Long live the fuse?

Almost everybody knows what a fuse is - what one looks like and what it does. They've been a crucial part of electrical protection in one form or another for about 150 years. Every 13~amp plug has one and older consumer units are filled with them. Almost every piece of electrical equipment includes at least one fuse somewhere. Part of their appeal is their simplicity: they are nothing more than short lengths of wire, designed to get hot and melt when a fault occurs. They can be easily manufactured in quantities of a million or more for pennies each.

So why did I include that ominous question mark in the title? Am I trying to suggest that the time has come for these humble devices to be retired? Surely I shouldn't be too dismissive of a piece of technology that has achieved such longevity while remaining essentially unchanged?

To an extent, their use -- at least in proportion to the number of other alternatives -- has been declining for years. Fuses might be cheap, but they're not very flexible and when one blows it must be replaced with a new one. As I've already mentioned, old consumer units used fuses; but newer ones tend to replace them with modern MCBs, which cost quite a bit more, but offer improved safety to end-users and the convenience of a device that can be reset rather than replaced. Most power supplies use other methods of protecting against faults, although a fuse might still be included as a last resort. When it comes to electricity transmission and distribution, fuses are still used, but only as a small part of the protection strategy, supplemented by an array of far more complex devices.

There are also certain types of fault where fuses and fuse-like devices aren't much use. They won't offer much help if you accidentally come into contact with a live wire, for instance, which is why modern installations usually include additional protection in the form of RCDs. They also won't do much to protect against arc faults - where electrical currents "jump" gaps between conductors - at least not with any useful degree of reliability, although the faults themselves can be quite destructive.

Fuses also find themselves being used in ways that I would argue are inappropriate. Solar generation is one example. It's not unusual to find fuses protecting the output from solar panels, but when you consider how they work it doesn't make an awful lot of sense. The problem is that a fuse provides over-current protection, which works well with supplies like the grid that are generally thought of as voltage sources (at least at consumer voltages): they are capable of providing very large currents during faults, which ensures the fuse will blow. Solar cells aren't voltage sources. Under fault conditions they act like current sources. There's not much point in putting a fuse on the output of a current source: they naturally limit the current they are able to provide anyway. The fuse might blow eventually, but there's no guarantee it will actually offer the protection we want.

There are plenty of other examples of situations where fuses are unlikely to offer any meaningful protection. Most modern equipment power supplies will have a limited output current. If they didn't, then a fault on the output would destroy the supply before a fuse could protect it anyway. Clever protection circuitry which can respond immediately is used to prevent equipment failure. That doesn't mean we don't want a way of removing the fault.

These may seem like fringe issues now, but with the rise of microgrids and off-grid power systems, the general usefulness of fuses is likely to decline. There's a lot of talk of using DC distribution in aircraft, ships and telecommunications. DC distribution tends to be sourced from specialised power supplies using current limiting which prevents fuses from operating. In these applications, fuses just aren't suited to providing the reliable protection that's required.

You might reasonably wonder why we cannot simply use existing alternatives, such as MCBs and their larger relatives. The problem is that these alternatives operate based on very similar principles to fuses. They typically use the same thermal principles to protect against over-current, albeit with some additional magnetic functions thrown in to speed up tripping in some cases. What is needed is a new alternative to fuses which takes into account the changes we're likely to see in the way we generate and use electricity.

The problem is not insurmountable. It's entirely possible to develop such technology. Something as cheap and simple to construct as a fuse is unlikely to appear, but for the emerging applications requiring reliable protection, the additional cost will be a necessity. After all, if circuit protection was merely a luxury then even the cost of a fuse would be an expense many would seek to avoid.

Protecting power systems is often overlooked, partly because we do it well -- serious faults that cause fires or electrocution are thankfully rare. I suspect it's also partly because the fuse is so popular that we take it for granted, almost forgetting what it's there for. It's also misunderstood and, sadly, I do think it's usefulness is in slow decline. Despite all of this, I suspect that owing to it's simplicity and low cost, the humble fuse will remain an important part of electrical protection for many years to come.