Hello all,
I'm trying to simulate the occurrence of direct beam convergence; where sun hitting a concave, east facing, glass facade is redirected to form a "hot-spot" on the ground plane. I have a photograph, and the time taken, of the building in question forming a hot spot as well as a 3D model. Using the "mirror" material type I can get an approximate facsimile, but not with the "glass" material. I've tried the glass using "mkillum" and setting -ar 0 but I still can't "fry" those ants.
I'm under the impression that Radiance only considers the glass' reflected diffuse component, and not the direct (although I can see the sun's reflection in the glass)? While close, I don't believe the "mirror" material will give the accuracy we're looking for due to the difference in effect the angle of incidence has on the amount of light redirected off the two materials. Any thoughts, advice or observations would be greatly appreciated. Thanks.and regards.
Chris Humann
Hi Chris,
Radiance doesn't find the illumination of the sun off the glass material because the sun is too small to locate with random (Monte Carlo) sampling. The "mirror" type is the one to use, and you can account for the effect of angle on reflection by including an appropriate brightfunc pattern. Assuming standard 88% transmittance glass, this would be:
void glass glass_alt_mat
0
3 0.96 0.96 0.96
void brightfunc glass_angular_effect
2 A1+(1-A1)*(exp(-5.85*Rdot)-0.00287989916) .
0
1 0.08
glass_angular_effect mirror glass_mat
1 glass_alt_mat
0
3 1 1 1
The Fresnel approximation for unpolarized light was lifted out of src/rt/normal.c. To change the reflectance of your glazing, you have to change both the glass_alt_mat definition and the 0.08 normal reflectance value in glass_angular_effect.
Hope this helps.
-Greg
···
From: Christian Humann <[email protected]>
Date: March 17, 2008 6:51:09 PM PDT
Hello all,
I'm trying to simulate the occurrence of direct beam convergence; where sun hitting a concave, east facing, glass facade is redirected to form a "hot-spot" on the ground plane. I have a photograph, and the time taken, of the building in question forming a hot spot as well as a 3D model. Using the "mirror" material type I can get an approximate facsimile, but not with the "glass" material. I've tried the glass using "mkillum" and setting -ar 0 but I still can't "fry" those ants.
I'm under the impression that Radiance only considers the glass' reflected diffuse component, and not the direct (although I can see the sun's reflection in the glass)? While close, I don't believe the "mirror" material will give the accuracy we're looking for due to the difference in effect the angle of incidence has on the amount of light redirected off the two materials. Any thoughts, advice or observations would be greatly appreciated. Thanks.and regards.
Chris Humann
1 Like
Hello all,
I am currently trying to model simple glazing that transmits and reflects light by taking into account the angle of incidence without using a BSDF.
This method seems quite appropriate in my case. I would like information on the formula:
A1+(1-A1)(exp(-5.85Rdot)-0.00287989916)
What does each element mean? And where does it come from? Simplified Fresnel formula?
Regards
Antoine
This formula was one I developed years ago as an approximation to unpolarized Fresnel reflection, and was later rediscovered and published by Christophe Schlick in his simplified reflection model:
igorsklyar.com/system/documents/papers/28/Schlick94.pdf
I also found a Wikipedia page with a shorter explanation, here:
https://en.wikipedia.org/wiki/Schlick%27s_approximation
Cheers,
-Greg
Hi Greg,
Just to make sure I am modeling the correct material properties, to model an external glass with 20% reflectance, the material definition should be same as below?
void glass glass_alt_mat
0
3 0.8 0.8 0.8
void brightfunc glass_angular_effect
2 A1+(1-A1)(exp(-5.85Rdot)-0.00287989916) .
0
1 0.2
glass_angular_effect mirror glass_mat
1 glass_alt_mat
0
3 1 1 1
Thank you so much
Is there a mirror coating or something that causes the reflectance to be so high? Normally, glass reflectance peaks around 8% at normal incidence.
Your reflection will redirect solar and other light sources from the exterior only using the “mirror” type. You need to make sure the surface normal faces towards the side you want reflections from. The material will have a more ordinary 7% or 8% reflectance when viewed directly using the given “glass” alternate type. If you want to match the 20% reflection more closely for non-source rays, you will need to use a “trans” with appropriate arguments rather than “glass.” I don’t know how much transmission your glazing has, so I can’t tell you what arguments to use, exactly.
Best,
-Greg
Hi Greg,
We are only modeling external reflection for this project. Indoor daylight and transmission don’t matter.
The information we have from the glass manufacturer:
• Glass contains low-e coating
• External visible reflectance: 20%.
• Solar Energy Reflected: 40%.
• VLT is 42%.
We are only interested in quantifying the solar energy reflected, irradiance reflected (w/m^2)
Do you have any advice on how to model this external surface in Radiance.
Thanks,
In that case, I would just use the “mirror” material without an alternate type, and make sure the surface normal faces outside. (The more typical condition in Radiance is to face the window surface normal to the interior.)
Cheers,
-Greg