Thermal Fuse Protection Circuit Design

A thermal fuse is a one time device that opens when its temperature exceeds a set value, and it is the last line of defence in a product that can overheat. Used well, it converts a dangerous failure into a product that stops working; used badly, it either never operates or operates during normal use.

Thermal Fuse, Thermal Cut Off and Protector

A thermal fuse is a non resettable device with a metal alloy element that melts at a defined temperature. Once it opens it stays open, and the product has to be repaired or replaced.

A thermal cut off is a similar device, often distinguished only by the standard it is approved to, and the terms are used interchangeably in some industries. The important property is that it does not reset and it cannot be overridden by the user.

A self resetting protector uses a bimetal disc or a polymer element and returns to the conducting state when it cools. It protects against a temporary condition and it can cycle, which is acceptable for a stalled motor and not acceptable for a fire risk.

trip temperature and Its Meaning

The trip temperature is the temperature at which the device operates, and it is specified with a tolerance that is wider than most designers expect. A device marked at a hundred and thirty degrees may open anywhere between a hundred and twenty and a hundred and forty.

The device also has a holding temperature, below which it will never open, and a maximum temperature at which it can be used without opening. The design window is between the highest normal operating temperature and the holding temperature, and it has to accommodate the manufacturing tolerances.

The response time matters as well. A device that opens in a second at the trip temperature takes much longer at a temperature just above the holding value, and a slow rise can therefore leave the product above its safe temperature for a long period.

Where to Place It

The device has to see the temperature of the part that overheats rather than the temperature of the air in the enclosure. Mounting it on or near the heat generating component is the only way to achieve that, and a device on the far side of the board responds too late.

Thermal coupling is the practical problem. A device clipped to a heatsink follows the heatsink; one standing in air follows the air; and one bonded with a thermally conductive adhesive follows the surface it is bonded to. The choice should reflect the failure mode being protected against.

Where the risk is a fire started by a component, the device is placed close to that component. Where the risk is the enclosure becoming too hot to touch, it is placed where the enclosure temperature is representative.

Electrical Rating and fault current

The device carries the full load current and has to be rated for it with margin. The rating is set by the internal element, and a device carrying close to its rating runs warmer than the ambient, which reduces the margin to the trip temperature.

The breaking capacity is the current it can interrupt without arcing across the gap. A device that opens on a fault current larger than its rating may continue to conduct through the arc, which is the worst possible outcome for a protection device.

Where a large fault current is possible, a current limiting device is placed in series so that the thermal fuse only ever sees a current it can break. The two devices protect against different things and neither replaces the other.

Interfacing and Detection

A blown fuse removes the supply from the circuit, and the product then appears dead. Where the cause has to be reported, a second contact or an optical indicator shows that the device has operated, and that information is worth more than the few cents it costs.

Some designs use the thermal fuse to trigger a controlled shutdown rather than to interrupt the current directly. A sensor detects the overtemperature, the controller stops the load, and the fuse is the backup. That arrangement avoids interrupting a large current and keeps the product repairable.

A resettable protector used in the same role needs hysteresis, or it will cycle at a rate set by the thermal time constant of the product. The cycling is visible to the user and can stress other components if it continues.

Standards and Approval

Safety standards require that a single fault cannot cause a hazard, which usually means two independent means of limiting the temperature. A controller with a sensor is one, and a thermal fuse is the other, and the fuse must work even if the controller has failed with the load on.

The device has to be approved to the relevant standard for the application, and the approval covers the trip temperature and the breaking capacity. A general purpose part with the right number on the label may not be acceptable.

The documentation has to state which fault each protective device covers. A design where the purpose of the fuse is unclear tends to lose it during a cost reduction, and the product then relies on a single measure.

Enclosure and Airflow

The thermal design of the enclosure decides how much margin the protection has. A product in a sealed plastic case runs hotter than the same circuit in a metal one, and the internal temperature rise can be tens of degrees above the ambient even with a modest dissipation.

Airflow, including the small amount produced by convection inside an enclosure, changes the temperature distribution and therefore the temperature that the device sees. A design that depends on a particular airflow has to be tested with the airflow present and with it blocked. The wider approach to moving heat out of a board is set out in our guide to PCB thermal management design.

A product that has to pass a test at forty degrees ambient needs the device chosen from the internal temperature at that ambient rather than at room temperature. The internal rise is measured rather than assumed, and the margin that remains to the holding temperature is the answer the design has to produce.

Testing and Verification

Verify the protection with a single fault test. Disable the controller’s own protection, force the load on, and measure the temperature of the part that is at risk while the product runs. The fuse should open before that temperature reaches a value that damages anything.

Record the temperature at which it opens, and repeat on several samples, because the tolerance on the trip temperature is wide. A design that only just passes on the first sample will fail on another.

After the test, replace the fuse and repeat to confirm that the rest of the circuit survived. A protection device that works but leaves the product unrepairable is a design compromise rather than a failure. The release checks that keep such protection in place are collected in our PCB design release checklist, and the assembly points are listed in judging PCB quality.

FAQ

Should I use a thermal fuse or a resettable protector? Use a resettable device for a temporary condition and a one time fuse where the failure must not be able to repeat.

How close should the fuse be to the hot component? Close enough to follow its temperature. A device in free air on the far side of the board responds far too late.

Can the controller replace a thermal fuse? Not for safety purposes. The protection circuit has to work with the controller failed, so a physical device is required.