I’m interested in modeling an OLED display that’s displaying the letter P. The brute way to do this would be a series of lambertian point sources that form that shape. Is there a better way? Maybe with a dll or a CAD object?
Best answer by Angel MoralesView original
Thanks for posting! Were you specifically wanting a DLL approach/solution? Probably the quickest way would be to use a Slide object and load in a BMP/JPG of a letter, and illuminate it from behind with a Lambertian source.
Alternatively, you could also define a User Defined Aperture and probably construct the letter of interest using various LIN and ARC commands. It would be a bit of work to define, but then you wouldn’t be loading in an image file. I guess the trade-off would be knowing what resolution best suits your modeling of the letter in an image file, and if it’s quite high resolution, you might be better off taking the UDA approach to reduce memory usage.
It’s probably negligible if the image isn’t a crazy-high resolution, but just something to keep in mind.
Let us know if you have any more questions, and thanks again!
One good alternative to the masking of a source is to use a “Source Object”. The letter shape can then be described by another object (e.g., via UDA if your letter geometry is simple, or a CAD file as you suggested). The advantage of this method over masking a bigger source is that you can precisely control the power emitted, instead of having to compensate the losses by overpowering your source.
For using a mask (UDA aperture on another object or Slide), the Slide object main advantage is if you want to have spatial variation of power. Also, masking (via a UDA or a Slide object) gives access to more complex radiant intensity distributions (such as the LambertianOverfill DLL, Source Object offers only a power of a cosine).