I am powering an off-grid ESP32 project from a 12v battery (12v because that's what the connected pump and solenoids use). The ESP32 is directly on PCB; I'm no longer using Dev Kits. I'm needing a regulator that brings the 12v down to 3v3 for the microprocessor, one that uses minimal current to conserve battery power. The 12v input may rise to over 14v as there's a solar panel involved with charging the lead acid battery, which affects my choice of regulator.
The data sheet for the LF33 (https://www.mouser.co.uk/datasheet/2/389/lfxx-1849555.pdf) shows quiescent current 12mA or less (target < 1mA) but I'm seeing in the region of 20mA with or without components (such as the ESP32) downstream of it. Edit: the excess seems to be downstream of the regulator.
A direct swap on the same PCB with a DC-DC converter (https://www.mouser.co.uk/ProductDetail/CUI-Inc/VXO7803-1000) reduces current down to 7mA, so it's clearly something up with how I'm using the regulator. But I can't figure it out. The relevant part of the schematic is:
Edit; added full schematic (though the ESP32 isn't placed on the PCB for this test board)
Edit; the problem is downstream of the regulator, so I've updated the question title.
Best Answer
The answer was the CP2012 was not de-powering (going into suspend mode), and it was drawing about 20mA. I tried bypass capacitors, pull-down on VBUS and D-, pull-up on RSTb with no joy. Sometimes, just sometimes, it powered down; more often with the DC-DC converter regulator for some reason. As I only rarely need UART (debug mainly) I'll put a jumper in place to isolate the UART completely when on battery power. CP21xx devices can be made to run off the USB power only, so powering off the plugged-in cable is an option too.