Annual Daylight Analysis Inconsistency

ladybug-tools
,
1

Hi Group.

Riddle me this! I’m running a very simple room analysis with one window to determine the annual daylight impact within a room based on orientation. My location is London, I’m using the HB Annual Daylight recipe. My room reflectance are 0.8 for walls and ceilings and 0.4 for floor with a window having 0.62 transmittance. I’ve plugged in the Radiance Par component set to medium quality (so -ab 5 etc). I’ve then run a script to change the orientation of the room (via the compass) at 1 degree intervals. I’ve then post processed this into an image to give the sDA of the room at each degree.

All very straightforward. However as shown by the images attached the results jump around wildly. I’d expect some variation as obviously with radiance there will be a little based on the raytracing algorythim. However it’s jumping from an sDA of 97% at 1 degree, down to 54% at 2 degrees back up to 96% at 3 degrees. And then randomly in that format throughout 360 degree orientation. What’s going on here? Any help or pointers would be greatly appreciated as atm I’m not sure I can trust HBR results for Annual Daylight.

Thanks in advance

Dan

1
11920×1571 57.9 KB

2
21920×1571 58.9 KB

3
31920×1571 58.2 KB

1 Like
2

Hi @ladyloo, are you able to share your model / script? I expect there’s something in there that will give the answers.

Cheers, Charlie

3

Hi @ladyloo ,
Based on your description, something is coming up to my mind. But like what @charlie.brooker said, the script and model will be helpful for troubleshooting.

You may double-check:

  1. rotate degree.
  2. shade geometry position.
  3. sensor grid size.
  4. glazing performance setting.
  5. unit of all input numbers.
  6. make sure you don’t miss any hidden expression

Jiewei

4

Hi All,

Thanks for the responses. I think I got to the bottom of it. It came down to my room modelling. I originally modelled a hollow solid with a rectangle window cut out. This initially worked ok but when I extracted the floor, walls and ceiling, the results were blown. So I remodelled an HB room with no wall thickness and an HB window attached. This worked fine.

So it seems like the problem was light leaks. A schoolboy mistake but goes to show the importance of getting the basic modelling right.