[41south]

Ah hah 

Yes that indeed all makes sense and I did wonder from the start if it was a sensor issue of sorts. I will buy in another one (or 2  ) and experiment further. I also have a couple of VEML6075's on the way, which one assumes will be more accurate

Thanks for the explanation Werk 

All good fun!

[werk_ag]

Its all on the code...  

Another hint, before placing all the guilt on the sensor...
If possible try to measure the voltage at ADC0 input, then verify if the raw reading from ADC0 is consistent with the voltage you are reading on the multimeter using this formula:

voltage (in Volts) = (adc0_reading * 1.1 / 1024)

If the values don't agree for a significant amount, there is a possibility that the internal voltage reference of that particular Arduino Nano isn't accurate (by specs they can be between  1.00V -- 1.20V) which obviously have impact on the readings.
Think now you have all the bits of what is involved.

If I were you, I would just try to calibrate the sensor by comparing the readings with other sensors in the vicinity, once it has the coverage you want to use, then it is easier to play with only the value to assign to the UV_Transmission.

[41south]

Yes I agree to put it under it's glass or some such cover and then experiment more. I did that yesterday in the thought that the sensor may work better in it's 'low range' and did have some success in getting more accurate readings. I'll check the reference voltage as you suggest as well.

The interesting part will getting it calibrated across it's entire range of readings. I guess that is one of the challenges of using analogue devices.