Brownout Protection

The ESP32 will reset itself rather than run on a voltage too low to trust. That protection is a feature: it prevents the corrupted-state failures that a beacon absolutely cannot afford.

What the detector does

When the 3.3 V rail falls below the brownout threshold (about 2.8 V, configurable), the chip asserts reset before the flash and RAM become unreliable. The result is a clean restart instead of random corruption. The firmware can disable the detector, but the beacon never does: a beacon that resets cleanly is better than one that transmits garbage.

The reset signature

On the serial monitor, the loop prints:

Brownout detector was triggered

followed by a reboot. The trigger almost always coincides with the TX burst, because the burst is the highest current draw of the cycle (see Current Draw by Mode).

The hardware fixes, in order

1. Fresh, healthy cell: the sag test in Battery Troubleshooting.
2. Short, thick power wires: voltage drop is I x R; the burst current makes thin wires sag.
3. Decoupling: 100 - 470 uF at the board’s power input absorbs the burst transient.
4. Lower TX power: +17 dBm draws less than +22 dBm; in extreme cold this is the pragmatic fix (see Cold Weather Batteries).

The firmware side

The battery monitor reports the rail voltage continuously. A brownout during TX means the monitor was already near the warning threshold; the Battery Monitor Details page explains how to read the sequence.

When NOT to disable it

Disabling the brownout detector to “fix” resets is trading corruption for resets. The resets are the symptom; fix the supply. If a mission requires running on a nearly-dead cell, lower the power setting instead of disabling the detector.

Related pages

• Boot Loop Troubleshooting for the loop diagnosis
• Battery Troubleshooting for the cell side
• Power Measurement for the measurements