[hornychz]

We will continue... 

Hello,

I'd like to tell you something about my experiments ...

I've tried about 10 various materials. Two of them are "release candidates" - front CD cover (but only that very thin) and slide/glass from the old EEPROM:
 

I have many many troubles with my measurement, because of the very bad weather at our country now But tentative "research" findings are already clear. The transmission of the used materials is nearly linear - but it applies only for intensity of sun radiation. I have no option to identify UVA or UVB part of the sunshine. And I haven't had option to do enough measurements, so results are only approximate:
 

Another possibilities to build UV sensor cover is for example:

Would these do as a cover with the sensor mounted in, for example a , piece of 22mm conduit pipe with this as a lense?
http://www.ebay.co.uk/itm/Watch-Glass-Ac...5d5563789b

Or using any filter from Hoya U-340 glass for 365 nm UV-pass filter, for example:
http://www.ebay.com/itm/Hoya-U-340-37mm-...43dd21bbbb

I think it would be the best solutions but not cheap enough ...

But the best thing is that there is possibility to prepare (nearly) universal RX module code, with a paramater "transmission". Our formula then is:

[edit]This text isn't correct now, read more, please ...

IndiceUV = ((((ADC1 * 1 / Transmission ) - 135.0) * 11.0) / 1024.0) ;

[/edit]

Transmission measurement is very easy for everybody. Requires only to buy 2 sensors UVM-30A (not only one) and simultaneously do two measurement, with cover and without cover. The accuracy specified for UVM-30A is +-1 UVI, we have to think about it a do some corrections when the both sensor are not exactly the same.

I will buy that 365 nm UV-pass glass, when it arrive here, I'll publish another measurement. Maybe it will be summer in that time at us too ...