I did some more fuse testing today, and I have given the graphs to Trumpy to post for me.
I tested the 15A fuselink (a 22mm dia x 58mm barrel fuse) for temperature rise, first at 15A, and then at 20A. I had a temperature probe snuggled up against the barrel of the fuse, this being inside a fuseholder of the correct type for the fuse. This completely shrouds the fuse in a plastic case.
At 15A, the temperature rose to 55 deg C and held there, from a starting temp of 20deg C. This is within the allowable limit of no more than 40 Deg C rise at rated current.
The eye-opener was the 20A test. The temperature quickly climbed past 100 deg C and levelled out at 135 deg C! No wonder it made the office smell of frying fuse. One can only wonder just how close the fan-cooled fuseboard was to starting an electrical fire.
For anyone wondering about the test set used for these tests, here is a quick description : we have a 10A variac which plugs into a 240V mains socket. This feeds a transformer-type 140A arc welder which acts as the current source. The fuse (or breaker - we test them too) is connected as a short-circuit across the output of the welder with some nice fat 25mm2 cables we made up. A clamp-type multimeter provides ameans to measure the current. This setup lets us generate test currents of a little over 100A for short periods, and 50-60A all day long. A test current of 20A for today's test required the variac to be turned up to about 55V into the primary of the welder.
