parabolic reflector visualizer

For those of you whom it might interest, here is a link to a
little matlab script that I wrote which calculates the result
of a luxeon HD in a parabolic reflector.
Reflector dimensions are adjustable, as is the position of the emitter.
The program shows a "beam-shot" look-alike, and the intensity distribution at 5m.

To install it, download

this
executable called 'compiled_reflector2', and
this
info file.
To run it if you don't have matlab6.5 on your computer, you need the matlab run-time graphics library
which my program requires (the library is self-installing)

Any comments, recommendations, etc are appreciated!

Olivier

I got the error message:

ERROR: Can't open file compiled_reflector2.fig

EXITING

What's going on?

oops, sorry, you do indeed need the .fig file: it's
here
.Put it in the same directory as the exe.

Another remark (once you get it to work /ubbthreads/images/graemlins/crazy.gif ) :
the intensity values are not "physical"; I might try and get around to it some time but it's not the case yet.

Olivier

Whenever the download for the matlab runtime library reachs 2 + megs, the connection with the server will be resetted... I think there's some problem with my connection.

But here's the really weird thing. When I open up what's left of the 2 megs I downloaded, I get some war craft 3: Frozen throne things. /ubbthreads/images/graemlins/eek.gif

!!!

It's working now. I think the server sent the wrong file to me. /ubbthreads/images/graemlins/eek.gif

Another thing I forgot to say: the graph axis units for the beam shot are in cm, likewize for the intensity distribution.

Hey Nerd, sounds strange, the server sending you the wrong file /ubbthreads/images/graemlins/confused.gif

I got the install exe., compiled fig and compiled exe.

"Missing dll file libmmfile.dll". No Joy.

- Jeff

Waiting for peoples findings with this! /ubbthreads/images/graemlins/popcorn.gif

IceBreak

You will need to configure the PATH variable. In Win2000 system, you right-click on the "My Computer" icon, choose properties, then go to "Advanced" tab. Click on "Environment Variables", then "User Variables", then "New". Type in or choose "PATH" as the name, and the installation directory as the value (including the bin\win32 part). Then click OK/OK and log out. Re-login, and things should work.

dukeleto

The first input box was covered by the drop-down menu. Can you spread them out a little---probably you have much higher screen resulution than I.

[ QUOTE ]
LED mods As Small As Possible said:
You will need to configure the PATH variable.

[/ QUOTE ]

Sorry to ask but, can anyone refresh my memory on how to do this with W98SE. I'm just a novice PC user but a great PC modder. /ubbthreads/images/graemlins/grinser2.gif
I know, I know, my OS is antique stuff.

thx

Pfft! I'm running win95 on a 266mhz 160megs ram 8gig hd with ~20megs free space with an addon on 12meg voodoo2 card.
And yes, I can run "win98 only" apps on it. /ubbthreads/images/graemlins/grin.gif

Olivier,

Very nice program, now if I can only figure out what to do with it.

BTW, thanks to LED Mods for showing how to delineate the path needed for the .exe to work.

I also found that the .fig file needs to be dragged into the bin folder to make the program work.

Any possibility of being able to choose the diameter of the beam at various distances? Say I wanted to know what the beam width would be at 50 meters instead of 5 meters?

If I remember correctly, in win95/98 you will need to edit the c:\config.sys to add the PATH variable---most likely there will already be one, just add a ";" at the end and add your new path.

LASAP,
I've updated the fig & .exe files, does the layout work better for you?

The zoom function also should work better: zoom in by clicking on the left mouse button or by dragging to outline the zone you want to zoom. Zoom out with the right button.

oeo2oo: I'll try & work on it.

Another amusing thing: on the intensity plot, you can see that the "hotspot" actually has a small hole of darkness.
This is physically understandable: the light hitting the very central zone has come directly from the emitter, without touching the reflector, & therefore it is still divergent: its intensity will decrease with distance.
On the other hand, the reflected light is perfectly parallel and its intensity stays constant with distance.
This hole disappears when the emitter is raised VERY slightly on top of the focal point.

I don't know whether this effect is observable in the non-ideal "real-world"?


Olivier