Page 1 of 1
DevKit LED and Power
Posted: Tue Nov 24, 2020 10:02 pm
by bkgoodman
I have an ESP32 DEVKITV1 board I wanted to experiment with some low power/battery powered stuff.
I took 2 AA batteries and ran them between GND and 3V3. The board worked fine - but the red LED came on, and I did not expect it to.
According to all the schematics I could find - the red LED is tied right to 5v EXT power (USB power) and is on the other side of the LDO than the 3V3 rail - i.e. there is no way this LED should see power or come on.
I am assuming something is different here than I am expecting - i.e. wrong schematic or something?? Any ideas??
Re: DevKit LED and Power
Posted: Wed Nov 25, 2020 3:07 am
by ESP_Sprite
From the AMS1117
datasheet:
"Diodes between the input and output are not usually needed. Microsecond surge currents of 50A to 100A can be handled by the internal diode between the input and output pins of the device."
In other words, the AMS1117 has an internal diode allowing current to flow back from the output to the input.
Re: DevKit LED and Power
Posted: Wed Nov 25, 2020 3:20 am
by embedic
ESP_Sprite wrote: ↑Wed Nov 25, 2020 3:07 am
From the AMS1117
datasheet:
"Diodes between the input and output are not usually needed. Microsecond surge currents of 50A to 100A can be handled by the internal diode between the input and output pins of the device."
In other words, the AMS1117 has an internal diode allowing current to flow back from the output to the input.
I agree with your opinion.
Re: DevKit LED and Power
Posted: Wed Nov 25, 2020 4:13 am
by bkgoodman
So.....that's all the more reason why this should NOT happen. But when I put 2 AA batteries to 3.3V rail - the red LED lights up! This shouldn't happen. Any reason why it does?? (A diode in which direction?)
Re: DevKit LED and Power
Posted: Wed Nov 25, 2020 8:24 pm
by becorey
Put a multimeter on the 5V pin. See if you measure about 3.3-0.7V, that would tell you there's probably a diode drop going from the LDO output to input.
Re: DevKit LED and Power
Posted: Thu Nov 26, 2020 1:38 am
by ESP_Sprite
There is an internal diode going from the output to the input (so opposite to the normal current flow in a LDO) - presumably to make sure the input of the LDO always has power if there is a voltage on the input; I seem to remember LDOs can self-destruct if that's not the case. See figure 1 in the datasheet I mentioned but imagine that diode is within the AMS1117 package.