Agree 100%, I’m not in favor of any of these tweaks that report data that doesn’t exist 
I’ve just installed a UVN800 and I believe that it has to mounted vertically.
Max UV 3.0 yesterday & today in Eastern Spain.
Hi…
The forecast is about 5-6 UVI Barcelona had 4,9 UVI…
look here http://uv-saf.dmi.dk/Europe.html
stars…
That would make sense as I only stick the UVN800 in the garden which is shaded for part of the day.
I’m just waiting for a sloar panel to arrive for the Anemometer and then it’s up on the roof and install it there!
Hopefully I’ll also be able to understand sunshine hours by then.
I don’t think you will get very good sunshine hours using a UV sensor. As you can see from the graph the Solar Radiation is an estimate generated by multiplying the UV index by a fixed number. That is a very approximate number.
The sunshine hours calculation compares the measured (or in this case estimated) watts/m2 value to a theoretical maximum solar calculated for your latitude, longitude, and time of day as it would be measured by a “real” global solar radiation sensor. You set a threshold percentage above which it is “sunny”, eg if the measured value is 75% of the theoretical maximum, and WD counts the time the percentage is above that value and reports it as sunshine hours.
I’m sure there can be a fine line at times between getting the most out of a sensor and going too far…
There is measurable UV & solar at moderately high lattiudes in winter. The difficulty seems to be in usefully measuring it with relatively cheap hobby grade equipment… 
Have you tried the “temp sensor in a jar” thing?
Yes. 
As it happens I did I consider including in my message a comment to the effect that temp in a jar seemed to be a reasonably useful compromise but then decided to leave that out.
In due course when (if) I get a wire out to my sensor location (the pipe I aim to use for the underground wire run arrived yesterday) I hope to be able to compare temp in a jar, Hobby-Boards solar and a Breitling style PV sensor. A year or so’s comparative data for those might be interesting to study.
Oh the things that could be tried given the time (and money)… :lol:
Sounds like a good plan :thumbright:
Thanks for that info Niko - I’m still learning.
I was wondering what ET was but I guess from your message it is Estimated.
What type of sensor should be used for measuring Sunshine Hours?
ET is Evapotranspiration related to water use by plants and used for agriculture.
Sunshine Hours are correctly measured with a Campbell-Stokes Recorder or electronic equivalent, but those don’t come with weather stations. The Davis and some other stations have Solar Radiation Sensors available, they measure the full spectrum of solar energy received on a horizontal surface and WD uses that data for its sunshine hours calculation. I don’t think such a sensor is available for your station so the simplest couple of options that involve a little tinkering are:
“Temp Sensor in a Jar” = an extra temp sensor for your station located in a glass jar in full sun. WD calculates the solar from the temp difference between this sensor that is heated by the sun and the regular sensor that is shaded.
“1 wire Solar Sensor” = an electronic sensor that connects separately to your PC and can be read directly by WD. This requires more tinkering, running a wire to the sensor, extra software on the PC.
Either one can generate good results if you are prepared to put some time into the project.
So what’s the conclusion?
Tilt the UVN800 a bit, more facing the sun, of have it laying flat (horizontal) !?
The conclusion, mine at least, is that you can only get a very rudimentary estimate of global solar radiation from a consumer level UV sensor that has a limited viewing angle and only outputs UV index.
I was looking at putting an extra sensor in a glass jar, I even bought an additional sensor (THGR800) for my station(WMR200). I then read somewhere that the temp sensors were not lasting long in the jars before malfunctioning, so haven’t done it. I would still like to do it, so if anyone has more insight or special tricks to this. Would appreciate it.
I imagine that the biggest problem for a sensor would be moisture in the jar?
When I was using the WMR928 equivalent of that sensor in a jar I was careful to use the smallest jar that the sensor would fit in, to get the jar nice an warm and dry before hand, put some freshly dried out silica gel sachets in the jar along with the sensor, then use extra silicone sealant with putting on the lid to seal the whole thing up.
Now I use a “temp probe on a wire” type sensor, so the main electronics are no longer inside the jar. Even so I still off care to keep moisture inside the new jar to a minimum. So far I don’t think that OS have produced a WMR200 compatible “temp probe on a wire” sensor.
Looking at the pictures of NorcalDan and Jose’s temp to light conversions for the OS sensor it looks like it would be pretty easy to remote the temp sensor component. You wouldn’t want to extend it too far, but I don’t think extending it far enough to put it in a jar would be a problem.
I just received a shipment of cds sensors and in the coming weeks I will be working on getting my THGR810 outfitted to be outside and weather proof. I plan to remote the sensors about 6" from the THGR810. That should allow enough distance to get the THGR810 shielded from the elements and have the sensors in the sun with a diffuser for rain protection.
I checked out your sensor recently and think it is awesome. I don’t think I would trust myself to make such major changes to the sensor without causing damage. The sensors are very expensive in Kiwi dollars and isn’t helped by the shipping costs. I think the sensor in the jar will meet my needs for now.
It’s an easy modification. Not much chance of hurting the THGR810. I cut the temp sensor leads to the PCB so I wouldn’t have to solder on the board. I could easily remove the cds sensors and re-install the temp sensor.
Just found this thread and had asked the same question myself, Orientation of UV Sensor
Comparing the UV readings I was obtaining, compared to those by the BOM, it was 3-4 lower than it should have been. Tilting gives me readings that are directly comparable. Must be an issue with how the lens on top of the sensor collects the radiation. I plan to mount the UV sensor at around 52.5 degrees, the sun angle at the equinox, half way between minimum and maximum angles. Will see how that goes in the middle of summer in terms of accuracy / comparison of the UV reading.