PPE is tested (actually the cloth, not the PPE) for a particular incident energy level and over time using the Stoll curve. Since the working distance determines the incident energy, this determines the PPE required. One of the easiest ways to get an arc flash energy reduction is to set up the job to reposition the worker farther away from the equipment. Obviously the PPE requirement drops when you do this.



The hazard assessment (calculated or tabular method) gives the incident energy and a working distance. IEEE 1584 uses the distance between the face/chest area (where a fatality due to burns is most likely) and to the buswork at the back of the enclosure as the working distance which is surprisingly relatively uniform from one manufacturer to the next. Thus the labelling is typically a worst case analysis. This distance works fairly well for "typical" maintenance conditions but would be overkill for instance while racking a breaker, operating the handles, or using even a short hot stick to insert or remove medium voltage switches in a panel. If you really want/need to analyze the work on a task-by-task basis then you would have to look at adjusting the working distance. For instance at my facility we have a huge amount of substations with group operated outdoor load break switches mounted a minimum of 10 feet from the worker. So we recalculate the incident energy at this working distance which in most cases reduces the incident energy to <1.2 cal/cm^2.

Quote:

Quote:

Quote:

The written opinion will be to talk to IEEE. Except for the tabular method (and even then, it is pretty weak), NFPA doesn't really pass judgement on any particular arc flash hazard calculation methodology. They list 8 methods in the Annex and other than listing their respective ranges, they don't pass judgement on any of them. So you can freely implement any of them. However you also have to apply engineering judgement when doing this. For this reason the vast majority of practitioners are using IEEE 1584 for most cases, and then Ralph Lee or NESC only for cases outside of the valid range for IEEE 1584. NESC is actually using ArcPro for the calculated values in their tables in the medium voltage cases.

Quote:

Quote:

Quote:

The 70E Technical Committee has in numerous ROP's openly stated several times that "just walking by" does not carry any appreciable arc flash hazard. This is also noted in the definition of an arc flash hazard as well as elsewhere. The key is that interacting with the equipment in such a manner that an arc flash hazard is likely is where any of this matters. That requires a risk assessment, not simply a hazard assessment. The tables in 70E are actually a risk assessment which is the reason that they don't have a single value but include a range of tasks. When you do your own risk assessment using a calculation, then you need to follow suit and provide a range of values. However, this is where I diverge from the approach in the tables in 70E. Basically it comes down to this. Unless the task specifically does something to change the incident energy such as by increasing working distance, the task itself does NOT alter the arc flash hazard (incident energy) in any way. What may change is that depending on the task and the condition of the equipment, the LIKELIHOOD changes. Thus completing an arc flash risk assessment allows us to determine which tasks do not require arc flash PPE and which ones do require it. Risk assessments are required by OSHA regulation for all tasks. Performing the risk assessment will clear up the issues that you are having.

Quote:

Quote:

Quote:

This used to be a lot more clear in the 2009 edition. Section 130.5, Exception is what you are looking for. You either do the calculation or you do the table (following the exception).

Quote:

Quote:

Quote:

It is required by OSHA to do a risk assessment. If your engineering and safety group can't do their jobs, then find them new ones in another organization. There should be no argument about this.



Why stop at "Cat 4"? There is now 100 cal/cm^2 and higher incident energy rated clothing on the market. Back when 70E first took the "40 cal/cm^2" stance, there was almost nothing on the market rated that high.

Annex F is crap. It is in the annexes (which are 100% nonbinding) for a reason. I have actually attempted to use it. I strongly advise against using it. If you don't believe me, try to do one for any simple task. There are just too many undefined things in that methodology, and to put it mildly, it is utterly confusing. Depending on the edition, you have 4 risk factors to determine but it gives you no information on what to do with them. And there is no risk/likelihood matrix either unlike all other standards so you can't ever come up with a "result". The origin appears to be a very rough attempt to adapt the most common ANSI risk assessment method (TR1) but it is so poorly written that you can't possibly implement it. Use an ANSI or ISO approved risk assessment method. It's pretty simple with arc flash because without PPE you look at the equipment arcing fault rates which are already given in IEEE Gold book, or by human performance which is generally considered around 10% failure rate depending on the task. For example if you go through IEEE Gold book, opening/closing disconnect switches is better than 10^-6/year probability of an arcing fault. Then you move on to the hazard side of things. If you are exposed to >1.2 cal/cm^2, you get a fatality. Depending on the risk assessment method you may have to count one or more than one victim. Normally doing a risk assessment, we get to analyze a wide variety of potential injuries. With arc flash we just have a binary decision (live or dead). If wearing adequate PPE, there is a 90% chance of no injury (as per IEEE 1584 results). This is more than enough information to apply to an ANSI or ISO risk assessment methodology. Following on with the example, any of the risk assessment procedures will give you an "tolerable risk" result if the failure rate is <10^-6. However if it is something like attempting to land a wire on a live terminal with an uninsulated screwdriver, a task that is clearly limited by human performance alone, then the likelihood of failure is around 10%/year. If the hazard (calculated from IEEE 1584) is <1.2 cal/cm^2, then again, this is a tolerable risk since at least in terms of arc flash (not shock), nobody dies. BUT if it is over 1.2 cal/cm^2, all you can hope to do is have a 90% or better chance of surviving (IEEE 1584) by wearing the PPE. This puts you at 10% x 10% = 1% chance of a fatality. Most risk assessment methodologies will conclude that this is still not a tolerable level of risk so we're done and the PPE doesn't even matter. But if you are say drawing out a drawout breaker where the failure rate is around 10^-5, then adding the PPE decreases the risk by a factor of 10, to 10^-6, which gets you across the finish line and makes this a task for which there is a tolerable level of risk, provided that you can spec out sufficient PPE. If not, then we are at a no-go.

Now if we walk down this path, you will find that again, OPERATION of the equipment is at a tolerable level of risk (10^-6 or better) for almost everything except for draw out switch gear. This is why in the arc flash definition, 70E states that properly installed and maintained equipment is unlikely to pose an arc flash hazard. They repeat this kind of statement in the note in 130.7, Note 2 where they state that the same 480 V equipment you are referring to is not likely to pose an arc flash hazard during normal operation. There are also dozens of recent "Public Inputs" (the new name for ROP's) recently submitted for the 2015 edition asking the 70E Committee to finally take a stand on this issue.

This is with doors closed. Once the doors are open, then we get into 130.4 and we have to start considering the possibility that a worker can accidentally slip with a tool and cause an arc flash without insulated tools. Similarly, if the equipment just tripped due to a fault, then obviously the assumption of "properly installed and maintained" is now compromised and that we must treat the equipment differently. Operations needs to get out of the way and let qualified personnel perform maintenance to get it back into good operating condition again. At this point since the state of the equipment is unknown, the arc flash suits come out unless we de-energize upstream in some fashion.

You can provide adequate guidance to operations about how to do this. If they don't see smoke, obvious physical damage, obvious environmental damage, and it didn't trip (fault), and maintenance is following an adequate PM program as per 70E, Article 205 (especially 205.4), Article 210, 215, 220, 225, and 230, then the assumption about equipment condition is valid. Otherwise they are not allowed to reset things as per not only 70E but also OSHA (1910.333) until the fault is discovered and fixed. Resetting a breaker without checking it is NOT acceptable. In fact this is also repeated in NEMA AB-4 which is freely available and is the maintenance standard for molded case breakers. There are similar requirements for other types but generally only molded/insulated case breakers require an inspection after EVERY trip. Fortunately the inspection is very quick and simple (basic visual inspection only). The same requirement is also needed for all starters except for those installed under IEC Type 2 (no-damage) requirements, so if you are actually following the standards, you are doing this anyways.



You can add arc resistant gear to that list. Two slightly different issues. First, frankly, I'm in the electrical maintenance department. All of this kind of technology does NOTHING for the person who has to work on the equipment for the most part. Making it "blast resistant" with doors closed does nothing for someone who has to open the door. It's a huge waste of money in my eyes because if we can't protect the worker that has to work on the equipment, then we've failed to protect the person that is most susceptible to permanent injury or fatality. Not that I don't like operators but the likelihood of an injury for them is much lower than for electrical workers. This is clear from the injury stats where utility, maintenance, and construction workers account for the majority of the electrical related fatalities.

Second, there is a practical issue. Arc resistant gear is a performance standard. Simply put, you take your equipment to a testing lab as a manufacturer and then intentionally blow it up. They hang pieces of "Tee shirt" material around it and check for burn marks. If nothing is severely burned, the equipment passes. It says nothing about equipment design.

In a similar approach, there is plenty of reliable material out there on constructing "blast resistant" buildings and rooms. Generally the biggest problem is that it is very difficult to actually outright contain a blast. The general approach is to redirect it away in some harmless direction whenever possible. That's the start. But we have a problem. There has been very little test work and almost no actual published data actually quantifying the force of the "blast" from an arcing fault. Without that data, none of the "blast resistant" technology does any good because there's no engineering to go by. Doing the test work is very expensive and time consuming, so that's why it hasn't been done up to this point. I am aware of Ralph Lee's theoretical calculation by the way, and I'm pretty sure based on the bare amount of published data that the theoretical calculation is grossly incorrect. His calculation doesn't even match the evidence I've seen from actual arcing faults so I wouldn't give it any credibility whatsoever.

So to date right now, although in various places you will see it stated that the three electrical hazards are shock, arc flash, and arc blast, you will also almost never see any further mention of arc blast. Arc blast today is in the same state that it was 20 years ago. The best protection is still to avoid "interacting with electrical equipment in such a way that it will cause an arc flash", or if you have to do it, stand to one side and face away when you do it whenever possible. That's not much protection but it's better than nothing. Frankly though I expect that within the next 2-3 years we are going to see a huge leap forward in available data and standards on this particular subject.

To quiestion 1: the working distance to me is pretty clear, the fact that if it's not on the lable because the other 3 options dont include it, to me appears to be an error in the program.

To your point on the risk assesment, is that if the Annex F is "crap" to which I have to agree, then I would find it very challanging to think that an engineer would be able to jump to other methods including IEEE 493 in its application to the determination of PPE level for flash hazards. And I certainly couldnt expect a safety representative to be able to do that. One would think that the NFPA would provide that clearer guidance to the risk assesment component.

I also dont see the ability in the standard to do a risk assesment after the arc flash labels are on the panels. I see the labels dictating the requirments without exception; thus unless the labels are written with several tasks and associated PPE then what's written has to be what's followed.

to your point on question 4; everything has a hazard, although it is proportional to the risk of occurance. motor control centers, which are commonly started and stopped by remote actions, have a potentially greater risk of failure than a dead front breaker panel. which means that all electrical equipment has a hazard, and unless the hazard assessment is completed it cannot be assumed as neglidgable.

My biggest take away is that there's a lot of good in the NFPA 70E, and the standard is very clear on the extreme end of working on exposed energized equipment; but it looses that good clear direction after that.

If I were to propose my greatest improvement to the standard it would be a better definition to the labeling and risk assessment aspect for determination of flash PPE for all work activities.

Also check out a great article that PaulEngr published at the Arc Flash Forum a while back on the subject:
[url='http://arcflashforum.brainfiller.com/threads/2197/'][Paul's Article][/url]

I partially agree here. In certain fields (many process industries, many assembly-line style operations), doing a risk assessment is becoming standard practice. OSHA is already holding these industries to doing risk assessments and slowly expanding the umbrella over time. The incident energy calculation method in IEEE 1584 provides ALMOST the necessary data to quantify the severity of the hazard using those methods. There are two missing elements however. First is an understanding that you basically have to look at the 2nd degree burn threshold as the same as a "life altering injury" or "fatality". If you can make this leap which is far less conservative than it seems at first glance, then we've quantified the severity of the hazard. We can't quite get to the minor injury cases ("1st degree burn threshold") yet, although there is data to support that threshold as well. The second issue is the likelihood. Once you go down the risk assessment path, one thing that most practitioners get really good at is finding obscure and not-so-obscure sources of failure rate data. Outside of instrumentation and valve data such as found in OREDA, the best sources of electrical data are IEEE 500 and 493, and 500 is way out of print and almost impossible to find. This gets us the basic data for likelihoods.



What you have to do is to quantify BOTH parts of it. For instance many plants put it in their safety rules that operating equipment under normal (unfaulted) conditions is acceptable. It is not unusual at all to see various "alternative method" policies such as making it clear that using remote operators, mimic panels, remote racking devices, etc., etc., can be done with no arc flash PPE. The label is just a backstop at that point. If not using one of the "special procedures", follow the label. The same thing is true of virtually all the labels that plaster almost all equipment in any plant. If you just followed the label blindly, you could never actually start and operate most pieces of equipment.



The greatest improvement would be to provide guidance using current safety practices. The problem is that 70E seems to suggest that you essentially have only one option when it comes to arc flash hazards: PPE. That is actually supposed to be the LAST option actually used. Prior to that, one must determine whether or not PPE is even necessary, or whether or not the PPE will do any good. The primary reason for making PPE a last option is that almost all other safety improvements (eliminate, administrative controls, engineered solutions, etc., etc.) reduce the likelihood of an injury. PPE can never do that. It only reduces the magnitude of the injury. It is clearly stated in IEEE 1584 that if adequate or more than adequate PPE quantified under ASTM 1959 is worn using the incident energy calculated using the IEEE 1584 method as a guide, that there is a 90% chance that an arc flash is survivable. This is saying that even if you calculated the incident energy using the best known analytical methods, wore the correct clothing in the correct manner, and essentially did everything exactly as required, if an arc flash were to occur, there is still a 10% chance that you would die from an exposure.

On the other hand, implementing high resistance grounding on average reduces the chance of the arc flash hazard from occurring by 90% regardless of whether the correct PPE is worn or the correct calculation is done or not. Based on IEEE 493 data the improvement is not the straight "90%" that is often quoted but many times it is pretty close to that. In fact using Gold book data, it appears that only breakers are likely to pose an arc flash hazard in the first place while switching, and only drawout breakers are a big issue. Implementing resistance grounding reduces this hazard to below the typical required threshold for a fatality in most risk assessment methods. Plus it does one better. Arc resistant gear, often referred to as somehow safer, only protects someone when the doors are closed and latched properly. Resistance grounding works whether the doors are open or not, protecting both the operator (who wasn't likely to be injured in the first place), and the electrician that has to work on the equipment.

Other imrovements to equipment design such as insulated bus, self-diagnostic microprocessor protection relays, solidly insulated switch gear, vacuum breakers, magnetic actuators, and bolted switchgear (vs. draw out) also provide major improvements in reducing the likelihood of an arc flash in the first place. Even something as simple as requiring insulated tools when doing "low" voltage work (<600 V), or switching from 120 V controls to 24 VDC controls drastically reduces the likelihood of exposure to both shock and arc flash. Yet this kind of thing is limited to a footnote in one section of the entire Code.

Annex F should definitely be expanded. And that would be the opportunity to highlight going after the likelihood of an arc flash in the first place. Right now all the methods being applied are only looking at the magnitude in a worst case scenario, and the "mitigation" methods are only looking at reducing the magnitude of the hazard. Arc resistant gear, "arc flash detection relays", "maintenance switches", bus differential protection, high speed "arc termination devices", various remote racking devices, and so forth all have a single goal in mind. They are all only intended to reduce the magnitude of an arc flash once it is already happening. To me, we should be looking for solutions to prevent the hazard from happening in the first place. That can only happen if we focus in on improving both the reliability of the equipment, and the methods of keeping it that way. Everything else is a bandage trying to slow down the bleeding after the fact.

Sorry that I've kind of gotten up on a soap box here but it seems like this is the original poster's point...that if we are just bandaiding things and not eliminating the problem, we leave ourselves open to as the OP put it, using 40 cal/cm^2 suits to sweep the floors.