We have a halogen oven in our plant that is powered by 480volts. It is a pretty basic setup. The line feeding it is protected by a 125 amp breaker. All the legs going to the halogen lights are protected by fuses. There are 5 small circulating fans on this oven and each leg of each of the fans is fuse protected (15 fuses).
My problem is this:
Last night during normal operation the 3 3amp fuses on circulating fan #4 failed. They literally exploded and burned up. It tripped the 125 amp breaker feeding the oven. No other fuses or OL's were blown or tripped.

This morning I and another LE were investigating and we simply could not find the cause. No short circuits, nothing. We replaced the fuse blocks and the fuses and turned it back on and everything is running normal.
I am still puzzeled over why this happened. The fuses were Bussman Low peak LP-CC 3a 600v class CC time delay fuses with an A.I.R. of 20k.

I replaced them with fuses with Ferraz Shawmut "amp-trap" ATDR fuses with a rating of 200k.
I figured the rating was not high enough on the original fuses or they would not have exploded.

I know I did not go into a lot of detail here but Does anyone have any ideas what could have been the real problem ?

Also would it be unreasonable to expect the oven manufacturer to have put in fuses with a 200k rating rather then the 20 k.?

The LP-CC fuse has a 200k rating, not 20k. Are you sure that they exploded and didn't just burn up? A poor connection at the fuse clip or wire termination point will often cause the fuse to burn up.
don

I would tend to beleive there was an explosion because some of the controllers that were mounted through the front cover it blew out onto the floor. The fuses themselves are pretty much intact except for some of the metal burnt off.
You are right about the 200k. The 20k rating is for DC.
The wire termination connection was definatly less then ideal. I noticed on some of them the insulation was not stripped back far enough and some of it was under the clamp which is obviously a problem. However, that did not seem to be the culprit in this incident because there were no signs of overheating at the terminations themselves.
The only place where I saw signs of overheating were the fuse clips themselves. And when I pulled the fuses out of the clips the clip had no discoloration where it made contact with the fuse.
That to me seems like the heat was not generated due to a weak fuse clip.

If the load on the three fuses is reasonably balanced, a millivolt-drop test might be done on the presently-installed fuses and fuseholders

See ecmweb.com/ar/electric_testing_contact_quality/index.htm—"Testing Contact Quality on Live Motor Starters". Understand that this involves dealing with energized circuits, so may not be appropriate unless you are adequately qualified and trained in live-troubleshooting practices.

Designers are sometimes pressured to make control gear as small a possible, and in doing so may choose class-CC over class-J fuses. I don’t think that’s always the best idea.