algorithmic modeling for Rhino
Hi all,
My question is more addressed to Clemens, but feel free to answer if you face the same kind of problem.
I would like to know if you plan on implementing links in Karamba. I am lecturing Karamba at l'Ecole des Ponts for a studio, and some groups have "scissors" connections (two continuous beams that can rotate freely with respect to each other). Beam joints don't work for this, and I think that scissors connections require 7dof instead of 6dof.
I was faced with this kind of connections for the modelling of elastic gridshells (video of the construction here), and I used rigid links in this study to connect the upper and lower layer. It worked well for me (I was using ADINA), and was wondering if it could be a feature to add into Karamba (maybe in the next release)? It's often useful to have rigid links to model eccentricity!
Best,
Romain
Tags:
Hi Romain,
I faced the problem of modeling eccentricity on a reciprocal structure. I think you have at least two ways of doing it in Karamba :
- Model short virtual beams that have enough stiffness (beware if they are too stiff it might lead to numerical errors).
-The second option is to use spring cross-secction instead of the short beams for the connectors. There is a scissor structure example file in the Karamba examples.
Hope this helps,
Best,
Hi Yousef,
I used virtual beams, but I felt like it wasn't the most appropriate solution. I forgot about the springs, I think it's the best way to go in my case.
It would still be nice to have different stiffness following x,y,z axis (or rigid links). It can be very important for the correct modelling of complex connection details. I think of a project I worked on, the coca cola beatbox pavilion: the connections were also modelled with links and there was a lot of discussion on how to model properly the releases.
a picture of the connection details (going out of the pillows).
Thank you!
Romain
Nice!
I don't know if that's what you are looking for but in the beam-joint component you can define different stiffnesses in the x, y, and z translational and rotational directions. You just have to specify which dof you want to release and then connect a vector to the Cr-Start, Cr-end, Ct-start and Ct-end with the values of stiffnesses you choose.
Hi,
yes that's the kind of thing I need to specify. The thing is, I prefer rigid link when it's available, otherwise you have to tweak the spring stiffness and it might make the stiffness matrix ill-conditioned if the spring stiffness is very large compared to other stiffnesses (like you said).
A rigid link would not act on the stiffness matrix and make the solver more robust (free of numerical errors) for this kind of situation.
Thanks for the discussion ;-)
Welcome to
Grasshopper
Added by Parametric House 0 Comments 0 Likes
Added by Parametric House 0 Comments 0 Likes
Added by Parametric House 0 Comments 0 Likes
Added by Parametric House 0 Comments 0 Likes
Added by Parametric House 0 Comments 0 Likes
© 2024 Created by Scott Davidson. Powered by