Hello,
Can someone help me model this MCC. I have attached the way I have done it on Easypower. I also have SKM if easier to do it on there. The MCC Size is 600A, The Breaker is Cuttler Hammer HMCP 50A. The wire size is 8 Awg @ 115'.

This is been debated a lot over the years. Some customers demand it. Others only label the line side of the MCC and not individual buckets. Most want you to decide for them.

My answer to your question is, I don't know. You have a lot going on in there. Multiple taps, multiple conductor sizes, multiple bus gaps, a transformer, overload, a drive and more. Not uncommon at all. I doubt if the IEEE lab tested every fault condition of every scenario and accounted for this in their formulas. I don't see how they possibly could have. I don't see this as much different from a control panel. As I've mentioned before, I don't think you can accurately model a control panel. There are too many variables. I've seen a fault inside a control panel that blasted the door open and the arc sustained for at least 10 seconds if not longer (caught on video) before the operator ran to open the upstream disconnect when the model said the fault should have been cleared very quickly by the upstream current limiting fuses. That particular panel was like your MCC bucket x 10 with what was going on inside it. Multiple taps, starters, relays, fuse blocks, a drive and more.

There are so many spots their for an arcing fault, who knows what the arcing fault current would actually be depending on where the fault occurred. Not what the software or the formulas say it could be but what it really would be in every scenario.

I'm making no recommendation here but one could choose to pick a point in that MCC, like the line side of the main, the load side of the main (if it has a main) or the line side of each MCP breaker or the load side of each MCP and fail it or the main or the upstream feeder and use the 2 second rule.

Or you could choose to say the MCP is clears the fault and label it that way. Or the 45A J fuses.

You could model every single one of those components too just like they are but, again, I'm not sure the formulas were meant and tested to be applied like that. Maybe Jim can weigh in on this one. He would know more than me about that.