It is currently Mon Aug 03, 2020 11:08 pm

Post new topic Reply to topic
Author Message
ekstra   ara
 Post subject: Does IEEE 1584 allow/consider internal separations?
PostPosted: Mon Feb 19, 2018 4:10 am 
User avatar

Joined: Mon Jan 15, 2018 3:55 am
Posts: 1
Location: Germany

I wonder if the IEEE 1584 is also considering internal separations?
For example when I have an MCC with protection against internal arcing (acc. to IEC/TR 61641 Ed.3) do I need to label all compartments with the highest incident energy from the main busbar or am I allowed to calculate the incident energy of each compartment individually?

 Profile Send private message  
Reply with quote  
 Post subject: Re: Does IEEE 1584 allow/consider internal separations?
PostPosted: Mon Feb 19, 2018 8:31 am 
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 2174
Location: North Carolina
IEEE 1584 does not give any guidance for this.

In practice on this forum and elsewhere it has been reported that between sections of switchgear no propagation has been seen. Within MCC's the situation is far different. Within a given section propagation definitely exists because I have seen reports and eye-witness accounts at least at two different sites. Both were triggered by removing and then reinserting the bucket onto a live MCC. So from this evidence it is clear that we need to model each section of the MCC as a single enclosure.

I have not seen any reports anywhere of section-to-section propagation in an MCC. Considering that there is almost no reason to ever work on the horizontal bus bars and those would be connected by splice plates not too much different from switchgear sections, I'm not surprised. Since arc propagation is powered by magnetic force and pushes the arc away from the power source, the only way to propagate "backwards" and into adjoining sections would be heating the air enough to cause the arc to "jump" to the lower impedance bus bar nearby. While this has been recorded across circuit breakers, I would have to believe that there is both a far shorter distance and there is a need for far greater containment of the gasses that would have to be channeled into the narrow opening behind the buckets and up all the way to the horizontal bus area. This is possible but it would seem far more likely that the hot gasses would simply expand out through the (now blown off) doors into the open space in front of the MCC. In the process of going up through such a channel the gasses would also be imparting energy onto the obstructions (heating the MCC) which would further cool and restrict movement in that direction.

So based on the evidence theoretically each section gets modelled separately. If there is a main breaker then obviously the incident energy in that entire section depends on the incoming bus and ignores any effect of the breaker. The next step depends on whether you go with an ultra-conservative approach and theorize without any evidence that propagation across sections is possible in which case model the entire MCC based on the incoming (upstream) connection, just like a panelboard. But absence any evidence that propagation section-to-section occurs then the model would treat each section separately. This means that except for the section containing the main breaker, the other sections are modelled with the main breaker as the upstream overcurrent protective device. Although one could model the slight impedance differences across individual splice plates and treat each MCC section separately, the difference is so small that in practice the rest of the MCC gets modelled as one enclosure.

A final consideration is whether or not to label the outgoing cables. In many cases the MCC is located in a relatively clean and dry environment. In this case it might make sense to model the loads including cable lengths and place the labels conveniently on the individual bucket doors. In almost all cases the incident energy will be under 1.2 cal/cm2 in the first place. Some plants just ASSUME this and don't bother with arc flash calculations at say a motor peckerhead, or assume that testing for absence of voltage will occur at the MCC where the wiring is easily accessible. This is obviously a bit of a stretch. It works most of the time. And there's the obvious issue that verifying voltage anywhere other than the worksite or within clearly visible range of the work site is flirting with disaster down the road but if the leads are all taped up properly in a peckerhead it's impossible to get access any way other than stabbing the probes through the insulation and hoping it is on something solid enough to get a voltage. With some THHN and XHHW-2 cables it is possible to do this through the cable itself by aiming the probe at a horizontal angle and sort of digging up under the insulation but obviously this does damage.

 Profile Send private message  
Reply with quote  
 Post subject: Re: Does IEEE 1584 allow/consider internal separations?
PostPosted: Mon Feb 19, 2018 9:25 am 
Sparks Level

Joined: Fri Apr 15, 2011 7:43 am
Posts: 177
Location: Colorado
I agree with Paul that the propagation issue is remote but I recently had a MCC with the main breaker in an open cubicle. By open I mean, once you open the door you were exposed to all of the bus bar that fed the rest of the MCC. If there were a fault the adjacent sections bus bar there could potentially be a secondary fault at the incoming side of the main breaker from molten particles filling both sections.

Our company takes the position that we are not in the business to determine the potential of propagation and will not take on that risk so for panels and MCC's we always use the upstream protection and not the main. 480V switchgear that is built like 5kV switchgear and has physical barriers (sheet steel) we can assume the propagation is minimal, if we do not know the construction we assume propagation may be possible.

If the upstream protection is set properly and designed properly, it will protect just as well as the main breaker with exception of clearing ground faults.

IEEE does not take that into consideration but here has been recent tests that show having isolation and distance in meter sockets affects the behavior of faults and propagation.

 Profile Send private message  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 3 posts ] 

All times are UTC - 7 hours

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
© 2019 Arcflash Forum / Brainfiller, Inc. | P.O. Box 12024 | Scottsdale, AZ 85267 USA | 800-874-8883