I wanted to bring this into its own thread because there is quite a bit of discussion and disagreement about the need for PPE when performing deadfront operations.

In the beginning, 2004, we were exclusively talking about needing PPE for live work. Dead front operations such as switching circuit breakers covers on, safet switches & mcc's door closed, on/off buttons, even switchgear on/off as being acceptable without FR PPE. Then in the 2009 edition we got this:

"...In several cases where the risk of an arc flash incident is considered low, very low, or extremely low by the task group, the hazard/risk category number has been reduced by 1,2, or 3 numbers, respectively. The collective experience of the task group is that in most cases closed doors do not provide enough protection to eliminate the need for PPE for instances where the state of the equipment is known to readily change (e.g., doors open or closed, rack in or rack out). The premise used by the Task Group is considered to be reasonable, based on the consensus judgement of the full NFPA 70E Technical Committee."

There was some discussion that the NFPA meant to reference switchgear exclusively. But I can't find the reference. Without searching, and taking the code as written, we now argue over what can and can't be done without FR PPE.

Lets all see if we can arrive at a point of reason.

I'll list some tasks in increasing risk, stop where you feel appropriate:

1. Turning on a light switch in an office

2. Turning on/off breakers in a lighting panel covers on:
120/208 (10KAIC)
120/208 (64K AIC)
277/480 (14KAIC)
277/480 (65KAIC)

3. Turning on/of safety switch, door closed:
250V 60A
250V 100A
250V 200A
250V 400A
480V 60A
480V 100A
480V 200A
480V 400A

4. Operating disconnect or on/off buttons on MCC, doors closed, low voltage, 65KAIC
Starter size 1,2,3,4

5. Cable tray work, all wires and cable with 600V insulation.
Seperate cables gloves off
Take amp probe reading gloves off

6. Switchgear - I think we all agree that any operation involving racking or switching requires PPE. But
a. reading gauges on deadfront

Am I alone in this or are there others who find 'the line in the sand' a little difficult to distinguish.




3.

I had the some of the same questions I typed up yesterday to further research.

1. Do I have to wear the posted level of PPE when opening/closing the CB of a bucket in an MCC to lock it out with all covers and doors installed?
2. Do I have to wear the posted PPE after locking a bucket out and am only working on the load side it? If so, once the bucket in an MCC is locked out and the top side of the CB is still energized, do I have to wear the posted PPE to verify the load side of the CB is de-energized? Once it is verified that the load side is de-energized, is it ok to work at a lower PPE rating and remove the un-required PPE?

I am of the opinion that the 'only' deadfront operation that requires PPE is operation of switchgear, opening or closing a breaker, racking or out doors closed.

However, since the theme here is targeting a devices that have a 'high' change of energy state, then I would carry that through to any device that switches an equivalent amount of energy. There are not many that I know of other than switcgear, but, and I say this with reservation, maybe a MCC bucket or safety switch, switching 800A or more.

For everything else, from the safety switch in the field, to the light switch in the office, if the covers are on then its not live work, nor is it a high energy change of state.

I'll agree that there should be some cut-off point for deadfront work.
The problem, of course comes in where that cut-off is.

Haze, you list operations with equipment and the equipment rated AIC. I don't know if you can make the AIC rating a cut-off point, for the very reason that the rating may not reflect the actual available maximum fault current.

We currently have a lot of old equipment. Some of it may be only rated 480V / 14KAIC. We have made the decision that when replacing it, we are going with 65KAIC rated equipment. This same equipment may or may not currently be overdutied, but to allow for future growth, to simply equipment, and to allow for some degree of flexibility later, we are specifying the higher AIC rating for new equipment. (We're a gov't site, it's not our money...)

The end result, of course, is that the rating of the equipment has no bearing on the actual conditions of available fault current. In fact, if you listed the AIC ratings with the thought that the higher rated equipment may pose a higher hazard, I would argue that, in our situation at least, that the higher rated equipment has a lower hazard, because it should be rated higher, in many cases, significantly higher, than the available fault current.

Now, back to the original question:
Where is the cut-off for deadfront operations? I'm not sure. I would put it somewhere between switchgear and panelboards.

At present we don't require PPE beyond standard issue level 2 minimum clothing (no faceshield) when isolating equipment for maintenance. Operators can throw a disconnect and perform LOTTO without taking special precautions beyond wearing site PPE.
If the same disconnect is opened - full arc flash PPE to the IE level is worn for testing voltage. The worker may gear down when absence of voltage is proven and there is no inadvertent exposure to hazardous energy.
All HV and MV switching is done with arc flash PPE to the IE level.
We require our workers to wear a faceshield and class 0 gloves when working inside a 120/208 volt panel regardless of KVA or equipment AIC. We don't have any restrictions when operating deadfront 120/240/480 breakers.
The long term goal should be to protect the worker at risk - not gear up because there is a hazard. Both NFPA and CSA recognize a risk based approach but design their tables to eliminate the risk with the table notes. A blended approach may be better but maybe not? Risk, probability and energy level need to be factored. I struggle with telling a worker that a 100 cal hazard is probably safe because the risk of failure is very low so I tend to fall on the side of following the IE values when possible. It sure would be nice to be able to predict the impending failure - I could make alot of money if I could do that!
I like to have my workers gear up when possible - extending that to operators and other unqualified workers doing normal day to day tasks makes it challenging in training and compliance. The hazard is there but the risk is so small that I don't see a payback. The previous statement is assuming proper maintenance of equipment, and training of all workers interacting with the equipment to recognize electrical hazards that may occur (loose doors, unknown condition, equipment failure, etc).

Haze, I agree with you. This seems to be a subjective and biased argument with no clear methodology leading towards a solution. As someone maybe you said, “I see more risk driving to work each day�.

IMO, The real problem here is safety management not qualified in electrical knowledge over complicating, for the sake of the company liability, when in the actual world the hazard is minimal. Not all cases but many.

Good discussion so far. I'd like to hear more views.

I used voltage and AIC of the equipment as an indication of increasing risk. The NEC 110 mandates that equipment be rated to handle the available fault current for the installation. We have to have some basis for discussion.
Obviously if equipment is not rated properly, or if doors are left open or with missing parts, then the whole basis of the question is out the door.

I am assuming, that a company that has a well designed, and well run arc flash program, will have the basics covered like having the correct rated equipment with good doors and covers that work.

If the equipment is under rated for its available fault, then the equipment would probably explode and the flying projectiles will cause more injury than the arc blast. This is why we will be next going to 'kevlar' thermal armored PPE.

Presently, manufacturing in the USA relies on its operators to perform basic lock out/tag out. Mostly, this is to clean equipment rather than repair it. Most businesses are looking to only have to pay for FR PPE for their electricians, and not every employee. This is not a 'little' hit financially to a company. Depending on well its designed, and how demanding the workforce, FR PPE for one employee can be $500 to $1000 a year. Have you ever had an electrician throw his 8cal shirt to you, and say "da...a button came off and I know this PPE has to be rated so I can't sew the button on myself".

Operators typically will not have an FR exposure. Performing a lockout on a 60a safety switch or size 2 MCC bucket, for most employers, does not warrant full arc flash training, clothing, PPE, and retraining. I can not see where it is required in Art 130. Yes, bad things happen to good people. But we have to run our manufacturing in a global environment, and that now means that every investment has to be justified.

So the question remains, where do you stop. If you have a better way to rate the risk, I'm all ears.

But that risk is mitigated by airbags and seatbelts, the danger is higher if you disable your airbags and don;t wear seat belts.

The danger is also higher if you are in a small compact, "fuel efficient" vehicle, but I don't see the government mandating armored cars for everyone.
In fact, they are going the other way...

I had a thought after the comment about me using AIC as a sorting filter. How about we use Amp rating instead. It occurred to me that the fault current might not be known, and that the equipment may not be rated with an AIC to which it might be subjected. But the cable size for its load is almost always correct, or at least very close to being correct. The largest impedance to any single load remote from the main supply is going to be the cable. If you have a 200A safety switch the cable is going to be 3/0 or 4/0. If you got 30A its going to be 12,10 or 8 awg. My thought here is that the cable, even if relatively short in length, is going to limit the fault current let through. So we could possible change our cutoffs based upon main or branch current rating. So I could switch any breaker if the main breaker or MLO is say 225amp or less, any low voltage. MCC could be buckets to size 4. Safety switches could be 400A, etc. Not picking those values, just asking if you think its a better rating method.

I can tell you, that in my opinion, a line has to be drawn if we want businesses to continue. Operators have to be able to perform LO/TO without FR on deadfronts, where reasonably acceptable.

See attached file
I would love to support the argument that the switch size could be used to determine the need for PPE but I have some problems with that scenario. Take an MCC with a 33 cal/cm² hazard out in the sticks with relatively small loads. The Arc Flash rating is high. 40 cal suit to lock out a 15 HP motor? You bet!
IMO: take a type of operation - simple lockout, apply a risk factor then assign the PPE level required. Just because there's an IE posted doesn't mean we suit up to the level if there is no exposure to energized parts and equipment is in good condition. I think it's all or nothing in this case, 40 cal suit or birthday suit
It's a little crazy trying to implement anything in between. How can anyone consider trying to defend using anything but the all or nothing approach to PPE? Either it's needed or not - plain and simple...
I think deadfront lockout should be exempt from the PPE requirements beyond common sense PPE unless there is a real need like poor or unknown equipment condition, obvious hazards (environmental, etc.)

Is this based on the CB at the top? 2 s rule? What's the OCPD protecting the 1-3/C 350 1243 ft feeder to MCC 74140?

Ia for MCC 74140 is over 20 times the normal current of MCC 74140, and over 12 times the current carrying capacity of the feeder. It should be relatively easy to engineer out the hazard or at least part of the hazard. I understand that until it's done, it's not and you have a 33 cal/cm^2 hazard lurking...

Protective devices

In my example, no instantaneous trip element in the MCC OCPD combined with a very high cable impedance leads to a high IE. I absolutely agree the hazard can be engineered out - all it takes is time and money! One negative about performing a study is that you find out how much work you need to do (and how much it will cost).
In the meantime, all Haze's comments are valid. What about the guys that go out and service machinery? Before AF regulations they could go into the field and lock out with impunity. Now they must abide by the label even though no-one can provide the data to prove their type of interaction is dangerous. It's not right but not completely wrong either. That's why I think risk ranking is the way to go regardless of the system characteristics. Safe or not - that should be the test. I'll live with my high IE's when exposed to the hazardous energies but why do I gear up when the equipment is being operated as it was designed (turning on and off a breaker or switch).

I will agree that the amp rating of the equipment is probably a better indicator. But formulating that in policy, and implementing it are two different things.
The machinist going to lock out a pump, or a welder is not going to check the amp rating of the disconnect or the MCC it's connected to. We have some equipment that is so old there are no ratings on the outside at all!
Even an electrician is not likely to check the rating before just pulling the switch.

There is enough confusion already with arc flash policy. From a practical standpoint, it may be better to make the cutoff the type of equipment, say switchgear, MCC, panel, disconnect. In this case I would be tempted to draw the line at the MCC, requiring PPE for MCC's and switchgear, but most electricians I've talked to say they've seen more incidents with disconnects, where one phase may fail to open or close with the handle, or a broken piece may fall out and onto the lugs during switching.

This is why

This is why...

Your guy can show me his stuff in a couple of weeks
Meantime, how do we keep the general workers safe? I still can't fathom why there wouldn't be an allowance for normal operation of equipment - after all that's what it was intended to do!
Call me crazy...

One could argue the arc flash safety program is unique when compared to other safety programs. This program requires a value to be determined and assigned to the worst case possible failure or accident. The worker is then required to be protected (100%) up to that level if mitigation is not possible. Can you think of another program with such requirements?

Typical industrial PPE is designed to provide a high level of protection well beyond nonuse but it all has limits. A hard hat will not protect for all possible falling objects. Safety boots will not protect for all crush hazards. Even the seatbelt and airbags, mentioned above, only provide a degree of protection; they are not designed to provided 100% protection for all possible accidents. There are hazards associated with an industrial setting no level of PPE can protect against. There has to be a balance between providing protection but not creating additional hazards while trying to protect. It continues to be obvious more testing is needed in order to better refine the science being used to make decisions and write policies.

In case of arc flash, the worker is not protected 100% against the worst case accident. In fact, mandated PPE from an IE study still maintain a 50% chance of having a second degree burn underneath.

Machine safety can usually be condensed to "don't let people touch dangerous things". If they can't touch, they can't be harmed.

Zog,
Are those photos not of switchgeaer. Can't tell if the one to the left is MCC.

Canuck,
Is not your IE for that MCC based upon the MCC bus. Any bucket is going to be connected with smaller cable. Plus the switching of the disconnect is only going to affect the load side of that cable, which would be a much lower IE.

Dean,
Of those safety switch accidents, the ones with the doors closed, what percentage resulted in injury.

Obviously, electrical equipment has failed in history, and will continue to fail in the future, with simple operations like routine switching. What I am trying to ascertain is how to incorporate the NFPA recommendation into a working program.

Let me ask it this way. If one wants to merely conform to the least restrictive interpretation, would you say the guideline is limited to 'Switchgear' only. MCC's, safety switches, panels, are not specifically required?

This section of the code is subject to much interpretation, so its a valuable discussion.

At my previous employer, we did NOT require FR for LO/TO except switchgear. We discussed it. But with some 100 operators, all with company paid 4 uniforms/year, it would have been a prohibitive expense. Operators NEVER had exposure to live work, only MCC buckets size 4 or smaller, and safety switches to 400A.

I get asked the question frequently by friends in other industries as to when does deadfront operation require FR. What do you tell them.

[quote="Vincent B."]In fact, mandated PPE from an IE study still maintain a 50% chance of having a second degree burn underneath.


Can you please elaborate? So if the calculation says 8cal/cm2 and an 8cal/cm2 PPE solution is used then the worker still has a 50% chance of being burned? What if the result is 3cal/cm2 and the same 8cal PPE is used, are the odds still the same?