On 16/07/2015 17:19, JohnD wrote:
"Brian Wakem" wrote in message ...
Really? The highest I recorded in the whole of December 2014 was 160w/sqm
and I'm pretty sure the sun came out at some point.
I asked in this newsgroup a few months ago if it was possible to
mathematically determine if the sun was 'out' using the w/sqm along with
time of day and time of year but I didn't get far.
=======================================
From memory, I think you'll find that the threshold is actually
120W/sqm _direct_ irradiance. I'm guessing you were probably measuring
with a sensor that records _total_ (ie direct plus indirect) irradiance,
which is obviously something rather different.
There are well-established algorithms for testing for bright sunshine
(ie if the sun is 'out'), which also need lat/long and TZ as inputs of
course if you are using a total irradiance sensor. It can only be an
approximate or rough estimate because of the limitations of using total
irradiance, but it works tolerably well. It's not possible to summarise
these methods in a short post here, but if still interested you might
want to Google a bit harder. To implement the algorithms in a script or
program does require some understanding of the trig functions involved.
John Dann
www.weatherstations.co.uk
Ha what you need is Gorzynski Normal Incidence meter. I used one of
these at Kew Observatory when I was there in the late 60s (it was still
there in the early 70s at least). It was a solarimter pointed at the sun
and had a circular disc (maybe more) with holes in. You pointed it at
the sun and that's all it measured. It was driven by clockwork (had to
wind it every day). When you did you had to unwind the signal cable. It
didn't move with the sun's height from day to day, so each day you had
to ensure that it was pointed straight at the sun. Interesting when the
sun came out after a few days of cloud.
But then KO had lots of strange meteorological instruments never seen
anywhere else.
I have a photo of one but I've not yet scanned it. Googling for it got
mentions but no pics.