AndrewKGentile wrote:
But wouldn't the sustainability of the arc, and it's ability to migrate, be based on the voltage and current magnitudes?
No. The relationship between voltage and incident energy is very weak and mostly has to do with the "arcing current" calculation. The "arcing current" is largely due to the fact that there is a minimum arc voltage. In low voltage cases particularly below 250-300 VAC arcing time is very limited because the peak voltage is not very far above the critical flashover voltage needed to restrike the arc. The theory is pretty obvious of course but since we lack the ability to model the air temperature in the vicinity of the arc, it's largely a theoretical concept right now. It is not plasma. Air doesn't dissociate until it hits if I recall correctly around 6,000 K. While this exists in the core of the arc, the actual core is only a few millimeters across. What you see in photos and is misidentified as plasma that I've seen looks a lot to me like light reflecting/fluorescing off of smoke particles, or the smoke itself glowing. Steel for instance turns a light pink and becomes visible in darkness around 1,000 F.
Quote:
Is there is a realistic chance that an arc started in a 15A breaker bucket in section 6 of an MCC would propagate all the way back to the main circuit breaker in section 1?
Much more realistic if you have an MLO compartment (main lug only) and a couple buckets in section 1. The majority of MCC arc flashes I've seen particularly when someone attempts to insert a bucket onto a live bus end up setting off an arc flash at the bus stabs and travels to the bottom of the vertical bus then blowing out the door at the bottom.
Quote:
And of course this would have to happen pretty much instantly, before the main breaker trips. Why would the arc move at all?p/quote]
Simple magnetic propulsion, like a rail gun. So whatever we're dealing with probably is NOT plasma. That word gets thrown around but the conditions can't possibly be right to form that stable of a plasma. Sure plasma exists in the arc core but it's only a couple millimeters in diameter. What is seen is just hot air as far as I can tell. That being said, arc propagation has been researched a little and discussed in the forensic literature because after an electrical fire arc tracks (little shark fin looking marks) are left behind. JHU has published some arc flash testing for Navy work and found that the arcs travel at around 400-600 feet per second. Given a typical "large" low voltage breaker opens in say 3 cycles or 50 milliseconds and using as an average say 500 feet per second, the arc can successfully travel at least 25 feet, more than enough to do what you describe but the physics are totally against arc propagation because it is magnetically propelled in the opposite direction. The only way for something to happen as you describe is if the smoke and heated air move out and engulf a section that is "upstream" of the original arcing fault location. Within a section sure this is easy because outside of the buckets themselves everything is all open inside within a section. There are definitely openings particularly at the top and bottom too between sections but they're not really much different from switchgear openings. I'm a bit dubious on the idea of arc propagation from one section to another simply because I haven't seen anything published to suggest that this is possible, and the mechanism for it is rather extreme. Within a single section I'd agree 100% that it is not only possible but multiple reports including a couple incidents at plants I've worked at confirm this.
Quote:
And why would it move to the incoming main? Is the incoming main somehow more electrically attractive to an arc?
No, quite the opposite. Arc tracking moves away from the source so it's not going to be propelled along a bus bar back towards the main. So horizontal movement from section to section seems very unlikely except under extreme circumstances. But remember that the arc is quenching as the voltage passes through a current zero then restriking once it gets above the flashover voltage. Assuming that the previous arc heated the air in the vicinity, it lowers the flashover voltage substantially. Within a given enclosure arc propagation from single phase to three phase is well documented and this is the reason that it happens. And it happens very quickly...within 1 cycle. Magnetic arc propagation as I explained earlier is also very fast. Travelling a full 8 feet from top to bottom should take about 16 milliseconds or one cycle, well before all but the fastest circuit breakers trip. Then as heated air rises, I'll buy it "jumping" within a single section as the heated air first blows the doors off and then flows upward until the flashover votlage is depressed enough towards the top of an MCC section that the arc restrikes there rather than at the bottom. What I have a hard time imagining is this same heated air emanating out, particularly if we are to believe that most of the heating is dissipated as radiant heat (aka incident energy), in such quantities that it actually engulfs even the section right next to it.
