[Radiance-general] Using radiance for simulating light pipes

[Rob Guglielmetti]

What Lars said, and:

Most TDD manufacturers provide performance tables based on some incident illumination on the exterior aperture, relating that flux to an expected avg illuminance inside a model space, assuming some room cavity ratio and TDD spacing. People have been known to use those tables, and an annual exterior (e.g. roof) illuminance schedule (ideally based on a weather file and Perez sky) to scale the predicted interior illuminance for an annual (or whatever) result. In this way you could use a light backwards ray tracer because you’re just solving the problem from the roof to the sky/sun. This whole process would be super fast with Lars’ recommended approach to deriving the annual illuminance schedule (i.e. using gendaymtx and a single point on the roof). This approach also takes into account local obstructions (rooftop units, trees, whatever). 

Several TDD manufacturers also provide photometry files that you could use in Radiance to compute a grid of interior point illuminances, and again scale those up or down based on the roof illuminance schedule. In this way you could do a fairly reasonable accounting of an actual interior. 

The performance tables and photometry files are all a bit nebulous, based on assumptions and best/worst case scenarios, and voodoo. Your mileage may vary. Proceed with caution. Look both ways before crossing the street. But hey, it’s something. =8-)

- Googs

On Jan 7, 2014, at 1:20 PM, Lars O. Grobe <grobe at gmx.net> wrote:

> Hi Alejandro,
> 
> the topic of light pipes appears from time to time, and unfortunately
> the simple answer is that Radiance (and any backwards-tracer) is not
> really suiteable for the task, if
> 
> 1) you consider sunny days (you probably want to do so)
> 
> 2) you do not have a diffusor placed on the top aperture of the pipe.
> 
> There are still some promising approaches to solve the problem. One is
> using descrete patch models of the sky, such as the Tregenza scheme,
> with the direct sun contribution being distributed over larger patches.
> This implies distributing the sun radiance over a larger part of the
> sky, which smoothes out sharp peaks in the transmission - which may not
> be crucial in many cases.
> 
> The other option is using a forward extension to Radiance, such as the
> photon map. This requires a modified version of Radiance, but gives you
> a tool which has been developed with exactly applications such as light
> pipes in mind.
> 
> Finally, you may even try to generate the BSDF of a light pipe using
> genBSDF and apply this to a surface representing the bottom aperture.
> 
> The latter two options keep the resolution of the incident radiance,
> especially the "sharpness" from the sun, intact.
> 
> Cheers, Lars.
> 
>> Hi everyone,
>> I am a newcomer in the radiance community. I am involved in a project to
>> assess the light output from light pipes. We are gonna simulate the
>> light output of 2 types of light pipes and then compared them to the on
>> site measurements. I would appreciate any guiding or advise on where to
>> start the simulations. I am considering backward retracing as a more
>> suitable option than forward retracing for this particular case, what do
>> you think?
>> Thank you
>> Ale
> 
> 
> _______________________________________________
> Radiance-general mailing list
> Radiance-general at radiance-online.org
> http://www.radiance-online.org/mailman/listinfo/radiance-general