Power out of the base
None of the four bases makes a volt of its own. The 3.3 at the edge is the dev board's regulator output and the 5 is USB power one diode down — which means the 5 V row is dead the moment you unplug the cable.
Everything comes from the dev board
A base has no regulator on it. Trace the 3V3 row at the edge and it goes through the socket to the dev board's 3V3 pin — the output of the AMS1117 regulator sitting next to the USB connector. Trace the 5V row and it reaches the dev board's 5V pin, which is USB power one Schottky diode down.
Two things follow, and both catch people.
The 5 V row is not five volts. It is a few tenths less, because of the diode. That is fine for almost everything a 5 V sensor wants and it is worth knowing before you measure it and conclude something is broken.
The 5 V row is only alive on USB. Power the board from its 3V3 pin — a regulated supply, a LiPo shield, anything that is not the USB socket — and the 5 V row has nothing on it. Nothing back-feeds it.
How much 3.3 volts is there
More than most sensors need and less than a screen with a backlight wants. The regulator is shared with the radio, so the honest way to think about it is: what the ESP32 is not using, you may have.
A WiFi transmit burst is the biggest thing the board does. If a sensor on the 3V3 row misbehaves only when the radio is busy, that is not a coincidence — put a capacitor of 10 to 100 µF across 3V3 and GND close to the sensor, and give anything hungrier a supply of its own.
Which base has what
| Expansion | PinPulse | Pluggable | Screw | |
|---|---|---|---|---|
| 3V3 | both sides | both sides | both blocks | both blocks |
| GND | both sides | both sides | both blocks | both blocks |
| 5V | both sides | both sides | none | left block only |
Ground on both sides of every board is the quiet convenience here. A sensor on the left and a sensor on the right each get their own ground wire, and neither has to reach across the board — which is how ground loops and long return paths start on a small project.
When it does not work
That row is the dev board's 5V pin, which is USB power through a Schottky diode. With no USB cable there is nothing on it — including when you are powering the board through its 3V3 pin instead. Plug the cable in, or feed your 5 V circuit from its own supply with a shared ground.
Everything on the 3V3 row shares one regulator with the radio, and a WiFi transmit burst is the biggest current the board ever draws. Add a 10 to 100 µF capacitor across 3V3 and GND near the sensor, and if it still dips, give the sensor its own regulator.
Feeding 5 V into that row does reach the dev board's 5V pin, and the board's own regulator will run from it. Two cautions: never do it with USB also connected unless you know both supplies can share, and never feed more than 5 V into a row labelled 5V.
It does not, and that is a design choice rather than a fault: twelve ways a side went on 3V3, a ground and ten GPIOs. A 5 V sensor on that base needs its own supply with its ground joined to the board's. The screw terminal base is the one with a 5 V terminal.
Fifteen ways, one ribbon, and a socket that only works face down.
The display port →Edit this page — content/books/esp32-base/power-out-of-the-base.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.