Hi,
I’m testing the direct sun hours in a courtyard for each day of the year. This is a mid rise building project in London. The building mass is a 50% glazing to solid ratio, as shown in the image.
The end result I’m hoping to see is : what percentage of the area of this courtyard (highlighted in the image) receives at least 2 hours of direct sun for each day of the year.
The result graph is very bizarre. I was hoping to see a gradual curve. Unfortunately, it doesn’t seem to be the case. What am I doing wrong?
Sun Hours_Forum.gh (1.1 MB)
@mostapha @chris
Any help to help me resolve the issue, will be amazing. Thank you in advance.
Without checking the file, I suggest using the ViewFromSun component, it can be your friend here. It allows to check what the sun sees. More than once this checking helps to understand the what and why of the results [the questions your are asking].
My assumption is that because this is not a simmetrical case the context is causing that some months/hours the open space will be more/less exposed, sometimes looking counter intuitive.
-A.
I tried changing the various dates to see this, and I’m not able to interpret what this means.
What you see in the image is what the sun sees. Meaning that you can have a better understanding why some areas are more exposed or shaded at certain times, and this can explain the results you are getting.
Since a whole year is a lot to check, I would take a more manageable period of time, maybe those where you identify/feel there is a problem.
-A.
Okay, got it.
I tried looking at the problematic times of the year, and the image seems to show at least some parts of the courtyard at all times. So the results of the HB Sun Hours analysis should ideally not show these results in my opinion. Let me know what you think on the basis of the attached .gh script, if I’m missing something.
I checked the file. So you are calculating daily the condition of solar exposure you want/need [at least 2 hours].
Also checked the ViewFromSun [VFS] for the day you set in the file [30.12]. I checked using different EPW since you didn’t specified which one you use. So i get 0% exposure, which the VFS confirmed. Checked another date [30.6 for summer] and i get ~90% exposure, also confirmed with the VFS.
In short i believe your workflow is fine.
Note: Consider updating the LBT components. There are various versions in your file, all of them relatively old [1.6 or lower]. For your case it doesn’t affect, but for other cases you never know.
-A.
Hi Deepthi,
I don’t have access to GH for now. But here are several thing that could go side way in your process.
- how refined is your window analysis grid. I suppose there is only one evaluation point per mesh. your can imagine is there is a 10m x 10m mesh, and only the center of the mesh is evalutated. that will fail to capture the shadings. I would say breakdown into 0.5x0.5m quad grids to ensure good resolution.
- the buildings do create gaps that lights can shade/transpass. check shadow analysis for specific time and day for validation.
- how refined is your timesteps in simulation? too big a timestep might fail to capture the abovementioned self shading effect.
- how is your percentage of area defined? are your averaging the number of points or the actual area of each simulated meshes?
- might have some bugs in your condition statement for finding the >2hours meshes.
Anyway, for sunlight hour analysis, LB model is not necessary. I would just do it in a simple shadow analysis from sun vectors, which is lighter to run and simpler workflow for agency to each parameters.
Help these can be helpful! Cheers, Will
