Yet another 3.6v battery charging question.

Just checking my logic and feasibility.

My project is a sensor application activated by a watchdog timer, so average draw is quite small.

I want to power it with solar, so my typical newbie question is this: can I get by without any charging controller without my battery exploding (or otherwise malfunctioning). But here's my reasoning and my sample configuration:

System: Arduino Pro Mini 328 - 3.3V/8MHz
Battery: 3.6v coin LR2032, full charge = 4.2v (actually probably 2 of these in parallel)
Solar: KXOB22-01X8F open circuit voltage = 4.7v, minus 0.7v for the diode = 4.0v
scc = 4mA

So my prediction is that the current will asymptotically drop to zero as the battery approaches 4.0v, and therefore not explode.

Am I right?

If you're using a silicon diode - probably yes.


Are you sure that the solar panel can only produce 4.7v even in very bright sunlight?

If not , some active protection may be necessary.



Well the danger threshold would theoretically be 4.9v (better leave a little margin though). Seems not too likely, since V increases as the log of solar intensity. The theory suggests that you’d have to about double the solar output to pick up that extra 0.2v. Doubling the solar output would mean that I’m no longer worried about my project.

The extra power from more sunlight mostly comes from increased current, which is not our concern since we are interested in OC voltage.

I suppose a possible no-cost software countermeasure would be to make the system stay on for a while if the input voltage indicates an unexpectedly full charge. Ensure that the thing is always guaranteed sufficiently discharged to be safe until the next wake up. It seems like a low probability scenario.

I guess I’ll have to order it and do some testing. The specs seem to indicate that the OC voltage is independent of input light level.

I found a couple nice sources:

Sounds OK to me - give it a try



That cell has a MPP point of 3.4V so its hardly any use for charging a 3.6V battery.

You want a cell with a MPP of about 5V (open circuit therefore more like 7V) and limit the output
voltage with an LDO regulator. The diode and LDO regulator together will drop around 1V--1.5V,
which is why 5V at MPP is the sort of spec for the PV array (ie about 10 cells if silicon).

Remember the MPP point for bright sunlight is higher than that for overcast conditions, so if
you want charging to happen on dull days you might want to consider even higher voltages,
or use of a DC-DC converter for generating the 4V output. Ideally a DC-DC converter that
could track the MPP for the cell as well as limit output voltage...

Ah thanks, that’s the kind of thing I need.

Now looking at: NJM2870F48-TE1#-ND 4.8V, 150mA (-0.7 = 4.1v)

But now I can’t seem to find the right solar unit. Lot’s of things at DigiKey orbiting my specs.

cost < $5 @ 10 units

vMpp around 4.5v-7v

But the current can be quite low. I think even just a few mA would do.

Alas this energy-harvester unit doesn't quite seem to be the right output voltage, but
it might be worth looking for other such chips/modules: