This week's question was submitted by one of our forum members. Here it is.

Consider this scenario: a qualified electrical worker is troubleshooting an energized 120 volt AC circuit while wearing all appropriate PPE and following appropriate procedures.

The worker discovers the problem is that an energized wire is loose at a screw terminal.

De-energizing the circuit will introduce a greater hazard in a continuous chemical process.

If the worker is wearing insulated gloves and uses an insulated screwdriver, do you believe an energized electrical work permit is required before tightening the screw terminal?

It seems trivial however, it appears NFPA 70E could be interpreted to require an EEWP for this task.

Do you believe an energized electrical work permit is required for tightening a live 120 V terminal?
Yes
No

If you have a question you would like to see asked, send it to jphillips@brainfiller.com

I believe the answer is yes. There's always the possibility that tightening the lug or screw would cause a momentary lack of conductivity (if the terminal is damaged), possibly shutting down the control system circuit or power circuit (not defined as to whether it is a control or power circuit). This would introduce the hazard mentioned.
At least management would have an opportunity to make the decision and procedures put into action to respond to the shutdown (if it were to occur).

Tightening an energized screw does not constitute "testing, troubleshooting, or voltage measuring." Some might argue you are just troubleshooting to see if you have a loose screw, but that probably wouldn't hold up with the AHJ.

In practice I don't want to have to go through the permitting process just to perform the safe operation of tightening a 120 V screw terminal, but I think technically you should have to.

While it may seem top be a pain, I believe a EEWP would be required to tighten the bolt with power still live.

I might add that one also needs to check the premise that shutting off power would "create a greater hazard". I'd be curious to see a "continuous process" that could not be shut down safely. One would not design such a process as all processes need to be shut down eventually for repairs or other contingencies. That is not to say the shutdown could be done quickly, inexpensively or conveniently. However, there is nothing in the regulations in which those items constitute a "greater hazard".

This of course puts both the electrician and the permit writer in an unenviable position of either going against code or having a costly shutdown. Which for at least the short term would mark these people as obstinate and hard to work with. I believe the goal of the code writers had something else in mind. This would be for someone very high up in the food chain to sign off on situations where one is claiming a greater hazard. This person would then tend to drive systems to either have redundancies or designs such that the power to that little 120V terminal (and others like them) could be more easily could be interrupted so one would not have to perform those live tasks.

This is a perfect example of what the "annual work permit" is ideal for covering.

Two issues here. First, there is the initial tightening of a screw. This would lead to the concept that someone is landing a wire onto a terminal while energized. While it can be done, the ensuing arcing can and often does damage the threads and/or other components causing burs and other mechanical defects that lead to an improper (loose) connection, setting things up for further failure.

The other possibility is that this is a case of retorquing a bolt. I don't know how to say this any more plainly other than the fact that this is incredibly bad practice that should not be done. The reason is because of simple bolt physics. Assuming we're talking about a screw terminal the desirable clamping force is in the axial direction. As the force increases, we increase the number of contact points which reduces ohmic resistance at the joint, up to a point. With copper in particular typically the metal is actually distorted and actually cold welds together...this is the reason that a proper joint whether crimped or screwed will permanently deform the strands in the wire in such a way that clipping and redressing the wire is the only way to ensure a "perfect" joint...ends shouldn't be reused. And yes, I know the practice is done all the time.

Getting back to the holding force thing...only about 10% of the force from torquing a screw fastener actually goes into the desirable clamping force. The rest is basically just overcoming friction in the screw threads. This is the fundamental reason that torque specs are so "loose" and the reason that measuring torque with a torque wrench/screwdriver is an imprecise measurement at best...because we're mostly measuring friction forces in the threads and not the clamping force itself. The second thing happening here is that the fastener actually stretches slightly and the microscopic roughness at the fastener surface is flattened out a little in the process. However within the crystal lattice of the metal in the fastener something happens where molecular-level defects actually slightly unravel or shift to reduce the induced stress in the joint. This is known as dislocation theory and you can physically see it happening with a TEM (transmission electron microscope) in very thin samples of steels. Here is a youtube video of one:
https://www.youtube.com/watch?v=BV1cxwxnhPs

About 25-50% of the force measured with a torque wrench actually dissipates over the course of a week but the loss in tension occurs even within minutes after tightening a fastener. The result is that if you test a fastener for looseness after tightening it, you will find that they are all loose. Thus the natural response would be to retorque it. However what is really happening is that the joint is being overtorqued and that over time repeated retorquing will CAUSE fastener failures.

A properly torqued joint will relax and do all the other things I described but will NOT fail. The hysteresis effect (mostly from cold welds) will cause electrical joints to perform well even under "undertensioned" (read: relaxed bolt) conditions throughout the live of the fastener. There are a couple text books on contact theory which explain this much better but suffice to say that two things:
1. Retorquing it an incredibly bad idea.
2. You can't test for torque with a torque wrench except at the time of application.

If you are really worried about fastener loosening then I have two suggestions. First off, through away all of those stupid split washers you get in all your fastener bags. They have been proven to actually CAUSE looseness. Second really the only proven systems other than safety wiring are constant tension joints such as the various small terminal blocks with spring clamps use, or the various "locking thread" type designs. For mechanical joints this means cheap nylon locking fasteners. For electrical connections this means defective thread designs (there are many).

And the second issue assuming that we're actually tightening a 120 V wire into a terminal block on a de-energized connection and not doing any of the "bad idea" practices is this. What if the tool or a part of the body or a part such as a nut or screw "gets away"? Is it going to fall down or accidentally get stabbed through into some place which will cause an arcing fault even though there was no hazard from an arcing fault with the 120 V circuit? Is there a chance that any of these sorts of events can lead to a shock injury? Granted with around 100 mA being the line where a fatal shock can occur and with approximately a 1 K-ohm body resistance as per IEEE Standard 80, that gives us 0.1 * 1000 = 100 Volts so we are very near the line where a fatality can occur but nevertheless the possibility exists. So if the task is being done with an insulated screw driver and the exposed tip is smaller than the gap between terminals where bad things can happen....the concern here is what we're doing and where. Inside a panel that contains nothing but 120 V equipment all on the same phase, there's no arc flash hazard and even the shock hazard can frequently be managed. But in an industrial control panel with a bunch of IEC contactors where 120 V and 480 V is mixed together the hazard can be significantly different.

Thus this is a case where it screams for a risk assessment. Many PLC control cabinets as well as the low voltage controls for switchgear are fairly safe and as long as the activity is tightening up a wire on a de-energized contact it can easily be done safely. I would not carte blanche allow ANY situation though because of the possibility of the presence of other energized equipment/terminals in the vicinity which may be an issue as well as the possibility of causing equipment damage whether or not there is a safety hazard (as opposed to just poor maintenance practices).

So my answer would have to be "yes" to the very general question posed, but it would be "no" for many reasons that the practice is done in the first place, but it would change to "no" for many other practical situations which occur especially with low voltage power and controls, in the right conditions.

If protocol dictates safety regulations for permission for an electrician to safely tighten a loose power screw live on a 120Vac system, or power down an expensive chemical process. then the design should change with redundancy so avoid intermittent loose contacts when being tightened. If the environment is explosive gas, is another thing, but power quality being what it is, the system should survive this glitch if it occurs during the tightening of a screw. It is unlikely power interruption of tightening a loose screw is going to be any worse than a normal power quality interruption. Even if it was 200A service, I doubt there is a safety issue.

Mind you I do all my home electrical wiring live, because of my experience, and I have no problems servicing an old TV with 35kV, so don;'t take my advice.

What I meant to say is...it depends so the answer to the general question is "yes".

And if it is for retorquing and/or landing wires on energized circuits again the answer is "yes". Mostly because my answer is automatically...no, don't do it. At best it does nothing but it also weakens the system and creates faults in the future.

And if it has nearby conductors which the activities could cause faults such as from dropped tools and other similar "oops" moments then again the answer is "yes" to ensure that the risks are mitigated.

Then we get to cases where my answer is a qualified no.

When it comes to "landing wires" here I'm assuming we're talking about 120 V industrial/commercial/residential wiring, not on overhead power lines where landing jumpers is standard practice and in which there are procedures and equipment such as saddles designed specifically for this purpose.

The second case is landing wiring on de-energized conductors within an energized panel within the limited approach boundary (but not necessarily the restricted approach boundary). Since the distance requirement is "avoid contact" and there is usually little or no arc flash hazard, we just have to avoid the "secondary effects" such as dropped tools.

So there are clearly case-by-case examples where on-site "job safety analysis" practices apply (whether it's an EEWP or not), some equipment specific examples where large classes of equipment can be qualified via an off-line risk assessment can be done, whether or not it is called a "annual EEWP", and finally cases where it should not be done (so EEWP should stop the job).

Two different situations. The flyback transformer has a very low current. It is easy with a TV and hurts a lot to get "zapped" but I doubt you can get to fibrillation (not enough current). Current is the hazard, not voltage. Otherwise the millions of volts generated by a carpet in winter would cause us to remove carpeting from all homes in the U.S.

Second though, may I interest you in OSHA's interpretation of "continuous industrial process"?

https://www.osha.gov/pls/oshaweb/owadis ... p_id=25559

Specifically, the questioner posed this scenario: "The manufacturing of our products involves many discrete pieces of equipment whose individual processes are part of the overall manufacture of integrated circuit components. For example, we have ten pieces of manufacturing equipment fed out of a 480-volt three-phase panel. A new project requires that additional feeders and a 225-ampere circuit breaker be added to the panel to supply a new piece of equipment. To perform the work in a de-energized state, it requires the power to the panel must be disconnected and appropriate LOTO devices applied. This activity would result in the shutdown of the ten pieces of equipment, causing a significant interruption to our ability to manufacture integrated circuits...Is the panel considered part of a "continuous industrial process," thus allowing the work to be performed while the panel was energized using electrical safe work practices, as per Note 2 in §1910.333(a)(1)?"

Does this sound like a typical expensive chemical plant process?

And here is OSHA's response: "It appears that your panel is not part of a "continuous industrial process." The term "continuous industrial process" was derived from its use in the National Electrical Code (NEC). In the NEC "continuous industrial process" is used in the context of situations where the orderly shut down of integrated processes and equipment would introduce additional or increased hazards.1 Therefore, to qualify for the exception found in Note 2 of §1910.333(a)(1), the employer must, on a case-by-case basis, determine if the orderly shutdown of the related equipment (including the panel) and processes would introduce additional or increased hazards. If so, then the employer may perform the work using the electrical safe work practices found in §§1910.331-1910.335, including, but not limited to, insulated tools, shields, barrier, and personal protective equipment. If the orderly shutdown of the related equipment and processes would not introduce additional or increased hazards, but merely alter or interrupt production, then the de-energization of the equipment would be considered feasible, and the exception found in Note 2 of §1910.333(a)(1) would not apply. Based on the limited information you provided, it does not appear that de-energization of the panel in question would introduce additional or increased hazards."

In other words, "infeasible" does not mean "costs a lot of money". Loss of production is not an acceptable justification in OSHA's point of view. Instead it means "creates greater hazards". Examples of where this applies would be for instance interrupting power to fire pumps or hazardous atmosphere ventilation systems. Obviously we can all recognize that there is no practical case where it is just plain impossible to shut down a process no matter what it is unless there is some element of time involved (the hazard increases the longer we delay) so this clause has a very limited use.

More importantly, OSHA has defined the terminology to be functionally identical to the first clause in 70E and OSHA 1910.333(a)(1)...it's exactly the same as the greater hazard rule. Removing the word "infeasible" and/or reference to a "continuous industrial process" has no impact on the net result because the "greater hazard" rule still applies. It doesn't give an out to those who see the words "continuous industrial process" and say "aha, we're a big expensive chemical plant so that doesn't apply to us as long as we pencil whip the paperwork...heck our plant is so safe that we haven't released a big cloud of ammonia in a couple years now and we haven't seriously injured an electrician in almost a year...we're the safest company in the world". I'm paraphrasing what was stated several times at a large Fortune 100 chemical company in the U.S. and Canada but last I knew the policy still stands...we just claim "continuous industrial process" and pencil whip the EEWP's.

If they are troubleshooting a live circuit already I would think that they should have an EEWP in hand before they started.

From experience, no. If you are doing EEWP's routinely for all jobs, then you are clearly not taking steps to minimize energized work and the whole purpose of the EEWP procedure has been neutered. That situation comes up very rarely in for almost all industrial plants, including chemical plants with expensive processes that they don't like to shut down. Just have the production manager sign off on the EEWP along with the above OSHA Letter of Interpretation and an understanding that they are putting people in harm's way in violation of OSHA regulation as well as safety. You'd be amazed at the response when it is clear that you are having them sign a document that amounts to a prison sentence (wilful negligence) if anything goes wrong.

The general rule is not to work on equipment while energized except for doing diagnostic type activities that can't be done while de-energized (can't test voltage if the power is off). Once the problem is discovered then the procedure should be to de-energize first and then take corrective action. An EEWP should be the exception, not the rule. That is precisely why a permit is required rather than normal, routine work, something that only occurs on an infrequent basis. The EEWP procedure simply recognizes that in a few instances energized work is necessary.

There are some situations that I've come across over the years that don't have good answers:
1. PLC panels and MCC controls. I'd say that 99% of panels meet or are very close to meeting a state where none of the wiring is exposed so it's really just an arc flash argument. In terms of arc flash most of the wiring is simply not a hazard but there's almost no guidance out there on this. So somehow, some way every site has to basically write a special procedure for these cases only. At 120 V, the "arc gap" is a fraction of a millimeter so even dry leather gloves (not oil/grease soaked) provide sufficient protection. The problem is that these can't be "voltage tested" because the test protocol is to fill the glove with water, immerse it in water, and hi pot it...so only a glove that is impervious to water (rubber) can pass the test at any voltage. So knowing full well that clean, dry leather gloves alone are sufficient protection do we go crazy and require rubber gloves (Class 00) plus leather protectors which makes it virtually impossible to handle common #14-#18 control wiring and the tiny screws associated with it or ignore a safety Code that violates common sense?
2. "Utility" style equipment, especially overhead. This stuff is made to be operated and worked on while energized. That's why 1910.269 doesn't have an EEWP rule. Most of the problems that large plants run into are addressed by realizing that a lot of the equipment of concern is distribution equipment and should be treated as such. There are only about 3 differences in the regulations from a practical point of view but people fear what they don't understand. Furthermore, please tell me how an EEWP would apply to this guy's job without becoming routine:
https://www.youtube.com/watch?v=FrgUZNCkfFI
3. Fuses on HVAC condensers and light poles along highways. It has long been standing practice to use yanking out the fuses as a means of disconnect without anything resembling a "true" disconnect. This violates the heck out of OSHA LOTO rules in terms of providing a means of disconnect but I don't know how many years it will be before we get rid of the last of these. At least with medium voltage "fuse cutouts" I can claim that it's a disconnect/fuse holder but not so with many common 240/120V fuse holders used as disconnects.
4. Landing wires on the de-energized side of circuit breakers, especially lighting panels. At least here arc flash MIGHT be possible as well as shock hazards but there are almost no reported incidents of major injuries or fatalities from this activity that I'm aware of. It is a relatively safe task but definitely pushing limits in many ways. Perhaps ultimately the answer is not to try to fix what's there but to come up with improved designs for lighting panels that make them inherently safer to work on.