Work on an ESP8266 based temperature, light, pressure and humidity sensor has moved to version 2. The new version includes a battery charger for LiFePO4 chemistry batteries.
http://sensornodeinfo.rockingdlabs.com/component/tlphnodeV2/
So this is an always on device? Choosing a regular instead on a buck converter although lowering the cost seems odd.
The sensor is not always on, it uses the deep sleep mode of the ESP-8266. Current draw at that time will be in the 50uA range. A couldn’t find a switching regulator with as low a power. Also the regulator has to ‘drop-out’ as the battery voltage falls so that full voltage range of the battery can be used. I also looked at buck-boost combinations to provide a fixed voltage, but the overall power drain was higher.
Really? I would have thought, given thats what regulators do, is that it would just constantly burn the energy over 3.3v, which is why they aren’t used in battery powered devices.
I must examine the specs of this one then!
You want to look at the quiescent current. That is the current the regulator itself draws. These regulators do a nice job of mode detection and can operate at very low quiescent current when there is no other (or very low) current draw.
Ok, so I went and read the first LDR - the LD39130S (i’m very interested, power is a real problem for these devices and I haven’t found any easy solution). So your ESP is going to need at least 10uA of power, which your regular needs to provide constantly. This implies it will need to be in green operational mode (1uA) - is that the plan? Until it wakes up, when the regulator will need to swap up to 45uA? That sounds pretty good?
Yes, at that point the Esp8266 will be drawing as much as 200ma while transmitting, so the 45uA is not much. The key is to quickly get back to sleep!