This week’s question is about the infeasibility of establishing an electrically safe work condition. NFPA 70E addresses what is considered infeasibility with:

130.2 Electrically Safe Work Conditions (A)(2) Infeasibility. Energized work shall be permitted where the employer can demonstrate that the task to be performed is infeasible in a de-energized state due to equipment design or operational limitations. There is an informational note the describes example.

Here is this week’s question.

Does your company/client(s) have equipment where de-energizing is infeasible?

What is the situation?

Yes - some infeasible conditions.
No - infeasible conditions
Doesn’t apply

None of my customers run their entire facility on a UPS. This tells me they are able, if not willing, to accept an unexpected power interruption from the utility.

We have numerous very large battery banks in various voltages that can never be "De-energized" upstream from the breaker or disconnect.

Continuous process plant where starting and stopping the process imposes greater risk than working energized.

I posted a comment on Oct 25, 2019 about having numerous large 130 VDC battery banks that can't be de-energized upstream of the breaker or OCPD I would like to add a few thoughts for us to consider about work on batteries.

While true the batteries themselves can't be placed in an Electrically Safe Work Condition (ESWC) in any real practical ways nor is it feasible when performing PMs or testing of the batteries we must look at our DC systems in two or maybe 3 distinct parts. The source (batteries themselves), the charging circuit(s) and the distribution/load equipment downstream of the batteries is another subject all together.

We recently had a significant arc flash event in one of our 130 VDC MCCs where the task was to replace a shunt from the positive line bus section of the MCC. Our plant was built in the late 70's and early 80's and the battery bank is directly hardwired from the positive and negative posts of cell #1 and cell #60 to the positive and negative buses on the line side of the main breaker feeding the MCC. Being an older plant an intermediate disconnecting device upstream from the MCC main breaker was not installed which essentially means sections within the MCC will always remain energized. And as you would guess the shunt is physically located on the line side of the main breaker, meaning it is always energized.

The shunt is fairly large and is bolted in series to the positive bus through eight 1/2" bolts (4 on each side of the shunt) through a layered "sandwich" type connection. When all the bolts are removed, you can't just pull the shunt off the bus because after 30 plus years, the sandwich layers are still very tightly compressed together. Therefore the removal of the shunt requires some "gentle persuasion" through the use of a heavy rubber mallet or a plastic dead blow hammer to tap it loose. The only available angle of approach is to tap the shunt on the right side which will shift it to the left as it breaks free, as your facing the bus work. Unfortunately the negative line bus is only a few inches to the left of the shunt and the positive line bus. Understanding the close proximity of the energized positive and negative buses to one another we made a determination to cover the negative bus with PVC plastic roll insulation rated to 7500 volts (Salisbury Roll Blanket Class 1 PVC RLBPVC1-48) to insulate it and to provide some level of "guarding".

Well as Murphy would have it, when the shunt was slowly breaking free and being moved to the left through soft and controlled taps from the rubber mallet it unexpectedly broke free from the connection. This caused the lower left corner of the shunt (+) to make contact with the negative (-) bus which was wrapped in insulation roll blanket. While the blanket is more than adequate for the nominal voltage of the bus (130 VDC versus Class 1 insulation rating of 7500 VAC) it isn't designed to protect against physical damage from a sharp piece of metal. And this is exactly what happened. The sharp bottom left corner of the shunt which is still energized at positive (+) polarity cut through the insulation blanket and made direct contact with the negative (-) bus which obviously resulted in a short circuit. To say it was significant would be an understatement.

Thankfully the majority of the arc flash was contained within the wraps of the roll blanket and the PPEs the electrician was wearing protected him from serious injury.

Learned much from this event and from now on we will disconnect all the leads (2 parallel runs of 750 MCM wires) from the battery to fully isolate the MCC from all sources of energy. Additionally, we are also considering adding an intermediate disconnect safety switch up stream from the main breaker of the MCC so it can be quickly and safely be de-energized.

We learned a very valuable lesson and were very "lucky" during this incident. Specifically the fact we should look at every possible way to de-energize equipment prior to performing any work whenever possible through a robust risk hazard assessment. Sure it would have taken a little longer to disconnect the wires from the battery bank and there's some risks in doing so but compared to the risks of working two adjacent energized buses side by side within inches of one another was the far greater risk hazard to our employees.

I share this to help others learn from our experience.