MaxSolarRad (from pyephm) is not alligned with Solar-Rad from my Weatherstation

Johannes Ebner

unread,

Oct 12, 2019, 11:37:00 AM10/12/19

to weewx-user

Hi,

I have yesterday installed pyephm on my Raspberry and enabled maxSolarRad.

I can see it already in my graphs, but the value is much below what is coming from my weather station. What could be the reason?

Br,

Johannes

vince

unread,

Oct 12, 2019, 3:36:53 PM10/12/19

to weewx-user

On Saturday, October 12, 2019 at 8:37:00 AM UTC-7, Johannes Ebner wrote:

I have yesterday installed pyephm on my Raspberry and enabled maxSolarRad.

I can see it already in my graphs, but the value is much below what is coming from my weather station. What could be the reason?

What kind of weather station ?

Many provide very incorrect values for radiation and brightness etc.

Johannes Ebner

unread,

Oct 13, 2019, 2:12:08 AM10/13/19

to weewx-user

Its a WS3080

Br,

Johannes

gjr80

unread,

Oct 13, 2019, 7:16:47 PM10/13/19

to weewx-user

Hi,

Your station has a sensor that reports illuminance in lux and the fineoffset driver approximates solar insolation using a simple linear multiplier. From the comments in fousb.py:

Illuminance and Radiation 
The 30xx stations include a sensor that reports illuminance (lux). The 
conversion from lux to radiation is a function of the angle of the sun and
altitude, but this driver uses a single multiplier as an approximation. 
 
Apparently the display on fine offset stations is incorrect. The display 
reports radiation with a lux-to-W/m^2 multiplier of 0.001464. Apparently 
Cumulus and WeatherDisplay use a multiplier of 0.0079. The multiplier for 
sea level with sun directly overhead is 0.01075. 
 
This driver uses the sea level multiplier of 0.01075. Use an entry in 
StdCalibrate to adjust this for your location and altitude.

Let's not start a debate about the validity/accuracy of deriving solar insolation from illuminance, I will let you google that, but you can see there is some variation in multipliers used.

Bottom line is it is an approximation and if you want to vary the multiplier used you can (effectively) do so via a StdCalibrate correction.

Gary

Xant

unread,

Oct 17, 2019, 12:10:22 PM10/17/19

to weewx-user

Gary is correct to not initiate an Insolation vs Illuminance debate... been there, done that.