[Radiance-general] rtrace vs. rview, mirror vs. metal or BRTD

[Greg Ward]

Hi Germain,

The "mirror" type allows Radiance to find a small source like the sun  
by creating a "virtual source" behind the mirrored surface.  The same  
is not true for the other types you tried, which is why your -I  
calculation came up empty.

With many mirrors in your scene, especially if they are not perfect  
mirrors, you will probably need to use rtcontrib to send rays from  
the direction of the sun and track them onto your object, estimating  
the flux density by ray counting.  I think I mentioned this  
possibility to you at the workshop.  Unfortunately, I don't have a  
simple example to offer you showing how it's done, and it's not what  
I would characterize as straightforward.

Essentially, you need to create a ray flux generator corresponding to  
the solar beam radiation striking your mirror array.  You send these  
rays into rtcontrib, specifying one or more materials to record ray  
strikes on the heliostat receiver.  These get counted up and the  
ratio of ray strikes to rays sent tells you the transfer efficiency.

Does this make sense?

-Greg

> From: Germain Augsburger <germain.augsburger at epfl.ch>
> Date: January 5, 2009 10:33:30 AM PST
>
> Hi, I wish you a happy New Year! Starting with a radiance-issue...lol
>
> In my Radiance scene I replaced my "mirror" materials with  
> "metal" (or even "BRTD"), and although rview obviously "sees" the  
> new material, rtrace doesn't "see" anything (irradiance measurement  
> -I). I tried to set the rview and rtrace options as similar as  
> possible (for instance -dr, -aa, -ar, -ad, -as) but so far without  
> success. Is there a way of getting "metal" or "BRTD" work in a  
> similar way to "mirror" with rtrace?
>
> My 3 scenes (field of mirrors reflecting the sun towards a central  
> receiver, 3 times: with "mirror", "metal", and "BRTD") are  
> available if needed (too big in a general radiance mail); and here  
> are some of my command lines (first "cd HelFiSim"):