VFD Output Clearing Time for Output Side Fault

[ThePunisher]

I am currently conducting an arc flash study and it also includes making arc flash labels associated with 4.16 kV isolation switches at the OUTPUT of VFDs.

I understand that when a fault occurs at the switch terminations (while switch is closed), the fault contributions would come from the VFD output (which would be limited to a maximum of 150~200% VFD FLA) and motor contribution (approx. Locked Rotor Current).

I would like to ask what would be a typical VFD clearing time or shutdown time when a short circuit occurs at the closed switch's terminals. I have difficulty getting this estimated from the VFD manual. I would think it should be as fast as a fuse as the VFD may have recognize a short circuit from an overload due to sudden voltage dip.

The VFD I am modeling is an ABB ACS1000i rated for 2,250 HP, 13.8-4.16 KV

Thank you to anyone who can support me.

 

[jraef]

Transistors can turn off instantaneously but fault clearing time is basically detection time and any firing board hardware scan time it takes to act to turn them off and thus is totally subject to the design of the drive. It's not something a VFD mfr will publish because it potentially exposes weaknesses to competitors. ABB should be able to tell you, but I doubt anyone here will be able to provide you with anything more than educated guesses. My educated guess is somewhere in the 15-20 msec range: faster than a fuse (but from experience still not fast enough to protect the transistors from every output fault). Bottom line in basically every case I have seen, the weak links are the transistors, not the down stream components.

But just so you know, the VFD fault contribution is likely higher than the 150-200% value. That value is closer to a "long time" trip setting in software so is for something like 60 seconds or 30 seconds. For a fault condition, the typical design will use the hardware overrides built into the firing boards that doesn't wait for the microprocessor to see the fault. That value is typically published in the design specs an as instantaneous current trip point, generally in the 300% range.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[ThePunisher]

Thanks Jraef for your prompt response. I already sent a query to ABB.

I am intending to check the schematics. Chances are the detection logic may also include an output trip to the upstream contactor. The input short circuit protection is a fuse and thus the only back trip we can make is from the VFD controls to the input contactor.

 

[ThePunisher]

I am thinking, if I do not get a response from ABB soon, I would use 50 msec (instantaneous VFD P/U) + 133 msec (contactor opening time) which will get me to approximately 200 msec