Fuse Traces and Overcurrent Protection on PCBs

A fuse trace is a deliberate weak point: a short, narrow section of copper that melts before the rest of the circuit is damaged. It costs nothing, needs no component, and fails more predictably than many engineers expect. It also has to be dimensioned carefully, because a trace that is too small will open on a normal surge and one that is too generous will let the real fault through. This article covers the calculation and the layout.

When a Fuse Trace Is the Right Choice

A fuse trace suits a low power branch that must be protected from a fault occurring downstream: a sensor input, a small fan, a status LED, a modular connector that a user can plug the wrong thing into. It is attractive where space is scarce, where the assembly must survive a high temperature process, and where a conventional fuse would add cost and a placement step.

It is the wrong choice where the fault current is not well defined, where the trace is exposed to mechanical damage, or where a resettable or serviceable device is required. Because the fuse trace is part of the board, replacing it after a fault means repairing the board, so the design should be used where a fuse is expected to be a last resort rather than a routine consumable.

Fusing Current and the Onderdonk Relation

The classical relation between current and melting time comes from Onderdonk and is widely quoted for a given conductor cross section. It states that the current required to melt a conductor in a defined time rises with the cross sectional area raised to a power of roughly three quarters. Halving the width of a trace therefore reduces the fusing current by close to 40 percent rather than by half, which is why a modest width reduction can create a usable weak point.

The relationship assumes an adiabatic condition, meaning that the heat generated by the fault current is not conducted away. That assumption holds for a short pulse of a few milliseconds but not for a sustained overcurrent, where the surrounding laminate, the copper plane, and the solder joints all act as heat sinks. Practical designs are therefore validated with a current ramp rather than by calculation alone.

Dimensioning the Trace

Dimensioning starts from the normal operating current. The trace must carry it with a temperature rise and a voltage drop that the circuit can tolerate, which usually means the width is set by the voltage drop rather than by heating. From that width, the fuse section is reduced until the fusing current sits at a comfortable multiple of the normal current, often two to three times, and below the current that would damage the rest of the circuit.

Length matters as much as width. A short constriction loses less voltage but conducts heat away more effectively, so it fuses at a higher current than a longer one of the same width. The practical approach is to keep the constricted section long enough to be thermally isolated from the wide traces on either side, typically several millimetres, and to verify the result with a test on the actual stackup rather than on a generic coupon.

Copper weight then sets the range that is achievable. On 1 oz copper, a constriction that survives a few hundred milliamperes is practical; below that the trace becomes so narrow that manufacturing tolerance dominates the result. Where a smaller current must be protected, a component fuse is more appropriate than a trace.

Layout Rules for a Sacrificial Section

The constriction should be placed where the failure can be contained. Keep it away from the board edge, away from connectors, and away from anything that would be damaged by molten solder or by a flame. Because the trace opens with a small arc, a keep-out area free of solder mask and free of nearby conductors is advisable, and the surrounding material should not be a combustible surface.

The return path deserves the same attention as the fuse itself. If the fuse is placed in the positive branch but the return path is a wide plane, the fault current still flows and the plane may be damaged before the trace opens. Protecting only one side of a branch is normal in low voltage designs, but the return path should still be sized so that it does not fail first.

Test points or a current sense element should be placed on the load side of the fuse, so that a fault does not destroy the measurement circuit. Where the branch feeds an external connector, a series resistor or a transient suppressor in front of the fuse protects against a surge that would otherwise open the trace immediately.

Voltage Drop and Thermal Considerations

A constricted trace has a higher resistance, and that resistance appears as a voltage drop and as heat. For a supply branch this matters, because the drop increases the further the load is from the source and can push a rail out of specification. Calculating the resistance from the sheet resistance of the copper and the geometry takes a minute and prevents a design that protects the circuit but no longer powers it.

Thermal coupling is the other consideration. If the constriction sits next to a hot component or on top of a plane that conducts heat away, its fusing current rises and its behaviour becomes less predictable. Isolating the fuse trace from copper planes with a small thermal break makes the fuse behave as designed, and it also prevents the plane from being damaged when the trace opens.

Where the trace runs on an inner layer, heat removal is different again, since the laminate above and below conducts differently from air. This is a reason to validate the design on the layer where it will actually be produced, and to record the current carrying capacity used for the calculation.

Validating the Design

Validation uses a controlled current source and a rising current, with the board instrumented so that the melting point is recorded. The test should be repeated on several samples, because copper thickness and trace width vary within a panel. The observed fusing current will usually be lower than the calculation for a sustained fault and higher for a fast pulse, and the spread across samples tells the designer how much margin exists.

Where the branch is safety related, the validation should also confirm that the trace opens before anything else fails. That means monitoring the temperature of nearby components and the integrity of the board material, not only the moment of the open circuit. If another element fails first, the fuse trace has not achieved its purpose regardless of its calculated rating.

Documentation and Marking

Because a fuse trace is invisible to anyone reading the schematic, it should be documented twice: as a note on the fabrication drawing with the required width, length, and copper weight, and as a legible silkscreen marking that identifies the protected branch. That marking is what allows a technician to recognise a blown trace rather than diagnosing a component failure.

The documentation should also state the normal current, the expected fusing current, and the layer. When the design is revised, a change to the copper weight or the layer stack can quietly change the fusing behaviour, and a note that quantifies the expectation makes that change visible during review. It also gives the fabrication review a checkable item rather than an assumption.

Additional Considerations for This Build

Practical attention to current density pays for itself here, because it decides whether the finished board behaves as the drawing intended. Where the requirement is not stated on the fabrication drawing or in the assembly notes, the shop has to assume a default, and that default is rarely the value the design was simulated with. Stating current density explicitly, together with the tolerance that applies, removes the assumption and keeps the result predictable from batch to batch.

Deliberate attention to overcurrent protection pays for itself here, because it decides whether the finished board behaves as the drawing intended. Where the requirement is not stated on the fabrication drawing or in the assembly notes, the shop has to assume a default, and that default is rarely the value the design was simulated with. Stating overcurrent protection explicitly, together with the tolerance that applies, removes the assumption and keeps the result predictable from batch to batch.

Related reading: our fabrication notes, board quality and design release notes cover the same ground.

FAQ

How accurate is a fuse trace? Within perhaps plus or minus 20 percent for a controlled geometry on a known stackup, which is usually adequate for branch protection but not for a safety critical limit. Validate on production panels rather than trusting a calculation.

Can a fuse trace be repaired? It can be bridged or rebuilt with a wire, but the replacement will not have the same fusing characteristic. Repair should restore function only, and the assembly should be marked as such.

Does the solder mask affect fusing behaviour? Yes. Mask over the constriction slightly changes heat loss, and mask that cracks when the trace opens can leave debris. Leaving the fuse area uncovered makes the failure easier to see.