Hello All,
I am trying to set the mullion types for curtain wall ends depending on the angle between each two adjacent walls. Now I got it to work almost perfectly, but there’s a small problem. The mullion types are wrong on both ends of the north wall. I suspect this may have to do with the shift indices node I use on the wall vectors list to get a list of vectors adjacent to it and then get the angle between them. Or could it be the angle values that are causing the problem? In either case, what would be the most efficient way to solve this?
Thanks
Are you able to post the dyn and an example file? It’s a bit hard to visualise how the graph is working without context
Hello Mike, thanks for your reply. So this is the whole content of the dyn file, I attached a screenshot from the plan marking the curtain walls that have the problem, regardless of their state (flipped/not) they have the wrong mullion types on both ends, while all other walls have the correct mullion types set.
I was after the files as I would be trying to recreate what you already have done and the image provided does not show what you are trying to do with the mullions?
Your graph gathers all the walls of a type in the file and, as you have multiple joined elements, by shifting the index by one you could be comparing the last wall with the first wall in the next joined group - potentially it could be any wall based on how the walls have been created
Use GeniusLoci Get Joined Elements, which gets the walls at join in the order start, end. Also round the values as there can be some fine errors when calculating vector angles
The mullions are also not guaranteed to come out in order start / end as the Mullion.ByElement node iterates the Curtain Grid. You may need to compare the location X & Y to the wall end points
Flipping can be tested with a short python script OUT = [w.InternalElement.Flipped for w in IN[0]]
2 Likes
