Not going to happen -> https://esp32.com/viewtopic.php?t=46102#p149079


so according to you is it not possible to bring the consumption down to microamperes using standard devkit boards? since most of them use a AMS1117 LDO?


I'm powering my boards with an external battery pack and a low ...

power consumed must be in microamperes. how to manage that using ESP32Devkitc V4 and a MOSFET?

Not going to happen -> https://esp32.com/viewtopic.php?t=46102#p149079


so according to you is it not possible to bring the consumption down to microamperes using standard devkit boards? since most ...

Okay, and how did you connect the other hardware to it?

(Note that by starting from the DevkitV4, you're already at a bit of a disadvantage... those boards aren't really optimized for battery use as they use a LDO with a fairly high quiescent current.)


in the deep sleep mode, the power ...

Also which specific MOSFETs can be used?

Any P-channel MOSFET with a V[th] significantly less than 3.3V (or greater than -3.3V depending on the convention used) which can conduct enough current for your needs at V[gs] = -3.3V.
The AO3401A, for example, with a V[th] of -0.9V (typ.) is cheap ...

What LCD?!
How much power?
If you want to actually completely switch off power (3.3V) of external devices (which you can't power directly from a GPIO pin like the AHT10), the easiest way is indeed to just hook up a P-channel MOSFET to 3.3V and a GPIO output ("high-side switch").


The LCD I ...

What LCD?!
How much power?
If you want to actually completely switch off power (3.3V) of external devices (which you can't power directly from a GPIO pin like the AHT10), the easiest way is indeed to just hook up a P-channel MOSFET to 3.3V and a GPIO output ("high-side switch").


The LCD I have ...

Try gpio_deep_sleep_hold_en() and see if that makes a difference.

Thanks! but it didn't work. I think the LCD and the AHT10 are still drawing power even when the esp goes to deep sleep. How to stop that completely.

I dunno? You can check if the device goes into sleep mode after a measurement (what that library implies) by either asking the manufacturer or by testing it. The alternative, as I said, is to power the sensor directly from a GPIO; the GPIO current should be enough to power it and that way you don ...

This driver hints that there is a measure + go to sleep command; you could try that. If you want to be really sure: the sensor only seems to draw a few tens of uA in active mode; you could just tie its power supply pin to a GPIO and pull that high to power the sensor.

So is using a transistor ...

According to the docs for that sensor, it has a sleep mode in which it only uses 0.25uA.

But there are no specific commands for entering the sleep mode using AHT10, right? The state in which it uses only 0.25uA is the state immediately after a power up and before you send the initialization ...