Ribbon vs. individual windows

radiance
openstudio
daylighting
retag add tags

I've modeled a ribbon window as both a single, long window across a facade and as fairly tightly-packed individual windows. The average annual illuminance results for the tightly-packed case are 17% higher (42 vs. 49 lux) than the single window case. Don't know why.

Model particulars: 40'x40' room, 3' sill, 9' wall, south-facing, 5' cubicle walls, average annual illuminance was calculated using the OpenStudio Radiance measure's daylighting control sensor just inside west exterior wall behind first cubicle wall, all materials (windows, walls, etc.) are exact copies in both cases. The only difference is 3x4 windows (13, <1" apart) vs. a single 40x4' window. Not exactly the same WWR, but close.

Radiance parameters: -ab 10 -ad 65536 -as 512 -dj 1 -dp 1 -dt 0 -dc 1 -lw 1.52e0-05, not 3-phase, just glass (i.e. "uncontrolled" in OS Radiance-speak). The -ar is the default which is 256 I think.

Here's the link: The results are ready to examine in DView. For the ...30_34 space, for the individual window file (...74.osm), the average annual illuminance of the daylight sensor is around 49 lux, whereas it is 42 in the ribbon window file (...75.osm). I've run them a few times to check for "stochastic" variance but it always turns out this way.

1.1k

Determinant

asked 2018-06-27 20:44:47 -0500, updated 2018-07-12 12:26:55 -0500

edit flag offensive 0 remove flag close merge delete

Comments

add a comment see more comments

That's insane. Especially with those parameters, which should be rigorous enough to capture the lighting accurately in each case. What is the -ar set to, anyway? And am I correct in saying that you are only calculating the illuminance at a single point? How are you calculating this? i.e., is this a point-in-time calc with glass material for the windows, or part of a three phase? Are you sure the materials are the same for each case, and only the fenestration is changing? Maybe some more details on the "tightly-packed case" elevation would help. If the net glazed area is the same, and in the same position on the elevation, and materials are constant, there shouldn't be such a big difference here.

Comments