what is the new light?

[Gordon-Loomberah]

I've looked at Cumulus help, and searched the forums, but see no mention of what the yellow/blank light is next to the UV index. I'm using beta 1.9.1, build 162.
I thought perhaps it was an indication of UV levels, but it seems not, as it is yellow now, with a UV index of 0. I'd be happy if someone can inform me

[steve]

Well, it's yellow, and it's in the solar section, so I thought it was obvious. It lights up when Cumulus thinks the sun is shining.

[Gordon-Loomberah]

Thanks Steve, I assume it is based on time, or maybe solar radiation level? Radiation is unlikely, since I suspect it was cloudy when I last noticed it was on (remotely logged on from work)

[steve]

It's based on the solar radiation level, as compared to the current theoretical maximum. See the 'solar' section in the station settings help. Note that you will need to adjust the values to get reasonable results. Note also that the Davis only updates the solar reading once a minute. Note further that the results will be worse at each end of the day.

[silverview]

"Note further that the results will be worse at each end of the day."

Steve, is there a way to minimize this effect? Cumulus is recording sunshine at the beginning and end of every day for my station (even on cloudy days with the sensor covered in snow).

Early and late each day, the monthly log file shows that the current theorectical solar max is zero, but the solar sensor is getting readings above zero. If the theoretical max is 0 and the solar reading is 1 w/m2, then sunshine is recorded. I have tried adjusting the transmission factor to either extreme. My lat, long, and elevation are correct. It seems the calculation for solar max places my location as being considerably further north, with shorter day length.

Thanks.

[steve]

Steve, is there a way to minimize this effect? Cumulus is recording sunshine at the beginning and end of every day for my station (even on cloudy days with the sensor covered in snow).

I think there will have to be a (configurable) 'cutoff' period for the start and end of the day where sunshine is never recorded. I added the sunshine hours facility because I wanted it for myself, having bought a solar sensor, but I haven't yet found time to experiment very much with the settings.

Apart from the beginning and end of the day, have you been able to find settings which give reasonable results?

[silverview]

I added the sunshine hours facility because I wanted it for myself, having bought a solar sensor . . .

And I bought the solar sensor because you added the facility.

I haven't really had a good opportunity to fine tune the settings. Since I installed the sensor, it has mostly been snowing and cloudy, or just cloudy. And the sun is so low on the horizon this time of year, it hardly counts as sunshine. That and the fact that I leave for work when it's still dark, and return home in the dark. I'll play with the settings if it's sunny this weekend.

[steve]

And the sun is so low on the horizon this time of year, it hardly counts as sunshine.

Yes, it's not the ideal time of year in the northern hemisphere to be trying to do any calibration. And I'm eight degrees further north than you

[Gordon-Loomberah]

It's based on the solar radiation level, as compared to the current theoretical maximum.

OK, I do not understand this, right now it is showing 25W/m^2, which is very dark and gloomy for the sun being so high in the sky, definitely not what anyone would call sunny (although maybe that counts as sunny in the UK! ) and Cumulus is telling me it is sunny.

I assume the default threshold value of 75% means that if the radiation level is 75% of what it might normally be when it is sunny (ie the theoretical max calculated by Cumulus), which in my case would be somewhere in the vicinity of 1000W/m^2 with the sun that high up. 25/1000 is 2.5% by my reckoning... bumping it up to 95% didn't make the light go off, although the radiation was then up to 33W/m^2.

Is my interpretation correct? and how exactly does the transmission factor come into it? I would think that a lower transmission value would allow "sunny" conditions at lower W/m^2 levels recorded by the VP2 sensor, to account for the sun shining through smoky/smoggy air, but I may be wrong there too.

I just cant see that with a threshold of 75% (or 95%) and transmission of 80%, it can say it is sunny with only 25W/m^2 with the sun high in the sky behind very thick clouds

EDIT: now at 6:30pm with only 5W/m^2, Cumulus thinks the sun is shining, The sun would be about 15 degrees above the horizon (but behind thick rain clouds), and the theoretical radiation must be some hundreds of W/m^2, so I think the calculation is broken somewhere.

[steve]

See previous threads for discussion of this. Which build are you actually using, 962? There was a bug where the longitude being used was negative when it should have been positive and vice versa, because of the non-standard way the library routine I use works. I can't remember which build fixed this. You can see the theoretical max values calculated by Cumulus in the log files.

Cumulus uses Ryan-Stolzenbach, see this page (the SolRad link) - http://www.ecy.wa.gov/programs/eap/models.html - for a spreadsheet which should give you similar figures to Cumulus. Basically, it calculates the theoretical max value at the edge of the atmosphere, and then applies the transmission factor to get the value at the surface. It then compares the current reading with that, using the percentage supplied.

[Gordon-Loomberah]

Thanks Steve, yes, using 962, will update and check the spreadsheet for comparison.