Converting Lux to W/m2

[steve]

The new Fine Offset station that I'm trying to add support for (which is reasonably easy as it is similar to existing Fine Offset stations) supplies a 'light' reading in Lux. The code that I already have, for Davis stations, uses the W/m2 reading supplied by Davis stations to try to decide whether the sun is shining.

I'd like to avoid writing new code to handle Lux, so I want to convert to W/m2. From the reading I've done, I see that the conversion depends on the frequency of the light, but an often quoted figure is that the equivalent for green light at 555nm, at the centre of the spectrum, is 683 Lux = 1 W/m2 and apparently other frequencies are much lower in power.

What I would like to know is: can I use this 683 figure to do the conversion and get a reasonable result, or would the result be meaningless? Is there an expert in the house?

[mcrossley]

As you probably know the relative 'strengths' of the wavelengths at the Earths surface varies as a function of the Suns elevation - hence Red Sun sets/rises. I have an algorithm somewhere for calculating the RGB centred wavelength values as a function of solar elevation. Also the detectors tend to much more responsive to blue light which is attenuated the most.

[philcdav]

hi teve.

As u say, frequency specific @ 555 nm

685 Lux= 1W/m2

[steve]

Thanks, chaps. So what do I need to do? Use the formula to work out the RGB components of the sunlight at a given time, then apply the Lux to W/M2 figure for each of those components to arrive at a total figure?

[captzero]

I saw this at: http://bccp.lbl.gov/Academy/workshop08/ ... re_Law.pdf

Converting Lux to W/m2
• There is no simple conversion….it depends on
the wavelength or color of the light.
• However, for the SUN, there is an approximate
conversion of 0.0079 W/m2 per Lux.
• Example: We read 75,000 Lux on our light
sensor. We can convert that reading to W/m2.
75,000 x 0.0079 = 590 W/m2

Not sure if its of any help

Cheers

[steve]

• However, for the SUN, there is an approximate
conversion of 0.0079 W/m2 per Lux.

Thanks very much for that. If I can get away with just doing that, that would be be brilliant. As the Fine Offset sensors presumably aren't particularly accurate for the price - compare the whole station price with just the sensor price for the Davis - then it may not be worth doing anything more.

[mcrossley]

Agreed - maybe someone can do a comparison between the FO and Davis so the 'factor' can be tweaked if required.

[avmourik]

Hi all

The FO stations also have the possibility of iving a readout of W/m2. Low and behold it uses the 683lux per watt factor.
As you all know, this is only for a frequency of visible light. So the station will give you a much lower reading than when the 0.0079 calculation factor is used, which accounts for all the electromagnetic radiation in relation to the visible light.
The point is, what do you expect to read, is it the total power of the sun on the earths surface, or just the visble light?

Some investigative dismantling showed that the PCB layout has been substatially altered. Als the lightsensor has a green filter fitted under the dome.

[steve]

The Davis stations measure "global solar irradiance", and it's on that basis that the theoretical max value and sunshine hours are calculated in Cumulus. So the conversion factor is an attempt to get a similar figure from a Fine Offset; I realise this is going to very approximate as best, as they are probably measuring completely different things.

[Werk_AG]

Hi

What formula was been adopted in Cumulus for Fine Offset stations conversion from Lux to W/m2 ?

685 Lux= 1W/m2 or 1 Lux = 0,0079 W/m2

Thanks

[steve]

What formula was been adopted in Cumulus for Fine Offset stations conversion from Lux to W/m2 ?

It multiplies the Lux value by 0.0079 to get a W/m2 value. This multiplier can be changed by adding a line to the [Solar] section of Cumulus.ini (with Cumulus stopped):

LuxToWM2=x

where 'x' is the value you want to use instead of 0.0079.

[Werk_AG]

Steve, many thanks for this info.

Regards