind. It always costs more to obtain an Arc resistance compare to the standard design since it is an option. I also don't think that wasting money on an Arc resistant for something about lowest exposure. As a matter of fact, obtaining and Arc resistant or deploying other method is an benefit vs. cost equation. Most the time Arc resistant cost more for less obvious benefit but in some case no other choice is better than. It just like having a liquid transformer in side the build or dry type one. Arc resistant compare to other Arc flash mitigation solution just like passive safety control vs. active safety control solution. The latest safety technique is passive one.

No maintenance mean Run-To-Failure (RTF) just like we run our light bulb in house. Most of Distribution equipment RTF since the impact of an event used to cost less than nowadays. Another word, regulation driving the cost up lead to RTF becomes less effective to operate the system. Repair is not maintenance even it sometime driven by maintenance activity. regreasing and cleaning are maintenance mode but can be Preventive (PM) or On-deman. Periodic functional testing is also a maintenance mode (PM). As I mentioned in early note, most of underground gear is "blind to the public" but not maintenance free. An submersible Network Protector, for instant, still need to be period exercise (PM) to avoid mis-op. The case can be seal but after year in operation, dust still build up inside; therefore still require maintenance but less frequently.

Most of the old equipment is designed mechanically as an single unit which is hard or impossible to replace a component from within. However, modular is new design philosophy that allow small component can be replaced easily as it plug and play. The change has made a huge impact on O&M cost and Capital investment. It is also left account and tax code behind.

I'm sitting on Etap training this week and went through scenarios study wizard which allow compare all possible modes and provide worst case scenario for labeling. Somehow, I am not really think we always need worst case scenario for labeling since we don't want oversuite when maintenance mode is turned on which allow faster FCT or lower Ifb. Labeling is truly a challenge for any large and dynamic system.

Megger has a new vacuum bottle test device but requires out of service for testing which is impossible in some case. We has a hard time to keep up and report SF6 gas; therefore switching to Nitrogen if possible but solid dielectric is more favorite. I saw Elastimode MVI/MVS has picked momentum last couple year but still missing visual separation view. Vistagear with SF6 is still holding a large market portion.
=============================================


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Thu Aug 28, 2014 12:13 pm 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 2174
Location: North Carolina
7604 wrote:
Megger has a new vacuum bottle test device but requires out of service for testing which is impossible in some case. We has a hard time to keep up and report SF6 gas; therefore switching to Nitrogen if possible but solid dielectric is more favorite. I saw Elastimode MVI/MVS has picked momentum last couple year but still missing visual separation view. Vistagear with SF6 is still holding a large market portion.


Megger is reselling something that CBS Nuclear came up with. Its' basically a miniaturized version of what they use in producing vacuum bottles. It puts a magnetic field on the bottle and measures voltage on the terminals. Offline teest.

Elastimold has already overcome the visual break problem. They use the seperable connectors as a means of demonstrating visual break and even have a special modified version to show this. It makes sense. Vacuum interrupters can never be visually inspected for open/close, only air (or gas) break. The separable connector is the role of a "disconnect" (LBS or non-LBS) with MVI/MVR gear so that becomes your visual break mechanism.

Note that visual break is not specifically mandated by OSHA at least, or even 70E. 70E only mandates inspecting the visual break if the equipment has one. Molded case breakers for instance don't have a visual break that can be inspected. So there is no reason that it has to be present except that it has become virtually an unwritten requirement in the utility industry. Part of this is because a crew working in another utility's area during a storm event can still see whether or not the job is safe.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Wed Oct 01, 2014 6:44 am 
Sparks Level

Joined: Sat Feb 27, 2010 5:59 pm
Posts: 71
Location: Cincinnati, OH
7604 wrote:
I'm sitting on Etap training this week and went through scenarios study wizard which allow compare all possible modes and provide worst case scenario for labeling. Somehow, I am not really think we always need worst case scenario for labeling since we don't want oversuite when maintenance mode is turned on which allow faster FCT or lower Ifb. Labeling is truly a challenge for any large and dynamic system.


The industry consensus is that a single label should be applied which represents the worst case arc flash and shock hazard for a defined list of normal operating modes. Typically considered among "normal" cases are things like:
- maximum utility MVA
- minimum utility MVA
- all motors turned off (plant shutdown/maintenance situation)
- typical motors (not redundant spares) turned on
- main-tie-main equipment with mains closed and tie open
- main-tie-main equipment with one main open, one main closed, and tie closed
- other switching conditions that would persist for a long time

Situations that are often not considered "normal" for developing scenarios include:
- main-tie-main equipment with the mains and tie all closed due to a closed transition state that exists briefly as part of an automatic transfer scheme
- all motors energized in a plant that has fully redundant (2N) spared motors


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Wed Oct 01, 2014 11:40 am 

Joined: Thu Jun 05, 2014 5:52 pm
Posts: 19
jdsmith wrote:
7604 wrote:
- main-tie-main equipment with mains closed and tie open
- main-tie-main equipment with one main open, one main closed, and tie closed


Jdsmith,

With that two cases were consider "normal", how can having only one label?
========================================


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Wed Oct 01, 2014 3:51 pm 
Sparks Level

Joined: Sat Feb 27, 2010 5:59 pm
Posts: 71
Location: Cincinnati, OH
Research in the psychology field has shown that humans can actively engage our minds one one "controlled" task at a time, that is a task that requires conscious effort. Task such as having a conversation, reading, and navigating through traffic are controlled processes. We can also handle one additional "automatic" task, a task not requiring conscious attention, at the same time as one controlled task. Automatic tasks include eating, drinking, or listening to music.*

When a calculated study has been performed, selecting arc flash PPE typically involves reading the arc flash label on the equipment. Since most workers are familiar with the classes of PPE, they are often only looking for a few characters. In the past a worker could quickly look at the HRC number. Now the worker is looking at an incident energy number and they often just need to determine if the number is less than 1.2, from 1.2 - 12, or from 12-40. From a human factors/psychology standpoint, one could make an argument that determining PPE from an AF label is a controlled task since it requires reading. We could also contend that it requires minimal, very simple reading and could be an automatic task conducted successfully while the worker is focusing on a different controlled task. All of this is true if one arc flash label with one incident energy level is applied to the equipment.

If we have a situation where multiple arc flash labels are present on the equipment, or one label contains multiple incident energy levels for different situations, now consider whether selection of arc flash PPE is a controlled or automatic task. Now the worker is required to read some description of multiple scenarios (normal, RELT switch activated, equipment fed from alternate source, etc.), determine which state the equipment is in, then select the associated arc flash PPE as described on the labels. This is now a controlled process per the psychologists' definition.

Why does any of this matter? In the field there are many controlled tasks that face workers as they prepare to work on electrical equipment. The worker may be concentrating on a maintenance task, thinking through a problem they will troubleshoot after the equipment is isolated, having a conversation, or any other multitude of tasks. If we can change our labeling practices such that selecting arc flash PPE becomes more like an automatic task, we can increase the likelihood that workers will select the proper type of PPE. This is the reason that many safety professionals advocate the use of a single arc flash label with one incident energy level that represents the worst case incident energy.

This isn't to say that RELT (a settings maintenance switch) or use of different switching scenarios with lower incident energy levels can't be implemented. Rather, if employees want to put the equipment in some condition that will reduce incident energy levels and therefore apply lower levels of PPE, we should design the processes and procedures such that it forces selection of PPE to the forefront of the worker's mind - in other words PPE selection becomes the one and only controlled process that the worker is focusing on. Once the worker is focusing on PPE selection we can have a higher confidence level that they will correctly ascertain the status of the maintenance switch, source breakers, etc. and select the appropriate level of PPE rather than inadvertently selecting PPE with too little protection.


*Paraphrased from a paper by Anna Floyd published in the May/June 2012 issue of the IEEE IAS Magazine.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Wed Oct 01, 2014 3:58 pm 
Sparks Level

Joined: Sat Feb 27, 2010 5:59 pm
Posts: 71
Location: Cincinnati, OH
PaulEngr wrote:
There are efforts to move forward on making equipment safer to work on such as the new IEEE 1683 standard or solidly insulated switchgear or SF-6/vacuum switchgear that is plug welded shut but for the most part this equipment doesn't seem to be going mainstream very quickly. The idea especially with the latter switchgear designs that there are no user serviceable parts whatsoever (run to failure) seems to scare a lot of folks off even though these designs have been operating successfully in underground vaults, even while submerged 99% of the time, with little or no failures to speak of.


Have you reviewed the new IEEE 1683 standard? I would be interested to hear anyone's thoughts about IEEE 1683.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Mon Oct 06, 2014 8:24 pm 
Arc Level

Joined: Wed Jun 04, 2008 9:17 am
Posts: 428
Location: Spartanburg, South Carolina
Going back to the original subject of reclosing, I would recommend that work rules require that reclosers be put on one-shot when anyone is working on the line. If this is done, you don't have to consider reclosing.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Tue Oct 07, 2014 7:12 am 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 2174
Location: North Carolina
jdsmith wrote:
PaulEngr wrote:
There are efforts to move forward on making equipment safer to work on such as the new IEEE 1683 standard or solidly insulated switchgear or SF-6/vacuum switchgear that is plug welded shut but for the most part this equipment doesn't seem to be going mainstream very quickly. The idea especially with the latter switchgear designs that there are no user serviceable parts whatsoever (run to failure) seems to scare a lot of folks off even though these designs have been operating successfully in underground vaults, even while submerged 99% of the time, with little or no failures to speak of.


Have you reviewed the new IEEE 1683 standard? I would be interested to hear anyone's thoughts about IEEE 1683.


Yes. Some good ideas in there, some not so good. IEEE 1683 is not really a "standard" since it doesn't mandate any particular thing but more of a source book of ideas. For instance insulated bus can completely eliminate any hazards with respect to dropped tools or rodent nesting, but it does not necessarily imply eliminating arc propagation or arc faults from equipment failures. The idea of being able to externally perform measurement tasks is also a good idea in general and has been available for some time now in various "smart" MCC's but at some point once the measurements are collected repairs still have to be made and once again, we are opening equipment and setting up for exposure risks.

Perhaps one of the most significant and beneficial ideas is that of having a true safety disconnect switch as part of the MCC design. It is well known that draw out switchgear is significantly less safe and reliable than bolted switchgear and all the failure issues are related to the drawout mechanisms themselves, not the breakers which these days are similar if not identical in form and design. In fact at least one manufacturer (Siemens) uses the exact same breakers in both types of switchgear. Considering that disconnect switches have a long history of higher reliability than even breakers designed as switching devices, that indicates to me that the disconnect idea allows for working on MCC's safely, possibly even to the point of removing buckets in energized gear, more safely than in draw out switchgear. This is good news indded.

As an example in the switchgear realm, consider the Holec (Eaton) solidly insulated switchgear design. As far as I know you can't buy this in the U.S. but it is widely available in Europe. There are actually a couple others, this is just an example. In this case, what we have is a vacuum bottle as the primary interrupter and an earthing switch arrangement, all 100% encapsulated in epoxy. Connections are made with elbow connectors which provide 100% shielding right up to the equipment. Now considering failure modes of this type of equipment we have:
1. Rodents nesting or tools slipping. These are eliminated as we are 100% insulated and shielded. There is simply never a shock or arc flash exposure. Even if one of the blades in the switch somehow fails and remains energized, we can test for de-energized state using the capacitive taps on the elbows that are built into the equipment.
2. Bus spacing is so wide that arcs can't be sustained. Thus line-to-line arcing faults are eliminated at least within the gear. That leaves possible line-to-line faults with the cabling but this is also pretty much eliminated because it would imply shielding failures. Introduce impedance grounding and there's not enough energy in a line-to-ground fault to even initiate an arc since the ground fault current is less than the normal line current.
3. Catastrophic failures within the equipment are the only thing left such as failed interruption of a line-to-line fault which would imply that the breaker failed to operate, and whatever backup protection is present also failed to operate.

I think at some point we have to give consideration to the specific maintenance activities that would or should be allowed with the gear we use. Once those are decided, it should be a relatively simple effort to design for safety around each one of those tasks. For instance, should we be able to replace the overload relay? What about the contactor in a starter? What about contact tips and/or eutectic overload elements? Should we be able to pull the wiring and/or put in new wiring while the MCC is energized even if the bucket is not? Should we be able to change fuses? Fuse blocks? Circuit breakers? With each of these operations, how do we de-energize? Finally, how do we safely troubleshoot to detect failures in the first place? Decisions about each one of these may push for a less modular design where there are no user serviceable parts, or more modular.


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Wed Oct 08, 2014 6:48 pm 

Joined: Thu Jun 05, 2014 5:52 pm
Posts: 19
jghrist wrote:
Going back to the original subject of reclosing, I would recommend that work rules require that reclosers be put on one-shot when anyone is working on the line. If this is done, you don't have to consider reclosing.


Work rules requires that relosers must be disable before conducting a maintenance task but not switching as normal operation. Also, power system (utility) constantly changes daily. I didn't see recloser active on substation breaker setting (System Protection doc) but the fault data recording clearly shows the T-C-T. Human error contributes largely in most of the arc-flash events.
==============================


Top
 Profile Send private message  
Reply with quote  
 Post subject: Re: Re-close: How does it effect the calculation
PostPosted: Wed Oct 08, 2014 7:13 pm 

Joined: Thu Jun 05, 2014 5:52 pm
Posts: 19
Jdsmith,

We are currently replacing a number of LV MTM from both GE and Eaton. The problems are not only cost higher for an arc resistance switchgear but existing foot-print and location most the time didn't allow. The switchgear itself may not able to contain the flash/blast; therefore, incident energy must be draw out with exhausting channel which not only increasing physical size but also creating a big question of where to blow the arc within the building?

I like modular and draw-out breaker since most of the part can be replaced easily. Draw-out breaker is an dangerous task and most the time require highest Arc-suite protection. In some case, such and Network Protector in spot Network, there may be no suitable protection PPE. However, as soon as the breaker is racked out to the disconnect position, it is safe to perform necessary work under no or low voltage condition. Financially and spare part strategy, modular design provides huge benefit.
======================================


Top
 Profile Send private message  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 18 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:  
© 2019 Arcflash Forum / Brainfiller, Inc. | P.O. Box 12024 | Scottsdale, AZ 85267 USA | 800-874-8883