Photosynthetic Spectral Quality Calculator (PSQC)

CCCmints

Well-Known Member
I'm having trouble understanding how to digitize the spectral distribution. If anyone could help I'd greatly appreciate it.
 
Last edited:

CCCmints

Well-Known Member
If you could explain the goal perhaps someone can be of more assistance?
I'm trying to build COB fixtures equivalent to 1000W HPS DE. On the COB efficiency Spreadsheet the latest data I could find for Vero 29 and CXB3590 rated CXB3590 @ 49.71% efficiency when ran @ 2.1a. That would allow me to effectively match (nearly) the PPFD of a 1000W HPS DE. However, a user posted that this data was outdated and showed graphs where the CXM22 is better now. I'd just like to learn how to use this calculator so I can figure out exactly what is the current best COB to use @ preferably 80(+/-) watts since I want either 3x3 rows or 4x4 for even coverage over a 4'x4' area. If its smarter to do 4x4 then I'd like to go with 16 COBs so I'd be reviewing my options for that as well.
 

CCCmints

Well-Known Member
Why do I need this if I can simply use alesh's excel file without the need of installing additional software?
https://www.rollitup.org/t/math-behind.868988/
Well, in my situation, I think a calculator would help a lot. Maybe you can also do this in that excel file, which Ganjineer linked, but if you want to compare a large amount of COBs to determine the best one for your situation then having a program probably makes that process a lot easier.
 

ttystikk

Well-Known Member
I built modules that I can use singly or in arrays. Each one is 4 x CXB3590 chips driven by a Meanwell HLG185H-C700B, for a total of 225W each. Three of these modules would do well for your intended purpose and if/when circumstances change they can easily be reconfigured to work separately.

You won't save much by getting a bigger driver instead of several small ones but this way you will never be in the dark.

There are more modern chips available now with slightly better performance at much lower cost but the basic approach is still sound.

I hope that helps.
 
Last edited:

CCCmints

Well-Known Member
I've been trying to digitize the spectral distribution and I'm definitely doing something wrong. When analyzing the CXM22 the calculator tells me QER = 4.72414 umol/j and LER = 355.463 lm/w

That can't be right...I'm using WebPlotDigitizer on the chart posted by luminus for the CXM-22. I pick the color of the line I want to use as the foreground color and trace it with the pen tool to create the data I put into a spreadsheet for the calculator and the numbers are off. If anyone has done this before I'd appreciate some help. Once I figure out how to do this I'll do a bunch of the COBs (even by request) and post em' here.
 

CCCmints

Well-Known Member

CCCmints

Well-Known Member
So what do you guys do to get the flux values for the COBs when ran at different currents? And for relative power spectral distribution for temperatures that aren't graphed in the data sheets? Would you need to purchase that COB and test it yourself with a meter or is there another way?

Here's the CXB3590 (CD) 36v 2.4a (86.4w)

QER = 4.82549 umol/W/s
LER = 326.78 lm/W
Efficiency = 40.3772%
PAR = 34.8859W
PPF = 168.341umol/s

 
Last edited:

ttystikk

Well-Known Member
So what do you guys do to get the flux values for the COBs when ran at different currents? And for relative power spectral distribution for temperatures that aren't graphed in the data sheets? Would you need to purchase that COB and test it yourself with a meter or is there another way?

Here's the CXB3590 (CD) 36v 2.4a (86.4w)

QER = 4.82549 umol/W/s
LER = 326.78 lm/W
Efficiency = 40.3772%
PAR = 34.8859W
PPF = 168.341umol/s

We generally used Cree supplies charts for current droop calculations.

Can you label the Kelvin temperature of the chips in the graphs? It does make a substantial difference in output.
 

CCCmints

Well-Known Member
We generally used Cree supplies charts for current droop calculations.

Can you label the Kelvin temperature of the chips in the graphs? It does make a substantial difference in output.
Yes of-course I thought I had done that. The CXM22 was 3500k and the CXB3590 was 3000k.

Could you elaborate a bit on droop calculations? I have no problem buying and physically testing a bunch of different COBs. I want to get this shit right.
 

ttystikk

Well-Known Member
Yes of-course I thought I had done that. The CXM22 was 3500k and the CXB3590 was 3000k.

Could you elaborate a bit on droop calculations? I have no problem buying and physically testing a bunch of different COBs. I want to get this shit right.
The chips operate more efficiently the less current that's passed through them.

The relevant charts are around but buried.
 

HydroLynx

Well-Known Member
I can't get any data file I made to open, what am I missing? Using WebPlotDigitizer and exporting the data file via .csv and .xlsx file names, but the calculator always error dings the file selection, not the example file, yet my format is the same.
 

CCCmints

Well-Known Member
I can't get any data file I made to open, what am I missing? Using WebPlotDigitizer and exporting the data file via .csv and .xlsx file names, but the calculator always error dings the file selection, not the example file, yet my format is the same.
I don't think you can export the data from WebPlotDigitizer and put that file directly into the calculator. Try sorting the data by X and making your own excel file, then load that into the calculator.

Make sure the A and B column is only filled with the data. No headers like "wavelength" or "relative intensity". Also make sure you rename the sheet of the excel file to "Spectral Distribution", not the file itself.
 
Top