500A fuse burned in half. Any thoughts on pix?

[rawelk]

The pictured Bussman FLS-R 500A fuse was installed in a Square D I-Line bus plug-in disconnect feeding an ABB DCS401.0500 drive (rated 408 AC amps/phase) driving a Reliance "Super RPM" 250 HP, 500V/400 FLA motor in a plastics extrusion screw application. The drive itself is protected with fast-clearing semiconductor fuses, and they were not affected.

I was not involved in the fuse replacement, and don't know which pole it was installed in, or it's specific orientation (although it is installed along the horizontal axis, rather than vertically).

For approximately two weeks before the fuse burned in half the drive would shut down approximately every two or three days. Unfortunately, instead of checking fault code(s) after shutdown the drive was power-cycled to clear the fault, and diagnostic information was lost. Since fuse replacement these "mystery" shutdowns appear to have ceased.

My best guess based on heavy oxidation on only half of the links is this fuse was leaking filler material, exposing progressively greater fuse element surface area to the air, and these oxidized elements developed higher resistance. Over time this process avalanched, and eventually the fuse dropped enough power for it to burn apart.

http://files.engineering.com/getfil...c7e-4441-93ff-abf2e1cfe757&file=fuseview1.jpg

http://files.engineering.com/getfil...a75-4ac2-b15c-6cceced86764&file=fuseview2.jpg

Does that scenario seem likely? Has anyone else seen a fuse failure like this?

 

[itsmoked]

That sounds like a reasonable theory but you need to back up further. I suspect that fuse was truly cooking prior. Likely that is why the fuse may have been "leaking". The clips could've been loose and oxidizing => higher resistance. Same with the cable connections. Toasty critters. The fuse probably went from heatstroke rather than any form of overload.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[controlsdude]

Fault Codes on VFD
Every type and manufacturer of drives that I have used has a list of history faults on the drive. Powering down the drive does not clear these faults. Usually there are at least 5 faults in the history section of the drive for display.

 

[Zogzog]

What is the available fault current at that point of the system, a large motor like that can contribute quite a bit to the fault current.

If your fault current exceeds the interuption rating of your fuse it will end up looking just like the picture you posted.

Have you tested the cable from the bus plug to the VFD? Since the VFD fuses were not affected I would guess the cable faulted, the motor contributed to the fault and the fuse interuption rating was exceeded.

 

[Skogsgurra]

Not possible for a drive to contribute significantly to SC. Not even if in regen, which is never the case if load is an extruder.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[dpc]

I would send the photos and fuses to Bussman and ask for an explanation.

My first GUESS would be the same as Keith's - the fuses overheated due to high contact resistance, either external or internal.

 

[LionelHutz]

Side pics showing where the failure occurred would help. Was that at the top of the fuse or the bottom?

It could have been a bad fuse. I've seen a burn out failure like that before with a few different medium voltage fuses.

Also, not familiar with that particular fuse. Does it have a minimum melt current? Some fuses do not provide full-range protection.

 

[roydm]

It looks fairly obvious that it was a loose connection, Notice how discoulored one end is whereas the other end and fuse elements are OK. I imagine there was also damage to the holder.
Roy

 

[rawelk]

Thanks for all of the replies. I'm going to be in the middle of an installation project for the next two days, and won't have the opportunity to post more pictures, but wanted to update on several items I've learned today about the fuse failure.

In response to Controldude's observation ... checked the manual, and they list the 'faultlogger' function as non-volatile, and 16 items "deep". Forgot to look at the log today (the tech mentioned he received a "Mains Undervoltage" fault, so that should at least be in the log), but did survey it about a week before the fuse failure, and all 16 items were '0' (no fault).

The failed fuse had been mounted in the upper position of the bus switch.

All three fuses were replaced with FLS-R 500A fuses.

However, a Bussman NOS-600 (!) single element, 10 KAIC, class H fuse was originally in the center position, and another FLSR 500A was in the bottom position.

I was wrong about the FLSR fuse being manufactured by Bussman. The failed fuse label was mostly destroyed, but checked the other one, and learned it is actually a Littelfuse part number for an RK5 dual-element time delay fuse, 600V, 200 KAIC.

http://www.littelfuse.com/part/FLSR500.html

Maybe I'm looking in the wrong place, but can't find the characteristic curves for it, and gave up after about 5 minutes. Everything seems to point back to the above cut sheet.

The Square D plug-in disconnect uses spring-tensioned fuse holder fingers (rather than a bolted connection; guess they start using them at 600A and higher), and the technician related all three replacement fuses required considerable force to push in.

This suggests inadequate force on the fuse tabs didn't play a role in the original failure. Also, while there is some heat discoloration on the failed fuse tabs it appears to have migrated from the overheated fuse body, rather than being generated within the holder fingers. Nevertheless, it won't hurt to shoot it with an IR camera next time time allows.

I'm having a hard time with my original failure theory. The notion of heavily oxidized elements being due to loss of filler doesn't sound likely unless the fuse was in a 'bolted' orientation (which puts the heavily oxidized element towards the top). Plugging it into a spring clamp holder throws orientation out 90°. Also, it uses rather large grain silica sand (dissected the other one salvaged from that night) and I'd have a hard time believing it was prone to leakage.

On the other hand, the mystery NOS 600 fuse was leaking finely powered borax like crazy.

Tending towards believing it was an internal high resistance connection where the corroded set of element foils connected to the center piece.

In a couple of weeks I'm going to be doing a couple of changes to this drive, and will inspect everything with the proverbial fine-tooth comb.

Backstory. The DCS401 was installed maybe 4 years ago for emergency repair after the original Reliance MaxPak failed in an uneconomical-to-fix fashion, and was stuffed in the existing Reliance enclosure somewhat inelegantly. The line reactor sits on the bottom, and (no room for holders) we bolted the semiconductor fuses to it. Ugly, but it has been running fine.

I'll be mounting the reactor in an external enclosure, installing fuse holder for the semi fuses, and doing a couple of other odds and ends.