Arc Flash - Take it seriously

[megloff11x]

Regarding Arc Flash evaluation, I can't get a number for bolted fault current from the power company so I try and assume the worst. I've been using the type RK1 fuses to minimize the hazard, but if anyone has any other mitigating and minimizing solutions, post them here.

Matt

 

[dlhoule]

I don't think this is the correct message. The message should be don't work it hot. The cases where OSHA and NFPA 70E permit hot work are rare. Of course, until you have proved it is locked out and dead, you must have the correct PPE...even just to check for voltage.
Don

I'll 2nd Don on this: do not work it hot.

 

[kingpb]

Regarding Arc Flash evaluation, I can't get a number for bolted fault current from the power company so I try and assume the worst. I've been using the type RK1 fuses to minimize the hazard, but if anyone has any other mitigating and minimizing solutions, post them here.

Matt

You can't simply assume the worst case fault level because Arc Flash is Current and Time dependent. Assuming a higher current may cause the calculation to decrease the AF level because it may make the fuse appear to clear the energy quicker, thereby decreasing the overall level.

To be safe, and still work hot, you really do need to know the info.

 

[macmikeman]

I do not want to start any arguement here, and you will have to forgive me as I am only your standard run of the mill electrician, I don't teach any seminars and so forth like alot of my friends here do. On this very site I have watched several video examples of high voltage ark faults that went on for quite a few minutes at poco substations. The incident energy was by simple observation, quite a bit more shall we say energetic than what usually comes from a 480 volt bolted short circuit. I say usually because I do know all about 480 volt energy displacement, have seen enough to know it isn't worth it to not have the proper gear on when working on it live. There was a 480 volt transformer bank explosion at an auto assembly plant a few years back that killed at least two men. I am all for protection when it comes to real danger. Don is right. It should not even be energized period. But anyway, theory is nice and all, but just go watch the poco video's and then tell me about how 13,000 volts isn't packing the same punch as 480 volts.

 

[kingpb]

Unfortunately, a video will not be able to tell you the temperatures that are reached, or the energy released. By the naked eye, it is unlikely that a any person would be able to observe the difference between heat generated capable of causing a recoverable burn, and an unrecoverable burn.

Here is an example of the difference in LV and MV values:

A main breaker is fed with rated current of 1200 A @ 13.2kV. The available short circuit current is 15,000 A. The MV breaker is capable of operating in 5 cycles (0.083 secs). A 50/51 relay is used.

Now step the voltage down to 480V through a 3.5 MVA transformer with a 5.75% impedance. The 480 V switchgear main is rated for 4000A, the available short circuit is 69,554A, the breakers are capable of operating in 7 cycles, (0.117 sec).

In this situation, with relays coordinated and properly protecting equipment:

Fault on 13.2KV bus produces 3.66 cal/cm^2 @ 18", and has flash protection boundary of 2.62ft. This is Cat. 1 PPE.

Fault on 480V bus produces 375 cal/cm^2 @ 18", and has flash protection boundary of 26.52ft. This is ABOVE Cat. 4 PPE.

Hopefully this is a clear illustration as to why higher voltage does not mean higher Arc Flash.

 

[jim dungar]

Regarding Arc Flash evaluation, I can't get a number for bolted fault current from the power company so I try and assume the worst. I've been using the type RK1 fuses to minimize the hazard, but if anyone has any other mitigating and minimizing solutions, post them here.

Matt

Do not fall into the propaganda that RK1 fuses are always the answer to arc flash. Large RK1 fuses do not limit the fault for many non-service locations. A 400A fuse needs about 4kA of bolted fault current and 1200A ones need almost 14kA before they start to be effective (i.e. <40CAL/cm^2).

 

[MJJBEE]

You can't simply assume the worst case fault level because Arc Flash is Current and Time dependent. Assuming a higher current may cause the calculation to decrease the AF level because it may make the fuse appear to clear the energy quicker, thereby decreasing the overall level.

To be safe, and still work hot, you really do need to know the info.

Here comes the real problem with arc flash. I work for a utility and can't tell you a value at your service at all times. It changes because we feed services from diffrent sources we also get diffrent availble powers on the transmision system due to plants going on and off line.

 

[jim dungar]

Here comes the real problem with arc flash. I work for a utility and can't tell you a value at your service at all times. It changes because we feed services from diffrent sources we also get diffrent availble powers on the transmision system due to plants going on and off line.

If you can not predict a range of available fault currents on your system, how do you handle the coordination of your own protective relays and fuses?

 

[MJJBEE]

If you can not predict a range of available fault currents on your system, how do you handle the coordination of your own protective relays and fuses?

I can give you a range just not a value it will be all the time. However we are reluctant to give out anything but the worst case value that includes future upgrades. To prevent people from undersizing their equipment

 

[jim dungar]

I can give you a range just not a value it will be all the time. However we are reluctant to give out anything but the worst case value that includes future upgrades. To prevent people from undersizing their equipment

So, provide me with your three values and let me calculate the highest arc flash incident energy for each fault level. Then I will be able to determine the "worst case" PPE regardless what your system is "now". And of course all caluclations need to be re-done whenever there is a chnage on your system or in the customers facility.

 

[Onitram]

Danger

Danger

As employees work on or near energized equipment, movement near or contact with conductors or circuits, such as dropping a part or tool, may cause a phase-to-ground or a phase-to-phase fault. Also, an equipment failure that could produce a spark or arc could trigger an arcing fault. The resulting arc establishes plasma, which has a much higher degree of electrical conductivity than air.

Hazardous flash can occur in any electrical device, regardless of voltage, in which energy is high enough to sustain an arc. This includes many 440V motor control centers, panel boards, and switch boards. An arc of this type, while typically less than a second in duration, can reach temperatures in excess of 14,000deg.F--high enough to produce extensive first degree burns, permanent blindness, or death.

NFPA 70E specifies protective equipment including flash resistant clothing and face shielding. All personnel within the defined boundaries must wear specified protective equipment.

Onitram in chicago

 

[zog]

Onitram,

An arc can be up to 36,000 F, and can cause 3rd degree burns. A 1st degree burn is like a sunburn, not to serious.

 

[kingpb]

Here comes the real problem with arc flash. I work for a utility and can't tell you a value at your service at all times. It changes because we feed services from diffrent sources we also get diffrent availble powers on the transmision system due to plants going on and off line.

The service provided by the utility, at some point, will go through a transformer to the end user. The transformer may be owned by the utility or the customer, doesn't matter. In almost every instance the short circuit MVA will be so large, it will appear as an infinite bus to the Owner's system. That means the fault current will be limited by the stepdown transformer; and that is the point where the customer can start his analysis.

 

[don_resqcapt19]

In almost every instance the short circuit MVA will be so large, it will appear as an infinite bus to the Owner's system. That means the fault current will be limited by the stepdown transformer; and that is the point where the customer can start his analysis.

I don't think so. That would imply an infinite primary and that is not the case. A weak primary will limit the secondary fault current.
Don

 

[andy s.]

At a Facility where I work at, I have seen the results of an arc flash! Luckily for myself and co-worker we were not injured. During a shutdown (2 week) My co-worker and I were doing PM's on all electrical cabinets, contactors, switches, etc,... My co-worker re-engaged a 400 Amp disconnect after his PM. When it was time to energize the equipment, the disconnect exploded! Blew out of it's busbar! This knocked out a 1200 amp feed. Thank goodness that we were not near this device. I investigated the cause of this problem. The fuses to the disconnect were still good while the MCC 1200 amp bucket was tripped! I looked into the disconnect and seen that the stabs were burned, and the inside door of the disconnect was charred as well.

The root cause, in my opinion, was...The disconnect did not fully closed, but only was touching. Albeit, the smaller devices that was fed from this disco wasn't enough to draw the load, went unaffected until the motor was started.

I have discussed this with my colleagues much to their chagrin, for they are also Engineers but without field experience, that I believe they all should be at least a journeyman before their Degree (but that's a different topic)

In closing, If you have anyone working with 277 and above, please ensure that they fully understand the dangers of an ARC FLASH!!!!

A. S. (pending P.E.) 25 years later LOL

 

[JeffD]

To those who were wondering what NFPA 70E says about working hot. I don't know how to quote. This is for over 50 volts.

110.8(B)(2) - Written permission is needed.
130.1 - The employer needs to "demonstrate that deenergizing introduces additional or increased hazards or is infeasible due to equipment design or operational limitations."
130.1(A)(2) - Elements of a hot work permit
(1) - Description of ckt and equipment to be worked on
(2) - Justification of work and why it must be worked hot
(3) - description of safe practices to be employed
(4) - results of shock hazard analysis
(5) - determination of shock boundries
(6) - results of flash hazard analysis
(7) - the flash protection boundry
(8) - personal protective equipment needed
(9) - how you are going to limit access to the area from unqualified individuals
(10) - evidence of completion of a job briefing including addressing hazards
(11) - energized work approval with signature from personal responsible - management, safety officer, owner, etc.

 

[kingpb]

I don't think so. That would imply an infinite primary and that is not the case. A weak primary will limit the secondary fault current.
Don

First, let me point out that I said in almost every instance. I like to refrain from using, always and never.

Second, to make sure we are all on the same understanding as far as what infinite bus means:

An infinite bus is an ideal voltage source that can maintain constant voltage magnitude, constant phase, and constant frequency, regardless of the load connected to it.

It is fairly reasonable to assume that a 2500kVA transformer connected to a utility system that is generating 1000's of MVA will not alter the voltage or frequency of the system, regardless of what load is connected to that transformer.

Certainly, if that 2500KVA transformer is fed from an overloaded substation or line, then you could conceivably draw the voltage down temporarily during an event, such as large motor starting or fault. But the MVA capacity of the substation transformer, lets say its 20MVA, will still be capable of providing enough short circuit MVA (347MVA) such that the 2500KVA transformer (43.5MVA)will be the limiting factor. Therefore, the fault current on the low voltage terminals of the customers stepdown service transformer can be viewed as an infinite bus source for calculation purposes.

As far as a weak primary, lets say the voltage is as low as it can be at 95% of rated (really bad!) So, instead of 13.2kV, on the distribution line, it's at 12.54kV. The rated short circuit MVA of the 2500KVA transformer does not change, it will still allow 43.5MVA of short circuit contribution through it, and the system ahead of it will still be providing more then it can pass, so the low fault current on the low voltage terminals of the customers stepdown transformer can still be viewed as an infinite bus source.

Don, If you meant something different by a weak primary, perhaps you could explain further.

 

[don_resqcapt19]

Don, If you meant something different by a weak primary, perhaps you could explain further.

We are out in the sticks and at some distance from the substations. And high current events do drop the primary voltage. I expected that this drop in voltage would limit the primary current and in turn limit the secondary current. Are you saying that we almost never need to know information about the utility primary to do our arc flash calculations? If that is true, all we need to know is the service supply transformer KVA and impedance. This does not match up with the information that I have read on this subject from other sources, but I am not an engineer and do not fully understand all of the details of the arc flash calculations.
Don

 

[kingpb]

You have posed a complex situation where multiple things are happening, I will try to break into pieces.

Say your substation has a low voltage condition, for whatever reason, then you are correct, in that the low voltage on the primary will mean a low corresponding voltage on the secondary. The voltage will be further decreased during this large current draw, due to additional current in the line; due to voltage drop being dependent on current. At some point we have to assume that the system stabilizes under this condition and the substation can provide an acceptable level of voltage support.

But, the load (secondary of the transformer) will draw more current, and the transformer is still going to deliver the higher current to the load. Transformers are a strange beast because they can be overloaded for quite some time, before heating takes a toll on the longevity. (some sacrifice in voltage regulation does occur). Under steady state conditions (i.e. loads already running) as the voltage goes down, the current will go up, because the load is based on KVA. Remember the rating of the load cannot change, it is fixed, therefore the current and voltage are inversely proportional. What may occur at some point, operating the load at the higher current for a long time may cause the longtime protection to trip. For motors, that would be the overload device(s).

Now, that was steady state operation, prior to the fault. At the time of the fault, the lower pre-fault voltage on the primary line, will cause less fault current to be able to flow to the primary side of the transformer. (less pressure to force it down the pipe) This is analogous to motor starting, when starting at a lower voltage means lower current draw (i.e. soft starter) because during starting the motor impedance is predominantly reactance. However, because of the inductance in the transformer, the transformer is already limiting the fault current to the secondary side. The amount that the transformer is limiting the secondary fault to, is in most cases, less then what would be possible even if the primary line was at a normal voltage level.

So, in general, knowing the service transformer size and impedance should suffice to get started. In all cases, you should ask for the utility information, if you can get it that would be a plus. Certainly, if anyone suspects unusual circumstances, then the more information you have the better. But utilities usually will not perform a system study for each service installed (costs to much). What you may be able to get is the system planning information, but that probably won't do you any good, because it is based on a lot of system growth over 5 or 10 years, and I think you would find that the infinite bus would still be were you end up.

Obviously, if your building a substation, distribution system, or putting in a very large service, they are going to provide more information, but in those instances that would be the engineer's job to determine for you.

Keep in mind that the cable has not been mentioned in what we have talked about. Cable has impedance, so just like a transformer, it will choke the fault current down. Therefore, it is important to include it in your calc. Cable has more of an impact on decreasing fault current then people realize.

Sorry for being so "windy" but I wanted to try and explain it right.

 

[CarRamrod]

Those news articles said the employees were "shocked". Does the reporter just not know what he's talking about? It also seems amazing to me that these guys were working live on a building that's not even operational. I don't think we're talking "penalties", I think we're talking jail time. That's pretty negligent.

The absolute worst place to be working live is on the secondary side of a transformer, upstream of a breaker. If there is a fault there the protection, if any, will be on a higher voltage level. The fault current seen there will be much lower than it is and it will take longer for the breaker to trip.

That's another big reason why working on < 600 V is especially dangerous. Working on the line side of a 600 V breaker is suicidal.

My two cents.