modeling external glass lamella

Dear Radiance list,

I have build up a model of a room with external glass blinds. The aim with
my simulations is to obtain the light transmittance of the glass blinds, and
for the entire system (blinds + window).

The blinds are modeled as polygons with the trans material. This means that
the blinds have no thickness in the model, but the reflectance and
transmittance values correspond to the information from the glass
manufacturer (r=0.31 and LT=0.65).

My approach is to obtain illuminance at three different locations:

1) Before the blinds, externally;

2) Between the blinds and the window, externally;

3) After the window, internally.

I have therefore made a grid of points in these locations. The orientation
of my measuring points is outwards, perpendicular to the façade.

My results show that the illuminance level is increased at loaction 2
compared to location 1, which is a result I do not find reliable! I have
placed calculation points in the entire height of the window with a spacing
of 0.01, to be sure not to have the result affected by high illuminance
levels locally. I would have expected higher illuminance levels locally due
to the redirecting of light in the glass blinds, but I would not have
expected a higher average illuminance level. My simulation approach is
given below – please let me know if you can find any mistakes.

Thanks in advance

/Anne

# /az 30,

#Room.rad includes room geometry,

#LamelSystem.rad includes geometry of the blinds,

#Syst2.rad includes geometry of the window system for mkillum

oconv materials.mat CIE.sky Room.rad LamelSystem30.rad Syst2.rad >
./profil30az0alt30/scene_window_prof30alt30az0.oct

cat Syst2.rad | mkillum -ab 4
./profil30az0alt30/scene_window_prof30alt30az0.oct

./profil30az0alt30/scene_window_prof30alt30az0_i.rad

oconv materials.mat Room.rad
./profil30az0alt30/scene_window_prof30alt30az0_i.rad CIE.sky >
./profil30az0alt30/window_ii.oct

#points in grid inside the room
cat points.xyz | rtrace -w -h -I+ -ab 3 -aa 0.1 -ad 1026 -as 256 -ar 512 -ds
.3 -oov ./profil30az0alt30/window_2ii.oct | rcalc -e
'$1=179*($4*0.265+$5*0.670+$6*0.065)' >
./profil30az0alt30/prof30al30az0_2.dat

#points before and after the blind
cat extpoints.xyz | rtrace -w -h -I+ -ab 3 -aa 0.1 -ad 1026 -as 256 -ar 512
-ds .3 -oov ./profil30az0alt30/window_2ii.oct | rcalc -e
'$1=179*($4*0.265+$5*0.670+$6*0.065)' >
./profil30az0alt30/prof30al30az0_2ext.dat

Hi Anne,

You need to share your trans specification, as this may be a problem. It's too easy to get it wrong, and specify a surface that produces more light than it receives....

-Greg

···

From: Anne Iversen <[email protected]>
Date: March 25, 2009 7:45:57 AM PDT
Dear Radiance list,

I have build up a model of a room with external glass blinds. The aim with my simulations is to obtain the light transmittance of the glass blinds, and for the entire system (blinds + window).

The blinds are modeled as polygons with the trans material. This means that the blinds have no thickness in the model, but the reflectance and transmittance values correspond to the information from the glass manufacturer (r=0.31 and LT=0.65).

My approach is to obtain illuminance at three different locations:

1) Before the blinds, externally;

2) Between the blinds and the window, externally;

3) After the window, internally.

I have therefore made a grid of points in these locations. The orientation of my measuring points is outwards, perpendicular to the façade.

My results show that the illuminance level is increased at loaction 2 compared to location 1, which is a result I do not find reliable! I have placed calculation points in the entire height of the window with a spacing of 0.01, to be sure not to have the result affected by high illuminance levels locally. I would have expected higher illuminance levels locally due to the redirecting of light in the glass blinds, but I would not have expected a higher average illuminance level. My simulation approach is given below – please let me know if you can find any mistakes.

Thanks in advance

/Anne

Greg Ward wrote:

Hi Anne,

You need to share your trans specification, as this may be a problem. It's too easy to get it wrong, and specify a surface that produces more light than it receives....

Anne,

There is also an excellent flowchart for understanding how trans interactions are calculated; it definitely helps in defining valid trans descriptions:

http://www.schorsch.com/rayfront/manual/transdef.html

- Rob G.