The fuse and the diode
Every board puts two parts between the USB-C socket and your rail: a resettable fuse that holds half an amp, and a diode that lets current go one way only. One protects your charger from a short; the other protects it from your other supplies. Both cost you something.
Two parts, in this order
From the USB-C socket, current goes through F2, a resettable fuse, then D3, a diode, and only then reaches the rest of the board. It is the same on the standard, the step-down and the booster: after these two, the three boards go their separate ways.
The fuse: half an amp, then it trips
F2 is a Jinrui JK-nSMD050, a polymer PTC fuse. The maker's table gives two numbers for it:
| Holds | 0.50 A, indefinitely, at 25 °C |
| Trips | 1.00 A, for certain |
Between the two it may trip or may not, depending on how warm it is. When it trips, its resistance jumps from a fraction of an ohm to a very large one, the current falls to a trickle, and it stays that way until the fault is removed and it has cooled. Then it resets. Nothing is replaced.
What it protects is everything upstream: the USB port or charger, the cable, the socket. A wire slipping across the rails pulls a few amps from a charger that can supply them; the fuse limits the damage to a warm fuse and a dead rail.
It also sets the limit for everything the board can do. Every milliamp on either side comes through this one fuse. Both sides share it. The step-down and the booster share it with their regulators, which is why neither can deliver what its silkscreen says, as their chapters show.
The diode: one way only
D3 is an SS24, a Schottky diode rated at 2 A and 40 V. A diode passes current in one direction and blocks it in the other. Here it passes current from the socket to the board, and refuses to pass it back.
That matters the moment there is a second supply on your breadboard. A dev board plugged into your laptop has 5 V on its 5V pin, and if that pin is wired to the rail, the rail is live even with the adapter unplugged. Without the diode, that 5 V would push current backwards through the adapter and into whatever the USB-C cable is plugged into. With it, nothing goes back.
What they cost
Neither part is free. The fuse has a small resistance even when it is not tripped — the maker gives 0.15 to 0.7 Ω — and the diode takes a few tenths of a volt for letting current through the right way. Together they are why the standard board's "5 V" rail reads 4.7 V on a meter. The next article is that number.
When it does not work
That is the fuse tripping and resetting. Something drew more than it can hold — a short, a motor starting, a booster asked for too much. It cools, its resistance falls, and the rail returns. If it keeps happening, the cause is still there: measure what the circuit draws, or look for the short.
It is a 1206 surface-mount part and could be swapped, but the socket, the diode and a breadboard's contacts are what it protects, and none of them get bigger with it. If you need more than half an amp, use a bench supply or a dedicated regulator module wired with proper cable.
The adapter's diode stops the ESP32's USB supply pushing current back into the adapter's charger, so nothing breaks at that end. But the ESP32 board has its own diode, or none, and two supplies joined on one rail share the load unpredictably. Join the grounds, and feed the ESP32's 5V pin from the rail only when its USB cable is unplugged.
Hot enough to feel, which is how it works: the polymer inside expands as it warms and its resistance climbs by orders of magnitude. It holds that state for as long as the fault is there. Do not touch it while testing a short, and give it a minute after the short is gone.
Edit this page — content/books/bbpwd/the-fuse-and-the-diode.mdx
Questions about this product
See what other owners have asked, and read their solutions.
This page covers several products. Choose yours to see the right discussions.
Discuss this article
Ask about this page. The answer stays here, on the page it belongs to, for whoever hits the same wall next.