With proper use of deep sleep modes and sensible power budgeting, 1F is more than enough for overnight, as this fun tutorial demonstrates. No boost converter (which don't work well with PV cells) or charger IC is required.
Small value supercaps designed for memory retention can have a high internal resistance.
Look at the datasheets if it's able to deliver that current.
Leo..
I should work, with that IC or a normal boost to 5V, and an LDO like the MCP1700 to 3.3V, and deepsleep. Add a bulk capacitor in the ESP side, like 100uF, for the short current peaks.
You can also monitor the supercap voltage and increase the sleep time when it's below a threshold. Or take the samples and store them in RTC memory, but send with less frequency.
I'm not an expert in reading this datasheets, but it seems to deliver more than 200mA
And if uses all 42mWh (esr 0,23 ohm) it can run 2 sek 200mA 74 times, enough for me.
ESR is the internal resistance of the cap.
Simply use Ohm's law to calculate voltage drop.
The 1F cap with an ESR of 0.4 Ohm will internally drop 0.4 x 0.4A = 0.16volt at 400mA.
I think I told you about leakage current of supercaps in another thread.
The larger your cap, the higher the leakage you have to deal with (larger solar panel).
The closer to 5.5volt, the higher the leakage is.
Leakage during charging can be much higher, for a while.
Leo..
You could also consider sodium batteries. They work below -10c.
Regarding the supercaps, with 10F you have for 2 or 3 days at least of power, if you wakeup the MCU once per hours or so.
And when there is strong light the supercap will charge in a couple of minutes. With weak light or covered, it takes longer but not a lot.
@Wawa
I have done some thinking about the superCap test, and I plan to build it, (not shure if I have an Uno, must check.
I have 5.5V superC.
Also tried to understand your spoken schematic.
Here is my humble try to understand it:
Arduino Uno
12x 330 ohm from pin 2-13 to a star point
330 ohm from A0 to star point
6k8 between Aref pin and 3.3volt pin, to create a 2.7volt Aref, ! measure it !
=> NB if 5V cap use const float Aref = 5.0;, and no resistor
1N4004 diode (protection) between star point and 5volt, cathode (ring) to 5volt
supercap connected to star point (+) and two ground pins (-)
Consider also the use of a proper energy harvester circuit ( they are also more expensive tough ).
It will come handy when your circuit will 'run out of juice': when the ( solar ) power is available again the microcontroller could be unable to reset and start normally due to limited/unstable power available.
This case could be handled correctly even by the brown-out circuitry of the microcontroller, but better verify in advance ( and the energy harvester will deal correctly with these cases ).
The energy harvester will also help you reduce these cases as it countains an mppt circuit that help extract the 'maximum available power' from your solar panel, and they are able to start charging the supercap even with super low input voltages ( yes I know in these conditions the power extracted is very low, but what surprise me is that they work )
No, the bq25173 is simply a charger, you set charging current and maximum output voltage ( it is intended to be used with an adequate input power, so don't know if this will handle correctly the limited and varying power available from a solar cell, but it could do well nonetheless ).
You can find several ( every manufacturer has some in ther product line ):
even dfrobot has one ( I received an email this morning ) it is quite interesting as it does not need any external componet, for a standard usage ( but its datasheet is quite poor of informations )
...
