Kerry Wong Demonstrates a Process for Reliable Non-Destructive Testing of Single-Use Fuses
With a constant current source, a multimeter, and some clever maths, Wong's experiment shows that ratings aren't always to be trusted.
Kerry Wong has published the results of an experiment into non-destructive testing of single-use fuses — getting to within 10-20 percent of current required to cause a fuse to fail without damaging it in any way.
"A fuse is a sacrificial device for the protection the rest of the circuit in an over-current situation. There are situations we need to know with some reasonable confidence what the actual maximum current is for a given fuse without destroying it," Wong explains. "This is beneficial when the specifications for the fuse at hand is unknown (e.g. with cheap no-name brand fuse), or the circuitry to be protected is so critical that the interrupt current level must be specified precisely.
"This non-destructive testing of fuses can be accomplished by analysing the properties of the fuse material. In practice, all fuses are made of metal alloys with positive temperature coefficients. Although we do not know the actual composition of a given fuse material and thus do not know the exact temperature coefficient or the melting point precisely, we can use [a mathematical equation] to calculate the maximum current rating of a fuse with reasonable accuracy as when the current approaches the maximum, the thermal run-away causes the fuse resistance to rise exponentially and thus once the resistance doubles, it only takes very little current increment for the fuse resistance triple and eventually causes the fuse to fail."
Wong has found that the majority of fuses melt at between 250 and 750 degrees Celsius, and that when the measured resistance has doubled it's within 10 to 20 percent of the current required to make the fuse fail - providing a means of testing a fuse's suitability without actually making it blow.
"For the experiment setup, I used a high current power supply in conjunction with an electronic load to provide the constant current," Wong writes. "And a multimeter is used to monitor the voltage drop across the fuse."
The results are surprising: A generic glass fuse rated at 1A remained intact up to 1.3A, though Wong notes that "the fuse was glowing red hot and would [most likely] fail at that current level for prolonged period of time," while a 0.1A-rated fuse remained intact long after it should have blown. "The fuse’s maximum current limit is almost ten times that of the rated current," Wong notes. "This is another reason why this kind of non-destructive testing is useful as this fuse would not be useful at all for a circuit that requires a 0.1A fuse."
Wong's full write-up is available on his blog.