Archive of the former Yahoo!Groups mailing list: Homebrew PCBs

I did up a little program in scilab to add up contributions of a 1D
array of LEDs based on the intensity diagram from the BestHongKong
LEDs. You will have to download scilab to run it. Google for it if you
do not know what it is.

I am assuming that the radiation diagram is in W/m^2-sr or Watts per
meter squared per steradian, but values returned are just relative to
the what is in the radiation diagram.

For my case of a .5 inch grid spacing it looks like something around
1.9 inches looks optimal which is close to what Adam was recommending.

Disclaimer, I am not an optics expert so I could be doing something
wrong, but the results look reasonable.


//Simple 1D intensity calculator
//Use with scilab
//Change gridl,gridy,gridx to get different patterns
clear all;
gridl=12; //Number of LEDs
gridy=1.9;//Height above the board
//gridy=1.42;//Height above the board
gridx=.5;//Spacing of LEDs

//Calculate the intensity of a single LED based on
//angle
function result=getInten(rtheta)
//Angles step with intensity .1
//20 Degree LED/10 degree half angle
point=[0 5 6 7 8 9 10 11 15 20 90];
//40 degree LED
// point=[0 8 11 15 18 20 22 28 37 55 90];
inten=[1,.9,.8, .7, .6, .5, .4, .3, .2, .1, 0];
theta=abs(rtheta∗180/%pi);
// printf("angle is %i\n",theta);
prevVal=0;
currentVal=0;
if(theta==0)
result=1;
return;
end
for i=2:length(point)
currentVal=point(i);
if currentVal>theta then
dtheta=currentVal-prevVal;
dinten=inten(i)-inten(i-1);
result=inten(i-1)+(theta-point(i-1))∗dinten/dtheta;
return;
end
prevVal=currentVal;
// printf("%i\n",point(i));
end
result=0;
endfunction

//Calculate total intensity at a point from
//an line of LEDs
function result=getTotal(x)
result=0;
for i=1:gridl
result=result+getInten(atan((x-gridx∗i)/gridy));
end
endfunction

x=-3∗gridx:.05:gridx∗(gridl+3);
value=1:1:length(x);

for i=1:length(x)
value(i)=getTotal(x(i));
end

scf(1);
clf();
text=sprintf("%i LEDs %.2f spacing %.2f height",gridl,gridx,gridy);
plot2d(x,value,leg=text);