Radiation, ET & UV?

[VirtualSky]

Hi Steve,

As Cumulus is running that perfectly, I wanted to know if there was an easy way for you to add the functionality of reading the Solar radiation, the UV ant the ET from VP2...

Cheers,

Julien

[steve]

It wouldn't be too hard. I don't have the relevant sensors myself (but I'd like a solar radiation sensor so I could do hours of sunshine). The difficult thing would be to work out where to display the values. If I was to add them to this version of Cumulus, I think the deal would have to be that I just display the values and create web tags for them. They wouldn't get stored in the log files, all-time records, graphs etc. No sunshine hours calculations etc, just the raw data.

Steve

[VirtualSky]

Steve,

That would be great. If you get the raw data, the sunshine duration can then be easily calculated through a PHP script. And even be discplayed graphically.

Cheers,

Julien

[steve]

I'll just have to work out where to put it on the main display, and whether to allow it to be hidden for people who don't have those sensors. I want to get 1.8.3 out first, but I'll look at it after that.

Steve

[VirtualSky]

Great!

[broadstairs]

Steve,

That would be great. If you get the raw data, the sunshine duration can then be easily calculated through a PHP script. And even be discplayed graphically.

Cheers,

Julien

I'm not sure its as easy as you think. I have looked into this a lot and the only real way to estimate the sunshine hours is to compare the actual reading from a VP solar sensor with an expected max value for the correct latitude/longitude and time of day, which means doing this calculation every minute of every hour of possible sunshine between sunrise and sunset, you need to then set a threshold value (normally a %) over which is considered sunshine. Trivial its not. The Davis VP solar sensors are not of the kind where you can simply say over a certain w/m² value is sunshine, it does not work.

Stuart

[steve]

Weatherlink does that, though, doesn't it? Anything over 100 W/m² it regards as 'sunshine', I think?. Does it not come up with very good results?

Steve

[broadstairs]

Weatherlink does that, though, doesn't it? Anything over 100 W/m² it regards as 'sunshine', I think?. Does it not come up with very good results?

Steve

In my opinion that is complete rubbish. A Davis solar sensor measures global solar radiation on a flat surface which is the total of direct and reflected radiation. The measurement of 100w/m² for sunshine is only valid I believe if you consider direct radiation only, but a Davis sensor does not measure this. If you check what a solar sensor on a VP reads you will find bright sunshine in the early morning is considerably under 100w/m². I simply do not undestand how weatherlink can do this on raw VP solar data.

Stuart

Edit: I've just checked my solar data for 29th October when it was raining solidly all morning here and the raw solar value from my VP ranged from 5 just after sunrise to constantly between 100 and 180 from 9:28am until about 11:49 after when it brightened up and the sun did come out for a while. So during this time of constant rain Weatherlink would have accumulated nearly 2.5 hours of bright sunshine!!!!!!!!!!!!!

[steve]

I believe the thinking behind the Weatherlink calculation is that 120 is the standard definition of sunshine, so they subtract 20 to allow for reflected radiation. It does seem a bit arbitrary, particularly considering your example.

Steve

[broadstairs]

Steve take a look here for my maximum solar radiation script. It will calculate your max possible radiation in w/m² for any given location (not good for very high latitudes above about 70°) and you will see an approximation of the maximum possible value for every minute, You can see if you change the date the very different values between summer and winter, and bear in mind this is for global solar radiation on a horizontal surface. To measure direct radiation accurately you need a solar sensor shielded from all reflected radiation and one which accurately tracks the sun across the sky. Our sensor on a VP dont do this and often may even exceed the max expcted value shown in this script because of reflections amongst other things from clouds. So you will see their values are to say the very least very very inaccurate and that is being polite. I'd have expected a company like Davis to come up with a much better solution software wise.

You can also see a US Government site here which does a similar calculation, just set date lat & long and select Extraterrestrial Global Horizontal Solar Irradiance (W/m2).

Stuart

[steve]

Wow. I'll be sure to consult you when I do the sunshine hours calculation!

Steve

[steve]

Beta version with Solar Rad, ET, and UV coming very soon. Will probably display garbage

Steve

[Paul C]

Steve
I hope you can come up with something. I now have my solar sensor installed as of 0900hrs this morning. As you know I have a traditional sunshine recorder but getting hold of the card is increasingly difficult and expensive if I had to buy them myself, so I was hoping the solar sensor would be a way of overcoming this problem.

The supplier of the sensor provided me with a link to some software under development that claims it can convert solar data into bright sunshine data
www.prodata.nildram.co.uk

I have downloaded the programme this evening and eventually managed to produce some data - it says I had 4.2hrs of sunshine today. Will be interesting to see what the recorder shows in the morning.

Perhaps Steve you could look at how this programme comes up with its end solution, may be able to incorporate into Cumulus. Of course as I still have an active sunshine recorder I may be able to assist in any trials.


(Graph needs some work on it to remove values I dont need)

[steve]

Perhaps Steve you could look at how this programme comes up with its end solution, may be able to incorporate into Cumulus. Of course as I still have an active sunshine recorder I may be able to assist in any trials.

Hmm. It says: "Sophisticated algorithm for calculating Bright Sunshine duration from solar radiation data". It doesn't sound as though I'll be able to work out what it's doing just from the output. I could ask John, I suppose, but as it's proprietary I don't suppose he'd be willing to tell me the algorithm.

I've found this of John's : http://www.elyweather.co.uk/sunshine1.aspx which looks interesting, but I haven't got time to read it just now as I'm going to listen to the Elbow concert on the radio

And of course, Stuart is our local expert in all this so we'll need to get him involved. But I don't think I'll be doing anything until Cumulus 2.

Steve

[hungerdunger]

but I haven't got time to read it just now as I'm going to listen to the Elbow concert on the radio

Steve - if you've got Freeview (or possibly Sky), you'll find the Elbow concert is also in vision on one of the BBC's Red Button channels at the moment (on a continuous loop, but I don't know for how long)