[werk_ag]

Zdenek

I like the idea to include something in the formula to compensate the attenuation caused by the sensor covering material.
So maybe something like this would be more appropriate:

IndiceUV = (((ADC1 * 12.0) / 1023.0) * (1 / Transmission));

I know this formula is despising (again) the nonlinear response of the sensor output voltage between UV Index 0 and UV Index 1 (this zone of the readings is the less significative).

As said before, your work in this matter is greatly appreciated, so, please continue...

When I tested my UV sensor, I have separate Arduino Mega with output to LCD and measured it on this polygon.

This explain some things to me.
If possible, try to do your tests with the UV sensor connected to the TX unit, creating test conditions as close as possible to real use conditions.