Glare from a high-reflective tower facade

asked 2018-05-13 16:20:08 -0600

Arch2010
129 ●7

updated 2020-03-24 10:13:26 -0600

I would like to examine which parts of a high-reflective tower facade contribute to glare in rooms across the tower. Some patches of the facade reflect sunlight at different times of the day and cause glare in different rooms across the tower. Can I assign these patches (surfaces) a light source using the rcontrib or possibly using mkillumor what would be the best practice to do this?

Deleted a series of steps that are not right.

Another thought:

I already created an array of views from the opposite building, below is a link to one HDR view!

In the HDR image (below), the purple area is the source of glare (please correct me if I'm wrong), so all the facade panels covered in purple are offending my view. Shouldn't this be enough to answer my question?

if I can overlay this image with my geometry (I'll use Matlab for that) I will be able to determine exactly where the offending panels are!

Image herehttps://static.wixstatic.com/media/07...

edit retag flag offensive close merge delete

Comments

I think there are many missteps in the above. First of all, it doesn't seem that you know what you are trying to calculate. It is difficult to help you with this, but I would guess that you want something more like views from the opposite building if you want to know the glare contributions from different parts of the facade. For example, you might have an array of views looking through the opposite building's windows or from its surface towards the offending building facade. Your current array of surface irradiances is not useful.

GregWard ( 2018-05-15 12:10:11 -0600 )edit

You are right, I don't know what I'm doing. I'm completely lost here. I'll keep working on this and try to figure what the best solution is! Thank you so much for trying to help me though!

Arch2010 ( 2018-05-15 12:24:26 -0600 )edit

As discussed earlier, it is best to use a single modifier and the bin expression for a more efficient calculation. You can use your original material as an alternate modifier to the mirror primitive, but you need to figure out the average specular reflection amount. (See Radiance reference manual.) Is there someone more experienced who can help you first-hand? Doing this over Unmet Hours is nigh impossible, and the resulting post is a mess.

GregWard ( 2018-05-15 12:42:03 -0600 )edit

If a series of views give you the information you are after, that is certainly an easier approach!

GregWard ( 2018-05-15 13:09:38 -0600 )edit

add a comment

answered 2018-05-14 10:49:34 -0600

GregWard

1203 ●6 ●5 http://www.anyhere.com/gward

updated 2018-05-14 12:52:59 -0600

I have not tried this, but I believe you can use the "mirror" type to create virtual light sources for each glazing panel using a unique modifier (material name) for each. You can then give this list of modifier names to rcontrib or rfluxmtx using the -M option. The result should be a matrix of reflected contributions from each source for each evaluated irradiance position in the room, or per pixel if you are generating component images.

If your offending building's facade is planar, you could use a single surface/modifier and a bin calculation to tell you where you are reflecting the most glare, and that would be a lot more efficient, since it wouldn't create hundreds of virtual light sources to track, only one.

For example, the following expression would compute a bin number corresponding to left-right then bottom-top ordering of panes on a window wall divided in a Fw x Fh grid where each pane is Wp by Hp in world units starting at lower-left position of (X0,Z0) and extending in the XZ plane:

floor((Px-X0)/Wp) + Fw*floor((Pz-Z0)/Hp)

You could give this as the "-b" option to rcontrib for your window-wall modifier, and it would create Fw x Fh output values for each ray (bundle).

Comments

Greg, Thank you so much for your help. I have a question though, what do you mean by "bin calculation"? could you elaborate on this? as my reflective facade is planar.

Arch2010 ( 2018-05-14 12:33:07 -0600 )edit

Added to the original answer above.

GregWard ( 2018-05-14 12:53:28 -0600 )edit

@GregWard thank you, I'll experiment with this the next few days and hopefully, I can get this to work.

Arch2010 ( 2018-05-14 15:11:05 -0600 )edit

@GregWard so I think I'm lost and in need of help. I updated the original post with the workflow I used. It would be much appreciated if you could direct me through this process.

Arch2010 ( 2018-05-15 01:49:57 -0600 )edit

add a comment

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.