DIY COB LED Calculator

bggrass

bggrass

Well-Known Member
cxb3590cd36v.PNG IMG_3263.JPG
I was playing with some online tools trying to figure out the amperage formulas and I kinda noticed that there is a difference between the lumens per watt in the online cree tool and Supras spreadsheets. If you look at the lm/w at 1.4A it's about 18 lumens difference, per watt, both estimates at 50*C. Does anyone have any explanation to it, am I missing something? I think that it's always been an estimate, so maybe that's why the difference...?
 
CobKits

CobKits

Well-Known Member
its been discussed. has to do with both the wonky way PCT reports Tj vs Tc, and also im pretty sure supra used midpoint of bin flux where PCT lists minimum bin flux.

Hey bg, any chance of adding the other citi colors? all the formulas are in the second (hidden) worksheet, which is accessible in the unprotected file floating around
 
bggrass

bggrass

Well-Known Member
CobKits said:
its been discussed. has to do with both the wonky way PCT reports Tj vs Tc, and also im pretty sure supra used midpoint of bin flux where PCT lists minimum bin flux.

Hey bg, any chance of adding the other citi colors? all the formulas are in the second (hidden) worksheet, which is accessible in the unprotected file floating around
Click to expand...
So even if I figure out the formulas, the numbers will come out different than in the spreadsheets that everyone has been using all along. I think I got a good lead yesterday. The numbers were coming off by +-0.2 lm/w for the cree range 1.4A - 3.6A which is pretty minimal.

I'm not sure where to look for that file you say. I don't visit the rest of the forum much anymore.... Post it here I'll look at it.

Edit: Also, the cree online tool does not have an option to select which color the of that bin are the lm/w. Am I to assume that it is the same lumens per watt for every color of that bin?
 
Last edited:
CobKits

CobKits

Well-Known Member
well different cree colors have different bins
so for top bin
3000K = CB
3500K = CD
4000K 70 CRI = DB
4000K 80CRI = CD
etc

citi doesnt bin like that but their simulators calc from presumably empirical formulas
 
bggrass

bggrass

Well-Known Member
JorgeGonzales said:
https://www.rollitup.org/t/math-behind.868988/

I use a program called GraphClick, for OSX, 401 points between 380 and 780nm, etc. The rest is explained in that link, and @alesh has his spreadsheet uploaded for after you extract your SPD.

Actually, he has a csv uploaded too with all his digitized data, but I had to do all the Citizen stuff myself.

Not sure what numbers you need for 1750, but the simulator from Citizen certainly has that data covered. http://ce.citizen.co.jp/lighting_led/dl_data/sim/CITILED_SelectionSimulator_Ver5_P3728_0616.xlsx
Click to expand...
@JorgeGonzales I looked at that spreadsheet, is there anywhere in the sheet where the formulas are written? The formulas that take the amperage and give the lm/w and voltage for a given amperage. I tried to make sense of it but it might take a deeper look to figure it out... I thought I'll ask, if you know...
 
researching

researching

Well-Known Member
So the input for amps is the MA rating for the driver correct? So for example an HLG-240H-C700B input would be .7 correct?
 
bggrass

bggrass

Well-Known Member
Alright..... I made a new calculator that calculates the numbers for the citi cobs. Thanks to @JorgeGonzales and @CobKits for the input and help with it. The link is in my signature. One thing to notice is that this calculator takes any number for amperage. Enjoy it.

(if something comes out wrong in the numbers post a picture here. I haven't done extensive testing to see if it is all working right)
 
coreywebster

coreywebster

Well-Known Member
bggrass said:
Alright..... I made a new calculator that calculates the numbers for the citi cobs. Thanks to @JorgeGonzales and @CobKits for the input and help with it. The link is in my signature. One thing to notice is that this calculator takes any number for amperage. Enjoy it.

(if something comes out wrong in the numbers post a picture here. I haven't done extensive testing to see if it is all working right)
Click to expand...
Bloody awesome!!! Your a goddamn sexual tyrannosaurus!
 
coreywebster

coreywebster

Well-Known Member
@bggrass Some issues comparing the cob calc and the citi calc, when inputting 1825s and same data(except Tj which I didn't know) get a difference between PPF which is understandable because of Tj I assume but look at the PPFD.

cobcalcscreenprnt.jpg
 
Last edited:
bggrass

bggrass

Well-Known Member
coreywebster said:
@bggrass Some issues comparing the cob calc and the citi calc, when inputting 1825s and same data(except Tj which I didn't know) get a difference between PPF which is understandable because of Tj I assume but look at the PPFD.

View attachment 3762699
Click to expand...
Good catch. The way the old calculator works, the umol value is fixed and it gets applied only to the par watts. The new one the umol is dynamic, it changes based on efficiency, which in turn depends on Kelvin, Celcius, amperage and voltage. I fixed it and I changed the one in the dropbox. The cobs in the old version are the 3000K80Min ones. I just checked with the excel file and the lumens and voltage come out the same.... The link in my signature for the update.
 
tomate

tomate

Well-Known Member
Now when I hit the Calculate button, I get an error telling me I have to enter numbers in the input text fields.
But all the input fields are filled.
 
coreywebster

coreywebster

Well-Known Member
bggrass said:
Good catch. The way the old calculator works, the umol value is fixed and it gets applied only to the par watts. The new one the umol is dynamic, it changes based on efficiency, which in turn depends on Kelvin, Celcius, amperage and voltage. I fixed it and I changed the one in the dropbox. The cobs in the old version are the 3000K80Min ones. I just checked with the excel file and the lumens and voltage come out the same.... The link in my signature for the update.
Click to expand...
Yet again you have excelled yourself fine chap.
Seriously though your work is very much appreciated.
 
You must log in or register to reply here.
Top