Fuse Selection: Fuse and Protection Device Selection on PCB

A fuse is the simplest protection device on a board and the one most often chosen by habit. It sits in series with the supply, carries the normal current indefinitely and opens when the current exceeds its rating for long enough. Choosing one correctly means understanding the difference between the current the product draws and the current the fuse must survive, and between the fault the fuse must clear and the fault it must not be asked to clear.

What a Fuse Protects

It protects the wiring and the board from a sustained overcurrent, not the electronics from every fault. A short circuit inside an integrated circuit may draw a current that the fuse never sees because the trace fails first, while a fault on the supply input is exactly what the fuse is there for. Being clear about the target fault is the first step in selection.

The device also protects against a failure of downstream protection, such as a shorted capacitor at the input. Where the board carries a battery or connects to an external supply, the fuse is usually the last line of defence and its placement matters as much as its rating.

Rating Current and Derating

The nominal rating is the current the fuse can carry without opening under defined conditions, and it is not the current a designer should run through it continuously. Thermal derating reduces the usable current at elevated ambient temperature, and most manufacturers publish a curve that shows this.

A common starting point is to select a rating at least 1.5 times the maximum steady state current, then check that figure against the derating curve at the highest internal temperature the product will see. Running a fuse close to its rating shortens its life and makes the opening point unpredictable.

Surface mount fuse placed at a board power entry point

Inrush Current and Time Current Curves

Capacitive loads draw a large current at switch on. The inrush current can be many times the steady state value and lasts for a few milliseconds, which is enough to rupture a fuse element that would never open under normal operation. The fuse must survive that pulse for the life of the product, including thousands of switching cycles.

The tool for this is the time current curve, which shows how long the device takes to open at a given multiple of its rating. Comparing the inrush waveform against the curve, with a margin for ageing, shows whether the choice is safe. Slow blow and time delay devices exist precisely to allow that margin without raising the rating to a level that no longer protects the circuit.

Breaking Capacity and Fault Current

Breaking capacity is the maximum fault current the fuse can interrupt without being destroyed. In a low voltage board fed from a small supply, the available fault current is limited by the source impedance and is usually modest. In a product connected to mains or to a large battery, it can be high enough to require a device with a specified breaking capacity.

Underrated breaking capacity is a safety issue rather than a reliability one. If the fault current exceeds what the device can interrupt, the fuse may arc, shatter or fail to open at all, which is worse than having no protection. The available fault current should be calculated, not assumed.

Placement and Layout

The fuse belongs as close as possible to the point where power enters the board, so that the unprotected length of conductor is short. Anything upstream of it, including connectors, traces and vias, is outside the protection and should be sized to survive the fault current on its own.

Thermal coupling also matters. A device placed beside a hot component sees a higher ambient temperature than the board average and derates earlier, while one placed in a cool corner may not reach the temperature assumed during selection. Our notes on current capacity in copper describe how trace heating interacts with the surrounding design.

Time current curve chart for two fuse ratings

Resettable Devices and Alternatives

Polymeric resettable devices and electronic current limiters avoid the need for replacement, which suits products that must recover without service. They are slower, they have a wider tolerance band, and their resistance is higher, so the voltage drop and the thermal behaviour have to be checked in the application.

Where the current is already limited by an active circuit, a fuse may only be needed for a single fault condition such as a shorted pass element. The decision is a system one, and the broader investigation method for overstress failures is described in our article on electrical overstress investigation.

Coordination with Other Protection

Protection devices should act in a defined order. A small local device should open before the main fuse, so that a local fault does not shut down the whole product. That coordination is achieved by comparing the time current curves of the two devices and ensuring they do not overlap in the region where the fault current falls.

Where coordination is not possible, the design accepts that a local fault will clear at the main fuse and the whole unit will stop. That is a legitimate choice, but it should be a deliberate one recorded in the design file rather than an accident of component selection.

Verification and Certification

Verification includes measuring the inrush waveform, confirming the steady state current at maximum load and temperature, and testing the response to a simulated fault. Standards bodies will ask for the fault current calculation and the device datasheet, so keeping both in the design record saves time later.

The same verification applies to the input filter and bulk capacitance, whose charging behaviour sets the inrush the fuse must survive. Placement of that capacitance is discussed in our article on bulk capacitor placement, and it interacts directly with the protection decision.

Process Control and Verification

Reviewing the design before the data is released is cheaper than correcting it after the panel is in the tank, because every step downstream inherits the decision made at the front end.

FAQ

Can a higher rated fuse be used for safety? No. Raising the rating reduces nuisance opening but also delays or prevents protection. The correct fix for nuisance blowing is a time delay characteristic or a better inrush model.

Is a fuse needed on every board? Where the board draws power from an external source and a fault could heat conductors excessively, yes. On a module inside a product that already has protection, a local fuse may be unnecessary.

How is the inrush current measured? With a current probe and an oscilloscope capturing the switch on event at the worst case temperature, including low temperature where capacitor ESR is highest and the pulse is largest.

1 Comment

  • Trace Fusing And Current Limits On PCB

    2026年 9月 13日 - pm6:06

    […] Where the trace is intended to survive a fault, its fusing current has to exceed the fault current for the duration the protection takes to operate. That comparison belongs with the protection design, and the interactions are described in our article on fuse selection. […]

Leave A Comment