OK, I submitted mine as follows:

1. Get rid of Annex F. Going to the terminology "risk assessment" vs. "hazard assessment" is a good move. Leaving Annex F in there in an unusable state is not. I think the best option is to provide a table of alternative methods similar to Annex D but giving no guidance is better than bad guidance.
2. Testing for absence of voltage references IEC/UL 61010. This is actually a safety standard for test equipment, NOT for voltage testing. The only voltage testing standards I'm aware of are IEC 61243.x (several).
3. The cryptic exception to 130.2. Rewrite or get rid of it.
4. Getting rid of "H/RC 0". Partially good move. As written this encompasses two cases. First is cases where likelihood of arc flash is very remote, regardless of the hazard. Example: operating a light switch in an office. The other case is where the arc flash hazard is very low and should be addressed by a minimum clothing standard (formerly H/RC 0). Example: residential electrician working on a 15 A lighting circuit. As it stands the standard addresses the latter but not the former. We need to include both cases where arc flash PPE is "No" as written, and where arc flash PPE is "minimal" (old H/RC 0).
5. DC arc flash section. Still doesn't address a lot of "low risk" scenarios. The tables now start at 100 volts. Folks, the vast majority of DC systems where exposure occurs are 12 VDC automotive type systems, electronics (3.3 or 5 volts typically), 24 VDC controls, and sometimes 48 VDC telecom/controls. 90 and 125 VDC motors fall somewhere in there. Instead of addressing these "common" cases and then leaving it up to the end user to do their own arc flash hazard analysis for cases outside of these very common ones, 70E takes the exact opposite approach and provides guidance for the cases where a hazard analysis would be more appropriate while not addressing the low risk cases.

One I did not submit but probably should:
Annex D, references to tables in NESC were deleted. The issue here is that even though 70E does not address utilities and thus it would be inappropriate to include NESC references in that case, utilization systems in some very large industrial plants use >15 kV equipment for which the only available standard which addresses incident energy in those cases aside from Lee's theoretical equation which gets more silly as the voltage increases (70E inappropriately uses the term "conservative" when it is at 300% of test data from IEEE 1584 at 15 kV and still increasing as the voltage increases), is the tables in NESC.

I need to spend more this.

At first glance, if I do not do a study (i.e. use the tables), there are situations where no PPE is required. However, if I perform calculations, they will always end up with some value >0, even a miniscule one, therefore PPE is always required.

Second, I like the new Section 130.4 Normal Operation, in concept it seems to follow the idea of Risk, Hazard, and Probability as different but interrelated. But, I am trying to think of its un-intended consequences.

IEEE 1584 is a hazard estimation, not a risk estimation. The hazard of being struck by a meteor is a fatality. The likelihood is very low, thus the risk is also very low. Thus, even though it may be fatal, we do not erect meteor shields. In the case of arc flash hazards, if the risk is low enough, PPE is not required. The trick of course is understanding when this line is crossed. The current industry average for arc flash fatalities according to ESFI data is 1 in 100,000. That is also the typical cutoff that I have seen for acceptable fatal risks per year in risk assessments.




It depends on how far the term normal operation is taken. Using IEEE gold book data, arcing fault probabilities for normally maintained equipment is easily under 1 in 100, 000, with one glaring exception. Draw out breakers have roughly an order of magnitude higher failure rates. Data from ABB that I believe comes from CIGRE said that 80% of failures in draw out breakers are the draw out mechanism, and these seem to corroborate the reason for higher arcing faults but this is partly conjecture on my part. Using resistance grounding eliminates line-to-ground faults and fixes the problem, at least numerically. Or, just don't allow drawing out without appropriate PPE. It looks like this would not be considered normal operation since this would expose the bus anyways.

I would like to open a discussion about re-wording the "Infeasibilty" clause for permitting energized work, Article 130.2 (A) (2).
This clause as currently written could be taken to mean that it is infeasible to make a live measurement with the equipment de-energized. This is plainly wrong, since the measurement could be made in a safe manner if the equipment is first de-energised and made safe, according to article 120, and then a means of making a remote measurement is attached, before re-energizing. The emphasis should be on only doing energized work if no other way can be found of doing the work dead.
Not having the right equipment to do such a measurement safely should not be an excuse.
An example of this would be a phase rotation meter with simple prods. These need to be hand-held and applied live, since there is no provision to hook it up and view from a distance. But why not? We should have phase rotation meters (volt meters etc.) that CAN be attached and monitored from a safe distance.

I would also like to see a tighter control of the Exemtions to Work Permit clause Article 130.2 (B) (3) for similar reasons.
I don't think it is acceptable to be allowed to make an energized voltage measurement, using "Energized Work" practices without the full work permit. That would then force people to find creative ways to make the measurement using dead work rather than have to fill out the permit.

I can think of two cases where it applies. First, diagnostic work. But that already has its own exception that bypasses the EEWP section altogether, and your examples fall under that section. So you are right, we don't need it. Second, light poles along public roadways. Those are wired up such that the door is bolted shut and there are only fuses and no disconnect to shut them off for maintenance. This is intentionally done so that some joker doesn't go along the highway and shut off street lighting. To de-energize, the only way to do it is to pull the fuse.

The exception that is given is for a "continuous industrial process". An OSHA letter of interpretation on the subject articulating similar wording in Subchapter S explains what this means and it does NOT mean what most people interpret it as (loss of production). An example would be where shutting off power would in turn shut down a ventilation system which makes it possible to do the work in the first place (controlling a hazardous atmosphere or controlling a hazardous location) such as with force ventilated enclosures located within a hazardous location. Working on this equipment dead may be completely infeasible in the case of for instance a gassy coal mine with force ventilated power distribution...loss of ventilation makes it a hazardous location.

However valid the usage is though, in my mind we can just as easily articulate this as the same thing as the greater hazard rule and thus you are correct.



I still have to make voltage measurements on PRESUMED live equipment in order to test for the absence of voltage, in order to achieve a safe working condition (de-energizing). Getting rid of this exemption would generate nothing but confusion and silly things. The EEWP definitely should NOT be used in this way. This is crazy. We did that ONCE at my current employer. We were generating dozens of EEWP's per day. As a result it became routine and then nobody was even paying any attention AND they started filling them out for basically everything, so we lost the whole point of the process in the first place.

The point of EEWP's is to create an administrative process which reduces the amount of energized work to as little as possible. If you dilute it by making it routine, people will just fill out paperwork blind and pretty much ignore it. That is where we ended up at when it became necessary to do one for every single electrical lock out, because it got misinterpreted as being required for testing for absence of voltage.

Also, note that the way that the thing is written, you really need to make sure that it is a PERMIT...should require fairly high levels of approval such as safety manager, plant manager, engineering/maintenance manager, etc. Definitely never, ever just a foreman signing off on them. You don't want low level guys making decisions that put people's lives at risk. But if you make an EEWP necessary for nearly every single electrical job, especially every one of them that requires LOTO, that is exactly what will be required in order to make the process practical to implement.

The justifications by the way got to be insane as well. The infeasibility rule was being abused by either putting down reasons like "LOTO" (which is spot on), or "continuous industrial process" (someone read 70E but misinterpreted what the FPN meant).

No thanks, from practical experience this is really bad practice and policy.

letter of interpretation on the subject articulating similar wording in Subchapter S explains what this means and it does NOT mean what most people interpret it as (loss of production)