I have a UV sensor for a WMR200. However, it appears that it doesn’t work quite correctly.
Currently and for the last couple of months, during winter I am getting no UV readings at all, even in the middle of the day, sun directly overhead. I have the unit mounted vertically, sensor on top pointing towards the sky of course. In summer, works, picks up the signal and it varies over the entire day. At the moment, according to the BOM it should be 2-4 during the day, but get zero.
Also, in summer, reading maxes out at (from memory, don’t have records handy) at about 10. However, again the BOM readings were around 14.
I was starting to think that may be vertical isn’t the best orientation, may be, like with solar panels, it should be on the same angle as the latitude of the location.
Is this the case?
Sensor is fine, nice and clean, no bird presents, dust etc.
It’s a very inexpensive device and doesn’t have the wide viewing angle necessary to read correctly at low sun angles. Other users have posted on here that they get more acceptable results with it tilted towards the sun (North for you).
Thought that might be the case. Thanks 
Think I will move it down off the pole and start playing with some angles at a better height to see how things go.
Just an idea, how about the use of an LED lens to help collect the light better over a bigger angel? Has someone tried that?
The problem is that almost all plastics, and most glass, are pretty effective at blocking UV, so you would need something like quartz glass to make a UV lens, but what you actually want is a diffuser, not a lens, so look for a nice piece of milky quartz 
Yes, UV absorption definitely an issue, forgot about that.
Re the lens, remember you are using it back the other way, instead of using it to spread light out over a wide angle, you are using it to collect light over a wide angle and focus it on the sensor.
True, I imagine there is some complex lens solution, but all the solar radiation sensors that I have seen use a diffuser to get the correct cosine response.
Thanks for the help 8)
Got a few things to try out now, will report back once had a go.
But, I guess it’s an interesting question just what the optical response on a UV sensor should be. The global solar “radiation incident on a horizontal surface” would be good for roasting by the pool, but…
Had a play with the UV sensor, looks like putting on an angle is the easiest compromise. For my latitude, 37.5 degrees, minimum and maximum sun angle is 29 and 76 degrees, at at the equinox is 52.5 degrees, so will go for that angle I think.