While most other electronic and electromechanical components have moved into silicon devices, the classic cartridge fuse has remained the mainstay of circuit protection. However, even this bastion of the conventional is being displaced by new technology – the eFuse. Admittedly, they are not designed for use in circuits using alternating current or at the high voltages found in mains supplies. All the same, they do offer developers fuse functionality that provides a little intelligence.
While most other electronic and electromechanical components have moved into silicon devices, the classic cartridge fuse has remained the mainstay of circuit protection. However, even this bastion of the conventional is being displaced by new technology – the eFuse. Admittedly, they are not designed for use in circuits using alternating current or at the high voltages found in mains supplies. All the same, they do offer developers fuse functionality that provides a little intelligence.
Who designed the fuse?
We happily link the great scientists of the 19th century, such as Ohm, Faraday, and Henry, with the basic three passive components (resistors, capacitors, and inductors, respectively). But, the origins of the fuse will likely be a mystery to most, even highly-tenured, engineers. It seems that, after the all the effort that went into the big three, everyone had lost interest in this seemingly simple component. After all, the entire purpose of the fuse was to self-destruct when something went wrong.
Cartridge fuse in a multimeter.
Much correspondence went back and forth between the great and good of the day, trying to pin-down who had invented the fuse. At that time, the ‘electric light’ was a novelty and, to stop the wiring causing a fire in the event of a short circuit, fuses were often also being installed. Various designs were even patented including one by Prof. S. P. Thompson in 1883, and even Thomas Edison in 1890.
Early fuse designs included low melting-point alloy balls,
leaf-spring soldered plates, and weights.
Perhaps the most research, however, was undertaken by A. C. Cockburn resulting his paper presented in 1887. Concerned by the fuse's inconsistency in both function and their holder's physical construction, he undertook considerable experimentation.
Amongst other things he analyzed the materials that would be most suitable, along with the thermal effects the fuse-holder had on function, as well as the aging of the final device. He also defined that fuses should be rated at 150% – 200% of the circuit they were protecting, a little more margin than thought appropriate today.
So, what is an eFuse?
The big problem with the fuse is the issue of serviceability. While it provides protection against overheating and potentially fire, once blown it must be replaced. As well as finding the correct replacement it is also necessary to check why the fuse blew in the first place. If the appliance it is protecting has a permanent fault, this will need to be found and repaired first.
With the robustness of silicon power devices today, most short-circuit situations are not destructive. For example, ideally, plugging a faulty device into a USB socket should not cause the PC to shut down; it should just respond by removing power from the affected port. Then, having checked that the faulty device has been removed, it can be enabled again.
eFuse devices from Toshiba, STMicroelectronics, and On Semiconductor (to name just a few) offer resettable fuse functionality in silicon. Of course, an electronic fuse is not totally new and circuits for such devices have been covered in the past, but today's devices are compact, highly integrated, and easy to use.
As a result, such eFuse devices usually feature output voltage clamping and under voltage lockout, while a thermal shutdown cuts in should it get too hot. A dv/dt control is often provided too, allowing the turn-on rate of the output to be defined and thereby limit inrush currents.
How is an eFuse used?
A device such as the TCKE805NL from Toshiba is easily integrated with a microcontroller to provide a latched 5.0 V protected output. This could be implemented on any application from which other user-installed devices draw their power, such as SATA and USB hard drive interfaces, smartphone chargers, programmable logic controllers (PLC), or oscilloscope inputs that power probes.
The device in the circuit shown features over voltage clamping fixed at 6.04 V, while a 75 kΩ resistor at the ILIM pin sets the current limit to around 1.5 A (±11%). The EN enable pin is used to enable the output or re-enable the output after the latching fuse protection has been tripped. Finally, to provide extra robustness, an external NFET can be included that protects from reverse currents.
eFuse based upon TCKE805NL with microcontroller.
In combination with a microcontroller, it is possible to detect when the eFuse has tripped. This can be implemented with a simple GPIO pin or, to acquire more information, with an analog-to-digital converter (ADC).
Should a short-circuit condition occur, an internal fast-trip comparator responds in just 150 ns – much faster than a microcontroller alone could respond using a software algorithm.
Various options are typically offered, such as 5 V, 12 V, or no over voltage protection, and there are even versions with non-latching outputs that automatically try to reapply power at their output.
Output protection
As we are accustomed, the march of progress continues even in the word of the humble fuse. With so many applications these days being capable of providing power to charge other devices, or expansion features to be attached, a robust but resettable eFuse makes for a simple but capable output protection feature. I’m sure that A. C. Cockburn et al. would be excited to learn what has become of their humble fuse!
