Sizing cable for medium voltage starters (internal wiring)

[paulengr]

Having some trouble understanding the Code requirements for internal wiring on medium voltage equipment. More specifically....we recently had a failure so I'm checking up on it.

First, understand the jurisdiction a little. We are talking about a surface metal/non-metal mine. MSHA != OSHA. In fact underground and coal mining equipment has to be tested by a national testing lab but there is only one that can be used...MSHA themselves. And just to indicate how much of an agency rivalry there is, guess which national testing lab is NOT on OSHA's list of NRTL's? Yep, even though separate department of OSHA (MSHA) actually runs an NRTL, it is a required NRTL under MSHA but not even recognized as an NRTL under OSHA!

There are about 25 TOTAL surface non-coal regulations for electrical. I know because I copied them all into a Word document that I supply to vendors who are unfamiliar with MSHA when they ask about what regulations they need to comply with. The regulations are almost all very vague. For instance one regulation states that overcurrent protection is required but there are no details on how to implement this, and that's the point of my question.

Since there is nothing in the way of a clear direction on implementation of MSHA regulations for surface metal/non-metal mines, virtually all of them adopt the existing standards that would apply to an OSHA jurisdiction. That is, NEC, NFPA 70E, UL, etc., as a consensus safety standard. This is not required but it's either that or draft your own version of NEC. This is done even though NEC very clearly and specifically exempts mining equipment. However because it becomes more or less an engineering standard, mines are free to comply or to ignore for instance Listing requirements. This in itself is very necessary since a lot of mining equipment would be rightly Listed using MSHA's own national testing lab which is specifically not recognized by OSHA.

Now the question...we have a medium voltage starter assembly that a mining equipment manufacturer put together. As is typical with these things it contains a disconnect switch, a medium voltage starter, a second disconnect switch, and a 4160:480 transformer that is used to supply some auxiliary starters and a control power transformer for the industrial control system.

Just as with typical industrial control panels, the wire size inside the starter assembly is different from external wiring. The wiring seems to have failed from overloading recently after a couple years of service.

Now if it is a <=600 V industrial control panel then unless it is specifically tested and Labelled then it would generally fall under UL 508A. The wiring going to and from the panel would fall under NEC while the internal wiring would fall under UL 508A.

However there is that pesky voltage limitation. In medium voltage (5 kV rating in this case), externally again NEC would apply. NEC includes ratings for cable run in conduit, run in "free air", and run underground. What NEC doesn't include of course is cable run inside of panels. It appears that this would be regulated by UL 347. I don't have a copy of that standard but it looks a lot like UL 347 is more of a performance standard (the wiring is acceptable as long as it doesn't burn up when tested under 100% rated load). The question though is...is there an equivalent to the service and branch conductor tables contained within UL 508A for medium voltage equipment? Or are we stuck with simply using the NEC tables for "conduit" as this is the nearest equivalent? By the way, the manufacturer used the "free air" rating even when the wiring went through openings between cabinets and was all inside of a cabinet...a location where the definition of "free air" is tenuous at best.

 

[Jraef]

You have indeed stumbled across a "quirk" of testing standards. UL listing of MV gear is relatively new (early 1990s). Before that, there was only ANSI C37, but ANSI is not a testing authority. So UL 347 was orinigally created just to test against the ANSI design spec, that's why it looks more like a performance spec. To understand the ANSI C37 design spec, you also must start with what TYPE of gear it is, metal clad, metal enclosed, etc., although stand-alone starters usually use metal enclosed designs.

Either way however, I don't think ANSI design specs ever called out for cable ampacity ratings, most likely because they would have been considering bus bar in switchgear. So any cable runs inside were only relevant to UL from a performance standpoint. In fact, people are often shocked (no pun intended) to find that the cable inside of MV gear is usually not shielded either. When I worked for a MV starter mfr, we got red tagged several times on that, which always ended in the AHJ getting an education on this very subject.

That said, the UL test standard does include some very rigorous elevated temperature procedures, so if your supplier has listed that starter assembly, it's doubtful that the cables would be inadequate. The problem is, if they did not submit that SPECIFIC assembly exactly as built to UL, there is no "inferred" assembly listing procedure (as in UL508A) that would allow them to list it without re-testing. A common "cheat" by after market builders is to start with a listed assembly, disassemble it, and re-assemble it in a different format. This happens quite a bit in the low-profile mining "drag assembly" electrical gear industry. The OEM is not the starter mfr, it's a custom gear assembler. He buys a UL 347 listed starter, but tears it down and rebuilds in in his custom shortie enclosure. He is SUPPOSED to re-submit that to UL for re-testing, but since it is only used in mining, he submits it to MSHA instead, who likely sees the original (now invalid) UL listing and accepts it, but nobody re-tested the assembly under heat stress again.