On Tue, 27 Aug 2019 at 14:08, Herbert Graf wrote: I've been interested in integrating an illuminance sensor. The issue is all > the ones I've been able to find don't have near enough the dynamic range > I'm > interested in. Specifically I want to be able to differential, from the > data, a cloudy day from a sunny day, and even how late in the day it is. > None of the commercial solutions I've found offer that sort of resolution= .. > Here is something which is easy to try and may work well enough for you. Any silicon (and many other) PV panels (solar panels) are close to illuminance controlled constant current sources. When loaded to substantially below their Vmp (maximum power voltage point) they are essentially a current source whose output is proportional to illuminace. Take as an example a say 5V Vmp, 100 mA Imp PV panel. It's Isc will probably be 110-120 mA. At 5V, 100 mA Rload =3D V/I =3D 5/.1 =3D 50 Ohms. Load it wit say 10 Ohms and in full sun Vout will be in the order of V =3D = IP =3D 0.11 x 10 =3D 1.1 Volt. This should work reasonably well down to 5%-10% of full output and somewhere below that the panel open circuit voltage will start t drop off with falling illuminance. Full midday sun is in the order of 100,000 lux. So 5% =3D 5000 lux which is still "very bright" for artificial lighting but getting noticeably dimmer for daylight. Unloaded or lightly loaded voltage will remain "reasonably high" until levels fall to somewhere under 1000 lux (1% sun). Loaded levels when measuring current will probably be well off a straight line when loaded as above, but still useful. If you are prepared to do some curve tailoring (table or whatever) you can probably get highly useful results from full sun down to very overcast. Russell --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .