I want to create an inflated dome structure just like the example here:
Using the naked points as the anchor like this is exactly what I want.
I understand how to use the Unix Z vector, EdgeLength factor/strength to change the height of the structure. How can I calculate these to correspond with real world values like 3 feet or one meter?
I mean… you would just open your original file document and set what units you are in. those are the parameters it goes off by.
The original file is in inches however I am sure the math is not that simple.
And each setting affects the overall height. If you look at the screenshots just changing the Target count of the Quad Remesh means that I have to change the force vector accordingly.
Both of these screenshots with the settings equal a tent height of approximately 4 feet.
Both of the geometry screenshots have the 600 target and .09 force vector but a smaller footprint will give a shorter than 4 feet (selected vertical line) total height.
Currently I just try to eyeball the tent height to match the 4 foot line but I need more precision than that to use it for construction.
Can anyone help me figure out the math for this so I can calculate an exact height based on the dimensions of the base and all the settings. Or can I set up the grasshopper file to give an exact height?
GH Settings:
Geometry:
You’re right that the resulting max height of a form-found shape like this does not depend on the input parameters in a linear way.
One approach would be to script a custom Goal which checks the highest point on the mesh and adjusts the load accordingly, in an iterative fashion, so it converges to that exact height.
However, a much simpler approach which would give a near identical result to this would be to set the form-finding parameters to get a result close to the desired height, check the current height, then scale vertically by (desired height)/(current height).
Thank you, that definitely gives me some direction to work on.
Still to green to know how to do any of that but, I will see what I can come up with.
An alternative approach that I’ve been experimenting with recently is using the Zombie Solver in combination with Galapagos.
For example you could measure the output height and have a desired height, which can allow you to work out a fitness expression = abs(desiredH-outputH), absolute value as you might be above or below the target. This can be your fitness that you minimise and your genome can be the force vector applied. You need to be careful with the measurement precision (i.e. round up to the relevant decimal place) and use enough decimal places in your input so that you can effectively minimise the fitness to 0. See below finding a hanging chain with an exact height, using K2.
The reason for the Zombie is that you save time in not displaying any internal K2 iterations. Just check that it has converged afterwards, or give it enough max iterations. Depending on the complexity, the Galapagos part might take some time to solve.
1 Like