[hvalentim]

Three ideas to ideally improve the measuring of solar radiation for your consideration and criticism:

Limitations to be addressed:
a. Capacitance issues limit the extension of wires to be used and the placement of measuring cells while potentially inducing a slight error due to voltage drops;
b. Arduino reference voltage is not always perfectly accurate;
c. Temperature of the solar cell ought to be taken in consideration in calculations an values read corrected accordingly.

Solution:
a. Use an additional Nano placed right next to the solar cell to do the readings and send them back as data (not voltage) to the TX unit (already there a couple of unused pins that might serve for this purpose?);
b. Use an external reference voltage feed to the appropriate pin by using a dedicated diode;
c. Stick a thermistor to the solar cell (does not seem easy to do to a diode) and program a compensation value to correct readings according to temperature.

This might imply having an additional board with its own power supply and regulator (?).