It is currently Sun Oct 22, 2017 3:08 am

Post new topic Reply to topic
Author Message
ekstra   ara
 Post subject: Cancer and electrical injuries: hazards vs. risk
PostPosted: Mon Feb 08, 2016 10:00 am 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 1779
Location: North Carolina
Time and again, the issue of "risk" vs. "hazard" comes up and not just with electrical safety but with other items as well. A similar problem arises whenever the big public fear, the big C, CANCER, comes up. Although we've made great strides towards improving both electrical safety, reducing the likelihood, and improving the survivability of cancer, there is a still a tremendous amount of fear and uncertainty about both. And with both of them part of it is the fault of so-called "experts" that if nothing else do a great job of failing to communicate the risk in a meaningful and understandable way.

The International Agency for Research on Cancer (IARC), a division of the World Health Organization (WHO), is the go-to authority on cancer hazards. Among other things, IARC has classified the following as carcinogenic:
Alcoholic beverages
Hormone replacement therapy
Baby oil
Outdoor air pollution
Working as a painter
Salted fish
Processed meats
Oral contraceptives

IARC has also classified the following as probably carcinogenic:
Fried, roasted, or baked food
Emissions from fried food
Hair dresser/barber profession
Cooked meat and fish
Grapefruit juice
Night shifts
Red meats

There is a third list which most substances are on, those which the IARC has not specifically ruled out as not causing cancer but neither have they claimed that there is yet sufficient evidence to show that it is carcinogenic. Among other things, water is on this list.

Furthermore, there is only a single item on the IARC's list of things known not to be carcinogenic. caprolactam.

IARC is very clear that they do not consider whether there is any likelihood of getting cancer, only whether or not the substance in question might cause cancer no matter how extreme the circumstances.

Taking as an example there is a growing amount of evidence that people that eat the most red and processed meats are 17% more likely to develop bowel cancer than those who eat the least amount. But the question is 17% of what? According to data from the UK, the average incident rate of bowel cancer is 61 people out of 1,000. The "high risk" group is around 65 cases out of 1,000 while the low risk group is 56 cases out of 1,000.

The reason for the issue with red meat in particular is that a particular form of the iron compound that makes up the hemoglobin in blood sometimes breaks down into N-nitroso compounds in the gut. These compounds can damage cells that line the bowel and when this occurs, it causes more cells to have to replicate to replace the damaged cells. With more cell replication going on, there is a higher probability of an error in copying the DNA, which is what ultimately causes cancer in the first place. On top of this processed meats have nitrite preservatives which also cause N-nitroso compounds to form.

Similarly, apples and pears are on the IARC's list of things that can cause cancer. Formaldehyde and amygladin are both substances which are known carcinogens. Both also naturally occur in apples and pears. An apple has about 22 parts per million of formaldehyde. Formaldehyde is also produced by your own body. But it's a matter of dosage and even competitive eaters could not eat enough apples to have an appreciable risk of cancer.

Sand is another boogeyman of the cancer crowd. How we get there is somewhat strange. Sand or more specifically :"free silica" in the right particle size causes irreversible damage to lung tissues. There are two critical parameters for this to occur. First the particles have to be small enough to make it all the way into the deepest parts of the lungs (the aveoli). Second the surface has to be chemically active. Chemically silica is a silicon atom bonded with 2 oxygen atoms or SiO2. Over time in the environment chemical reactions will occur that "neutralize" the silica. However in freshly broken surfaces such as from grinding or blasting, some of the silicon atoms will be temporarily missing an oxygen atom which makes them chemically active. By itself even as horrible as the disease known as silicosis is, the risk of cancer is not appreciably increased. However it was found that the rate of cancer in heavy smokers was increased if they were also diagnosed with advanced cases of silicosis compared to smokers alone or those with silicosis alone. Thus on this basis the IARC classified sand as probably carcinogenic.

Clearly we are missing something here. We clearly can't treat smoking or asbestos in the same way as we treat meat, fruits, vegetables, entire occupations, and play sand. Yet failing to recognize the difference has caused oddities such as including a cancer warning on bags of play sand. Not only does this cause needless public confusion and downright unwarranted fear but over time it is hardening our perceptions and causing us to discount the real risks because of the lack of credibility that it causes.

What is missing is the concept of risk. It is not enough to know that for instance that eating a hamburger can increase the risk of developing bowel cancer. It is far more critical to be able to put that information in perspective, to weigh the actual risk. In a similar way we can screen shock hazards based on their potential for causing fibrillation. In some cases such as the basic function of an EKG a shock is not only acceptable but necessary to operation of the equipment. Similarly we can also consider arc flash. Some equipment may not even be capable of generating an arc flash no matter what the magnitude of the hazard is. This would be akin to play sand that was mined, processed to remove dust, and bagged but never ground or blasted. And some equipment may be capable of producing an arc flash but the magnitude of the arc flash is not significant, equivalent to eating fruits and red meat in terms of cancer risk.

Furthermore although the IARC has a list of about 90 items on the "definitely carcinogenic" list and hundreds on the probably and maybe lists, it is interesting to note that there is only one substance on the "definitely not" list. Why is this? Simply because it is extremely hard to prove a negative. Sometimes the evidence is extremely weak such as the case of silica which only shows a positive result in the presence of another known carcinogen such as smoking.

Right now the thrust of most electrical hazard assessments is on just that, the hazard. This is the same thing that the IARC has been doing for years with cancer. However just as the IARC's evaluations are nearly worthless at best and cause needless panic and fear mongering at worst, we've got to keep electrical hazards in perspective. Performing an arc flash and shock HAZARD analysis is only the first step. In fact it could be argued that it is not even a necessary one in many cases. For example if the task in question is to push a "push to test" button and inspect the indicator lights on an MCC, there is almost no chance of an arc flash occurring during this task and thus the magnitude of the hazard is not even of interest.

On the flip side performing an arc flash hazard analysis on the potential from a cell phone battery could definitely be useful in determining whether or not any task, or any abuse, would even matter in the first place. However frequently especially with fringe cases such as "arc welders" it is extremely hard to prove that there is no significant hazard. Thus we fall into the same trap that the IARC is is extremely hard to prove that there is no hazard compared to proving that a significant one exists.

The next step is to screen out the tasks which place personnel at risk while operating and maintaining the equipment that MIGHT be a hazard. This is what a risk assessment is all about. In some ways this step is far more challenging because there are more unknowns. For instance we have the famous paradox if "Shrodinger's Cat". This is a thought experiment in which we have a box that contains a cat. Due to details of the experiment there is a 50% chance that the cat is alive. However we cannot know for certain until we open the box whether the cat is alive. Thus the question is whether we treat the cat as alive or dead prior to opening the box. In a similar way when doing an electrical risk assessment quite often we are challenged with only the external appearance of the equipment to determine whether or not there is a significant hazard. If we could hear the cat or shake the box and detect movement, we could probably improve our odds of knowing whether or not the cat is alive. Similarly if we consider recent history (did a breaker trip) and make an external cursory examination of the electrical equipment for tell tale signs such as burn marks or liquids running out of the panel, we can improve the odds on knowing the state of the equipment inside the panel. But at the end of the day until we open the panel and can inspect it or perform physical testing on it, there is no way to know with absolute certainty whether an arc flash can occur. But we may be able to determine this with some level of confidence.

Returning to the cat, what if the circumstances of Shrodinger's experiment had changed? What if we knew ahead of time that the odds of a dead cat were only 1 in 10,000? What if it was 1 in a million? Then prior to opening the box would one guess that the cat was alive or dead? Similarly if the chance of a serious arc flash accident for a certain task with a certain piece of equipment occurring was 1 in 100,000 and that if one were to occur, it would cause an injury that would require hospitalization, is this a risk that would be considered acceptable? The risk of death as a pedestrian as of 2002 is 1 in 47,273 per year. Thus the chance of dying while walking to the job location is twice as likely as our proverbial 1 in 100,000 arc flash. And driving to and from work has a 1 in 17,625 risk per year. Clearly on a comparative basis the risk is far greater of a serious injury or fatality from sources other than our theoretical electrical panel. These numbers are not made up by the way. The likelihood of an arc flash while racking a breaker in or out of a cell in draw out switchgear is approximately 1 in 10,000 while opening or closing a disconnect is better than 1 in a million. Thus we can determine in some cases that PPE is categorically required while in others it is not required in most cases.

Looking at it another way, what if the exact same electrical panel had just tripped due to an electrical fault and the task was to troubleshoot the system to determine the cause? Is the likelihood still 1 in 100,000? According to UL 489, a 100 A molded case circuit breaker must withstand at least 50 operations at 600% of rated current, and only a handful at rated interrupting current. NEMA AB-4 the standard for molded case circuit breaker maintenance referenced by nearly all molded case breaker manufacturers specifically requires a visual inspection after every trip. Clearly the odds of a failure occurring following a trip are drastically increased relative to the odds prior to a trip, even under relatively mild interrupting conditions. Under these circumstances does this change our perception of risk when it comes to this troubleshooting task? On a comparative basis is the likelihood of an arc flash now more likely than other work place risks? Does this then set up a situation where we may want to take a different approach to the task compared to the previous case? This is the second part of performing a risk assessment. With equipment which is in good working order, we can predict the likelihood of an arc flash with some certainty. However circumstances which can only be determined by inspection in the field can alter the likelihood drastically, just as shaking the box with the cat in it may provide clues to whether or not it is alive.

As a case in point around 2001 an electrician in a chemical plant was sent to troubleshoot a problem with a small 10 HP pump. The circuit breaker (not the overload) had tripped multiple times and the operator had reset it several times. On arriving at the scene, the electrician also reset it a few times to confirm before opening the circuit breaker and testing the motor leads. Eventually the source of the problem was found: the motor insulation was damaged and the motor was shorted out. The defective motor was replaced. On closing in the circuit breaker to put the pump back in operation the circuit breaker exploded and the electrician was hospitalized from the resulting arc flash. The circuit breaker was never inspected during troubleshooting nor before placing it back in service.

This case serves to make the point. Under normal circumstances the likelihood of a small molded case circuit breaker having an arcing failure is very small, much less than other industrial accidents that cause severe injuries. On a comparative basis requiring PPE for operating the breaker is akin to requiring NASCAR-style safety protection including roll cages, FR rated coveralls, helmets, gloves, 5 point safety belts, and neck restraint harnesses for the purpose of driving a vehicle to and from work and both should be required if PPE is required for operating the breaker because the risk (likelihood) of the automotive accident is far greater. On the other hand if the breaker just tripped on a fault then it should never be closed back in or put in service until it has been inspected. The circuit breaker is likely to have been under extreme stress interrupting the fault and it is designed to survive only a limited number of such incidents.

Finally, there is no "one size fits all". The hazards of working in the financial offices of an oil company would be very different from those of working for the same company in a retail gas station, or in a refinery, or on an offshore oil platform, or an on shore drilling rig under construction. On a comparative basis, what is acceptable for one environment cannot be assumed to be identical to the others.

In summary:
1. Hazard analysis whether it is in the form of the IARC's evaluation of carcinogens or an IEEE 1584 arc flash hazard study is incomplete when it comes to safety. Risk is equally critical in the analysis because the hazard analysis is only useful in screening ot those cases where a hazard may not exist, a task which is often inconclusive due to the inherent difficulty in proving a negative.
2. Risk is useful in screening out those cases where irrespective of the hazard, the risk may be inconsequential such as eating fruits and vegetables, or normal operation of electrical equipment.
3. Risk may be ever-present such as when smoking or inserting a circuit breaker into an energized cell in drawout switchgear. Risk may also change depending on circumstances, such as performing a chemical extraction of the toxins in apples or following an electrical fault.
4. Risk is never known with any degree of certainty...there is always an element of chance involved. Not all smokers die from cancer. Arc flash is rare even when equipment is defective or poorly maintained.
5. Keep things in perspective. Fear mongering over red meat, play sand, and toxins present in fruits and vegetables is just as crazy as fear mongering over electrical equipment. Given the right circumstances all of these can be deadly. It is only by comparing the risks that we can determine what is acceptable and what is not. And it is only when the magnitude of the hazard AND the likelihood of occurrence are elevated that the risk is significant enough to warrant action to lessen it.
6. At the end of the day, evaluating risk objectively is inherently hard and somewhat philosophical in nature. The only practical way to evaluate risk is by comparing known conditions to unknown conditions.

 Profile Send private message  
Reply with quote  
 Post subject: Re: Cancer and electrical injuries: hazards vs. risk
PostPosted: Mon Feb 15, 2016 9:51 am 
Sparks Level

Joined: Fri Apr 15, 2011 7:43 am
Posts: 151
Location: Colorado
Evaluation of any risk is a normal, daily task. Using 120V circuit is done by nearly everyone and is generally safe but it could kill you - just shower with a toaster, once. 95% of all the people knows there is some risk but they do not understand "real" risk. Like wise, I can come to your plant and issue an arc flash study but I cannot comment on the "real" risk of your plant. if you have competent people that have a a zeal for safety is a far different "real" risk that the plant that does not place the same importance on electrical safety.

I agree with your statement but how do we get there? Who is the one to determine the "real" risk? Who is really qualified?

 Profile Send private message  
Reply with quote  
 Post subject: Re: Cancer and electrical injuries: hazards vs. risk
PostPosted: Thu Feb 18, 2016 1:36 pm 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 1779
Location: North Carolina
engrick wrote:
Evaluation of any risk is a normal, daily task. Using 120V circuit is done by nearly everyone and is generally safe but it could kill you - just shower with a toaster, once. 95% of all the people knows there is some risk but they do not understand "real" risk. Like wise, I can come to your plant and issue an arc flash study but I cannot comment on the "real" risk of your plant. if you have competent people that have a a zeal for safety is a far different "real" risk that the plant that does not place the same importance on electrical safety.

I agree with your statement but how do we get there? Who is the one to determine the "real" risk? Who is really qualified?

This has been discussed ad infinitum. There are already many existing standards for doing this. It's just that the "electrical industry" may not have been exposed to them in the past. In terms of qualification you can get training and certification. Having hundreds of hours of risk assessments under my belt, I can honestly say that you want "everyone" on the team with representatives from all the major groups (safety, environmental, production, mechanical and electrical maintenance, engineering, R&D, etc.) The reason is partly for buy-in purposes and partly because everyone has a slightly different "take" on what they see as the risks. Thus with a good cross-functional team you get good coverage of the possible risks. Without this you will miss more. You start with a risk assessment procedure developed from a standard. There are many standards out there. 70E-2015 contains all the elements of this.

Generally the standards all revolve around a "risk ranking matrix" and they all claim that this is specific to a company/country/plant but from personal experience we all have roughly the same intuitive notion of what "risk" is (within an order of magnitude at least). Generally every time I do one at least 1 or 2 people on the team have ZERO experience. I do a quick training class with them which takes all of about 10-20 minutes. It really is that easy to do.

You need three things for doing the risk assessment itself once you have identified a specific hazard. First you need to know the magnitude of the hazard. For both shock and arc flash this is actually very easy to do because we really only need to know if we are above 1.2 or 2 cal/cm^2 or not. We can definitely quantify the point at which an arc flash or shock may result in a severe injury requiring hospitalization. With shock we can either use the 50 V cutoff, or dig into the time and current or voltage to look at Dalziel's information but again usually the cutoff is low enough that its a binary decision. We can't quite get there in terms of a fatality but you don't really need that.

The second piece of information we need is a measure of the likelihood of occurrence. This is where it gets tricky. As adults we generally have enough experience that we can intuitively guess at whether an incident occurs up to around once every 10 years or maybe once a century but our intuition doesn't get us any further, certainly not to the level of say once in a million. So this is the point where we need a source of statistical data to get us there. Safety data won't get us there but fortunately we hae lots of reliability data that works almost as good. The only tricky part is that it contains failures that aren't necessarily a hazard. Thus quickly gives us a good estimate of the data that we don't have. We don't have a lot of this but again, 70E-2015 has a big chart of this in the new tables.

The third and final piece of information is that we need to have a ruler to be able to judge whether or not the risk from the hazard is acceptable. For this the best available measurement is to compare the hazard to other similar hazards such as looking at the number of automobile fatalities per year or looking at the rate of recordable injuries per year for a particular injury. This gives us a comparable basis for getting there.

Taking this all together we get:
1. If something bad happens, will the person be seriously injured requiring hospitalization or a fatality? If not, then for electrical hazards (shock and arc flash) at least right now the standards go no further.
2. What is the likelihood that "something bad" can happen? For this we can look to reliability data as I mentioned or even easier, just look at the table that the 70E Committee already developed for this exact purpose.

 Profile Send private message  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 3 posts ] 

All times are UTC - 7 hours

Who is online

Users browsing this forum: No registered users and 1 guest

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:  
© 2017 Arcflash Forum / Brainfiller, Inc. | P.O. Box 12024 | Scottsdale, AZ 85267 USA | 800-874-8883