ESP32 - pico -nano fails to reset after multiple power -on cycles
Posted: Thu Mar 02, 2023 3:16 pm
Hello All,
I have a custom PCB with ESP pico-mini -02 module is mounted on.
It is powered by 3.3V with ~ 10uF of capacitance formed by 10u+0.1u+10n+1n capacitors cascade connected close to pin #3 (3V3) pin.
The reset pin is connected to 10K & 1uF network, in a regular manner.
the system power is not stable as it is fed form a PV panel so there is a period in dawn where the system power is very unstable, and we get multiple power cycles during a very short time.
During the power cycle sequence, we suspect that the EN pin fails to go below 0.6 thus the ESP "hangs" and doesn't boot.
The only way to restore operation is to force a hard reset, which solves the situation.
This problem is not very common, however, there are measures inside the ESP that are designed to force a reset on this type of start conditions, including brown-out detection and internal watch dog/s.
I was wondering if anyone encountered this situation and was able to prevent it by using a clever brown-out / watchdog configuration.
Many thanks in advance.
I have a custom PCB with ESP pico-mini -02 module is mounted on.
It is powered by 3.3V with ~ 10uF of capacitance formed by 10u+0.1u+10n+1n capacitors cascade connected close to pin #3 (3V3) pin.
The reset pin is connected to 10K & 1uF network, in a regular manner.
the system power is not stable as it is fed form a PV panel so there is a period in dawn where the system power is very unstable, and we get multiple power cycles during a very short time.
During the power cycle sequence, we suspect that the EN pin fails to go below 0.6 thus the ESP "hangs" and doesn't boot.
The only way to restore operation is to force a hard reset, which solves the situation.
This problem is not very common, however, there are measures inside the ESP that are designed to force a reset on this type of start conditions, including brown-out detection and internal watch dog/s.
I was wondering if anyone encountered this situation and was able to prevent it by using a clever brown-out / watchdog configuration.
Many thanks in advance.