Does a fuse protect the cable or the device?
A fuse primarily protects the cable, not the device. Its job is to prevent the conductor from reaching burning temperature in a short circuit or overload. That is why the fuse current is chosen below the carrying capacity of the smallest cross-section on the line and above the device's normal continuous current. Protecting the device's internal circuits is a separate job and is generally handled by the device's own internal protection.
The critical distinction is this: a semiconductor fails within microseconds, whereas a fuse needs far longer to melt. That is why a fuse cannot save the device's internal circuit. What it does save you from is the current flowing into a faulty device heating the cable and starting a fire.
The selection logic follows from that job:
- Lower limit: it must be greater than the device's normal continuous current, otherwise it blows at every start. Also allow for the inrush current drawn at start-up.
- Upper limit: it must stay below the carrying capacity of the smallest cross-section on the line. If the cross-section changes along the line, the thinnest point decides.
A suitable standard rating is chosen between these two limits. ATO/ATC blade fuses are produced in the 1-40 A range and are distinguished by their colours; for example red 10 A, blue 15 A, yellow 20 A, green 30 A.
Practical consequences:
- If the fuse blows, fitting a bigger one removes the very thing being protected; the previous rating had been chosen to suit the cable.
- Each separate branch gets its own fuse; thin branches fed from a single large fuse are unprotected.
- The fuse is placed at the point closest to the source the current comes from.
Full coverage of the topic: Installation, redundancy and field continuity