I'm setting up the code for a MailboxNotifier using an R3 MotionMote board with an R6 Moetino that will run from voltage produced by a solar panel through a schottky to a supercap, plugged into the battery connector. I note from the R3 schematic that the voltage divider for battery sense (1Meg between A7 and battery, 2Meg between A7 and ground) gives a battery sense voltage of 2/3 of the battery voltage. Using 2/3 gives a maximum battery voltage of 4.95 for an A7 voltage of 3.3. However the solar panel can charge the supercap much higher than that, which puts the sense voltage over 3.3 volts.
When I put the solar panel in direct 4pm sun with everything connected and on, the battery voltage is 5.775 but the voltage shown on the receiver display is 4.81 volts. 2/3 of 5.775 is 3.85 volts, well over 3.3 volts, so the analog reading is skewed. I don't yet know what the maximum battery voltage will be in direct noon sun, so I'm not sure if I'm in danger of exceeding the rated maximum voltage on A7. So far, it's been slightly over 6 volts unloaded around noon with the proper angle. Installed, it won't have full sun very often.
Does anyone have any field experience with the solar-supercap setup? Is there a way of correcting with the analog reading in software when the battery voltage is well over 5 volts to give an accurate battery voltage reading, or would it be better to just keep the solar panel in light shade to reduce the maximum voltage it puts out, which seems to be around 4.2 volts give or take?