Hello,
I have a Fire Pump Controller that is directly tapped of the main transformer (1500kVA, 12.47kV-480Y/277V) and as one would expect the arc flash hazard is 63 cal/cm2. This makes is impossible to do any work inside the controller as the circuit breaker is inside the controller. I suspect this may cause some issues in testing and maintaining the fire pump controller and fire suppression system.
I am sure other people have experienced this situation and would appreciate any insight on how you handled it.
Thank you

WBD:
You are right. If there is no OCPD ahead of the Fire Pump Controller there is not much you can do.

However, many times, when I consider the utility overcurrent protection at the primary of the utility transformer, the AFIE is reduced to levels below 40 cal/cm^2. Other times it doesn't. The only way to be sure is to perform the calculations.

So call the utility, get that information, run the arc flash study and see if the AFIE has been reduced.

Thank you. The results shown are with the utility fuses taken into account. The system setup is 100K fuse links, 700ft of utility cable, 1500kVA pad (12.47kV - 480Y/277V).

Even though it is not entirely my issue, I am concerned on how fire pump testing is done. From talking with an electrician he said motor currents are verified by clamp on readings taken from the motor leads inside the fire pump controller cabinet. Obviously this can't be continued and will probably have to be done in the motor peckerhead. I do not know what other readings need to be done in the fire pump controller for annual testing and certification.

I was thinking that a fused disconnect could be installed upstream but have not researched the code on that.

What you will find on fire pump controllers is that they are designed and installed so the motor will melt down before you clear a fault from an OCPD. The OCPD must carry locked rotor current indefinitely.

Have you considered mounting CTs in the controller cabinet and an external meter?

Fire pump controllers are considered slightly different from other equipment at least under NEC. Because the whole purpose is to fight a fire and it is considered emergency/essential, they are not required to have overcurrent protection but they are required to have short circuit protection.

Regardless, just measuring current doesn't fall under any standard for maintenance. At best it may be an attempt to deal with the special nature (no overcurrent protection) of the fire pump.

However, not working on the equipment is clearly at odds with 70E. The section in 130 requiring an arc flash risk assessment clearly states that improperly or poorly maintained equipment invalidates the arc flash study, so your "63 cal/cm^2" study would then be invalid altogether under any circumstance. Second, Article 205 specifically spells out the maintenance requirements required as a baseline minimum in order to be able to use the rest of 70E (Article 130 for instance). So NOT doing maintenance is far worse than dealing with arc flash and shock hazards because it significantly elevates the risk of a failure, often by a couple orders of magnitude of likelihood. 70E also contains several references to maintenance standards in Chapter 2. Of those, NETA MTS is pretty well written but unfortunately takes the "kitchen sink" approach because it is very self-serving (NETA is an organization of electrical testing companies who promote more maintenance, not less). NFPA 70B is much more reasonable but the version I have is poorly editted. Both are general maintenance standards and do not limit themselves to only the items required from a safety standpoint, and neither recognizes alternative or improve maintenance techniques very well. For instance, NETA MTS recommends using a torque wrench to check fastener tightness while in service and provides a table of standard torque values for initial tightening. This is clearly at odds with basic fastener engineering. It also recommends milliohmeter measurements, which is again clearly documented in multiple sources as not really providing any correlation with the thing that they are attempting to measure (detecting loose joints) and milliohmeter measurements have not been shown to provide any predictive power and may drift up or down with no relationship to the integrity of the joint itself. Fortunately NETA MTS provides for using the only known predictive maintenance technology for energized joints (IR scans). It also recommends taking apart EVERY joint and doing hi potting or meggering on everything to detect insulation failures. Neither test has been shown to predict anything except degree of contamination and again, neither has any predictive power whatsoever and at best only show current condition (as per IEEE Std. 4, the standard for these tests). And corona testing, a method which is both highly predictive and highly reliable means of detecting insulation problems in medium/high voltage equipment, is completely ignored.

I would have to disagree with that statement as it relates to this situation specifically. The only protective device upstream that affects the arc flash hazard on the Fire Pump Controller is the utility fuses as the Fire Pump controller is tied directly to secondary of transformer. I don't know what maintenance not being done could invalidate the results.

The question in the posting was related to what testing and maintenance is required for Fire Pump Controllers? I heard of one testing that was required by insurance company? which had the pump motor currents being verified by clamp on readings taken from inside the controller cabinet. This will be impossible now due to the arc flash hazard.

Lack of PM does not invalidate the arc flash hazard analysis. The NFPA 70E states "The arc flash hazard analysis shall take into consideration the design of the overcurrent protection device and its opening time, including its condition of maintenance."

IMHO maintenance requirements must meet the standards as PaulEng referenced AND the insurance company will have some more specific testing/maintenance requirements. After all, they will be out a bunch of money if the building burns down. I recently walked into a fire pump house and found the pump controller in the OFF position with an accumulation of dust and cobwebs on the handle operator. Ever wonder why some insurance companies require energy isolating devices on fire pump equipment be locked in the ON/OPEN position?

I forgot to add... have you thought about an arc flash suit with an ATPV rating >63cal/cm^2?

Just my opinion but I think the chance of causing an arc flash while walking by, opening the enclosure door, and assuming you are not physically pushing wires around to do it, applying a clamp on current meter should be the same. I might be persuaded to feel differently about closing doors if wiring is installed improperly and there is a risk of pinching a wire. But with no exposed conductors involved, I have a hard time seeing how an arc flash could occur in this scenario anyway except spontaneously.

Two facts in this thread are mutually exclusive.

First, When the AFIE at the fire controller panel is 40 cal/cm2 or greater you can't work on this panel even though you have PPE greater than 40 cal/cm2. The blast is too dangerous.

Second, you need to maintain the panel while energized.

Any solutions?

70E states that greater caution is required above 40 cal, not that it is "too dangerous". The arc flash hazard definition is also pretty clear that a more likely risk exists under certain circumstances, that of interacting with equipment in such a way as to cause an arc flash. It refers to the tables which give a rating of 0 ("no" in 2015 1st draft) for many activities including opening doors. There is zero substantiation for the 40 cal cutoff. There is also a requirement to do proper PM's. Thus it would seem in putting this together that doing a required PM that does not involve an interaction greater than opening a door, doing a thorough visual inspection, taking readings with a clamp on meter on insulated wire, and closing the door, passes the test for not interacting with the equipment in a manner that could cause an arc flash unless visual observation detects bad connections, bad insulation, or excessively loose wiring.

How do you visually determine a loose wire, if there is no signs of overheating? Indeed if a wire is excessively loose it is possible to jar it free trying to put a clamp on meter around it. Therefore I would contend that one is interacting in such a way as to cause an arc flash.

This does not reduce the AFIE.

Depending on how the cover is installed, it does make the area finger-safe, but the cover in and of itself may not appreciably reduce the liklihood of an arc flash incident.

While the probability of an arc flash incident is not high, we shouldn't allow ourselves to be lulled into a false sense of security based on making something touch-safe with lexan covers. If something does get behind the cover, or something becomes loose behind the cover and there is an arc flash incident, that lexan cover is likely to become a projectile or shrapnel.

Agreed, however NFPA 70E only applies to exposed live parts. The Lexan covers are primarily to provide shock protection.

That is why I started with "Depending on how the cover is installed" I have seen panels that use lexan covers where the "cover" is essentially a piece of flat lexan shield mounted on four standoffs situated in front of 460-volt connections or a piece of lexan placed in front of the primary side of a transformer. They certainly make the liklihood of accidental contact with the voltage more remote (from a straight-on approach, at least), but more often than not, they make a convenient trap-all for anything dropped above the upper edge of the sheet of lexan. I have fished out plenty of tools (or had to remove the "cover" to retrieve plenty of tools) that dropped between the lexan and the parts of the circuit that would have been live - had the power been on.

Terawatt... NFPA 70E applies to Electrical Safety in the Work Place and, if you read the definition of an arc flash (2012 Edition) if clearly explains the arc flash does not depend on being exposed.

Being exposed increases the risk of an incident. If it is finger safe, then the risk would be reduced, not totally eliminated.

As for the fire pump controller having loose wires... very true however, I would turn the controller off ( motor not running... handle to OFF position) verify absence of voltage on the load side of the switch, put the amp clamp on, then turn the controller back on (while standing to the side wearing an appropriate arc flash suit (>63cal/cm^2). Test the motor, turn it off in an orderly manner, place switch in the OFF position, test for voltage, remove clamps, close door, place switch to the ON position (again while standing to the side wearing arc flash suit).

Also, do not forget that while the arc flash suit protects you from getting burned, you are still exposed to arc blast. The suit will provide minimal protection from a 5 psi blast throwing you across the room and into a wall.

This was sort of a holdover. No longer true as of 2012 edition.

If properly installed and maintained, risk of loose wires is generally around 10^-12 and that is the assumed value in for instance IEC 61508 and 61511. Arcing faults in general under same assumptions are occurring at a rate of around 10^-6/year based on ESFI.org report. Although in an absolute sense just walking by could cause an arc flash if nothing else via the butterfly effect, it would seem that the primary cause resulting in a fatality is unlikely to be a loose wire which was not visible during the PM, again assuming proper installation and maintenance.