Math behind

RMF

New Member
Thanks SupraSPL, that was really helpful!

Still didnt grasp what the SPD graph really means.

I mean, if a certain wavelengh has for example 100%, is it 100% of what?

I apologize again for the basic question...:oops:
 

alesh

Well-Known Member
Thanks SupraSPL, that was really helpful!

Still didnt grasp what the SPD graph really means.

I mean, if a certain wavelengh has for example 100%, is it 100% of what?

I apologize again for the basic question...:oops:
100% marks the wavelength with the maximum output. The other values are relative to this one.

It would make more sense if the SPD were normalized to have integral of 1. Then the y values would be ratios of output of each wvl to overall output. For example the peak wvl in the chart below is 454nm with y value of 0.66%. That means that 0.66% of the whole output is being put out at 454nm.
example.jpg
 

CanadianONE

Well-Known Member
100% marks the wavelength with the maximum output. The other values are relative to this one.

It would make more sense if the SPD were normalized to have integral of 1. Then the y values would be ratios of output of each wvl to overall output. For example the peak wvl in the chart below is 454nm with y value of 0.66%. That means that 0.66% of the whole output is being put out at 454nm.
View attachment 3482264
One quick question with the math excel file, The only numbers we must change are column A & B which we import from digitizer once we digitize a chart? All other numbers remain untouched.
 

alesh

Well-Known Member
spectrum imported to the silvania Tool...all ya need

CCT for this SPD is 3500K
LER is 276 lm/W
and we have 4.83 µmol/W

View attachment 3486098
And now who should we believe...
The SPD chart is from the CXA3070 data sheet (3000K 93CRI although CXB3070 3000K 90CRI has the same). There isn't any 3500K chart in the data sheet.

And yes I'm getting CCT of 3400K as well, LER 276lm/W. There's some discrepancy, though, as I'm getting 4.91 µmol/J from the 380-780nm range or 4.51 µmol/J from the 400-700nm range.

With Veros v2 I'm always getting lower CCT than a chart claims. Such as 2532K for 3000K chart or 3278K for 4000K chart. The calculation works reasonable well for standard-CRI Crees.
 

churchhaze

Well-Known Member
Thanks SupraSPL, that was really helpful!

Still didnt grasp what the SPD graph really means.

I mean, if a certain wavelengh has for example 100%, is it 100% of what?

I apologize again for the basic question...:oops:
It's because it's relative SPD, not absolute. For absolute SPD, the area under the curve is total power output.

Edit: just realized this was answered...
 

CanadianONE

Well-Known Member
Yep Excel. Insert>Chart>Scatter>Smooth line in Excel 2007.
Thanks I am playing around with it now. Excel 2013 and I have a chart just trying to figure out how to only show the data I want and not have a blank section without data plotted for example its showing my wavelength axis from 0 to 700 but I want to omit the blank section from 0 to 400 and only show 400-700. I will get the hang of it eventually lol
 

alesh

Well-Known Member
Thanks I am playing around with it now. Excel 2013 and I have a chart just trying to figure out how to only show the data I want and not have a blank section without data plotted for example its showing my wavelength axis from 0 to 700 but I want to omit the blank section from 0 to 400 and only show 400-700. I will get the hang of it eventually lol
Right click on the axis > Format Axis
 

bggrass

Well-Known Member
Hi alesh, and all, when you run the integrals posted at the beginning of this thread, the PAR values and umol/s values that you get, are those the values at the source of the light? i.e at the cob's surface. Do we setup another integral with the par or umol values inside to get the values at 12" , 18", 24"?
 

alesh

Well-Known Member
Hi alesh, and all, when you run the integrals posted at the beginning of this thread, the PAR values and umol/s values that you get, are those the values at the source of the light? i.e at the cob's surface. Do we setup another integral with the par or umol values inside to get the values at 12" , 18", 24"?
No these values are only for converting luminous flux to radiant energy and/or photon flux. You don't calculate photon flux density this way.
 

speedyganga

Well-Known Member
Hello, Great share alesh, thank you !!
You calculate efficiency of each emiters, But what about efficacity ?
I mean, efficiency from a 660nm SSL @ 350mA is lower than a CXB3590@350mA but I guess we have to take into acount spectrum efficiency if we want to really compare them ?!
how can we calculate it ?
 

alesh

Well-Known Member
Hello, Great share alesh, thank you !!
You calculate efficiency of each emiters, But what about efficacity ?
I mean, efficiency from a 660nm SSL @ 350mA is lower than a CXB3590@350mA but I guess we have to take into acount spectrum efficiency if we want to really compare them ?!
how can we calculate it ?
This thread isn't about efficiency of emitters themselves but rather that of spectra.

It's very hard to compare different light sources from the horticultural POV. The current consensus is that the most accurate measure is photon output meaning that a better light sources produce more photons per energy dissipated and each photon - regardless of its wavelength - has the same weight.
If we're to add photon weighing by its wavelength we need a function that would tell us how to weigh them [ie McCree's curve]. Could be done in a very similar way as lm->W or W->µmol/s conversion.
 
Top