On 7/11/2019 7:04 PM, Jeff Liebermann wrote:
... the 20A that the variac is rated for resistive.

I would expect the fuse to be rated at MORE than 20A. Also, a 0.017
in thick spade lug is rather flimsy and will probably make a lousy
push on tab terminal. For 20A, it really should be thicker or the tab
terminal will arc, loosen, and/or fall apart. Tin is easily soldered.
I would therefore expect wires to be soldered to the tabs and not use
push on terminals. ...

It failed where the tabs were fixed to the body at right angles. It's
hard to tell exactly, but it looks like a spot weld & not much of one at
that. Pretty flakey & I'm glad to have it out of there.

My guess(tm) is if the fuse blew, the variac would be considered
totalled and require a factory replacement. That fuse was obviously
not intended for customer replacement. My conspiracy theory is that
someone forgot to design a fuse into the original device, and had the
variac winding blow up in the field. UL or the equivalent, probably
wouldn't like Ward Leonard modifying the design without having the
device recertified. Not wanting to go through that ordeal process
again, WL installed a fuse where UL wouldn't see it and said nothing
about the field failures.

That makes as much sense as anything.

Thanks again