[Radiance-general] Re: [Radiance-addons] Prismatic glazing/blind reflector system analysis

Phillip Greenup [email protected]
Wed, 25 Feb 2004 10:14:39 +1100


Hi Anthony, Jack,

What Jack says about simulating light pipes is right.  simulating these with radiance in general will not work.  i haven't tried the photon map, but sounds like it should work for you there.

regards modelling prismatics or similar materials, there are a couple of cal-files that can help.  prism.cal is part of the standard radiance distribution, and allows the modelling of prismatic glazing, using the prism2 material primitive.  lcp0.cal is another (not part of the standard distribution but i can supply it) that models laser cut light redirecting panels (LCPs).  this also is for use with the prism2 material primitive.

the use of each of these cal files is not simple.  it requires manual editing of scene files, careful orientation of geometry and material definitions and knowledge of the form of the prismatic or laser cut panel being modelled.

let us know if you would like the LCP cal file, or more help creating prism2 definitions.
Phil.

_______________________________
Phillip Greenup
Specialist Engineer
Arup Australasia
Level 10, 201 Kent St
Sydney NSW 2000
Ph (02) 9320 9426
Fax (02) 9320 9321


>>> Jack de Valpine <[email protected]> 25/02/2004 1:37:34 am >>>
Hi Anthony,

To simulate the perfomance of things like light pipes, you need to use a 
tool with forward raytracing capabilities, radiance in its classic form 
is a backward raytracer:

    * forward raytracing - trace rays from sources in scene to eyepoint
    * backward raytracing  - trace rays from eyepoint back into scene

One way to do forward raytracing is "photon mapping." Roland Schregle 
has written an extension to radiance that performs photon mapping:

    * http://www.radiance-online.org/ see "examples & extensions" in the
      software section
    * http://www.ise.fhg.de/alt-aber-aktiv/radiance/photon-map/ 

Since this is an extension to the core radiance engine, this will 
require moving beyond Desktop Radiance.

Note there are two prism material types in Radiance: prism1 and prism2. 
These would probably be useful for defining the properties of the 
prismatic glazing, the trick is acquiring perfomance data for the 
physical material(s) and correctly modeling the behavior with these 
radiance material types. I have not really used these material types 
before. However there should be others in the group who can provide 
feedback to you.

It is also possble the depending on how you breakdown your problem that 
there are simpler ways to do what you want. For example what are the key 
issues of what you are trying to simulate and evaluate:

    * properties of light pipe performance - then this probably requires
      the use of photon mapping
    * properties of prismatic glazing as enhancers of light distribution
      - can this be seperated out from the light pipe and evaluated
      using more classic techniques in standard radiance, in other words
      can you make some estimates for light output from the side
      apertures of the light pipe and thus remove the actual simulation
      of the light pipe behavior?

I expect you wil get other input, which I will look forward to reading 
myself.

Best Regards,

-Jack de Valpine

PS: I have cross posted to radiance-general, to give the discussion 
wider input.

Anthony J. Farrell wrote:

>Hi folks,
>
>I wish to analyse the light distribution to an office room supplied from a
>light pipe fitted with side apertures.
>
>The apertures will contain prismatic glazing or a blind reflector.
>
>I have drawn the light pipe using desktop radiance, and made the surface 97%
>spec ref. as from typical light pipe manufacturers data.
>
>I wish to include prismatic glazing or other novel glazing types such as
>blind reflectors to enhance the light distribution from the side aperture.
>
>Can anyone give me some insight as to how I may do this, since the glazing
>database that comes with desktop radiance only contains common glass types,
>and not prismatic or such like - perhaps one of you know of a patch for
>desktop radiance that provides several different prismatic glazing types, or
>of code for Radiance that includes for it.
>
>Thanks, and best regards,
>
>Anthony Farrell
>Dublin Institute of Technology, Ireland.
>
>
>  
>

-- 
#	John E. de Valpine
#	president
#
#	visarc incorporated
#	http://www.visarc.com 
#
#	channeling technology for superior design and construction