Why Include Springs in a Hanging Chain?
Real chains, like all materials, have some springiness associated with them, as defined by their coefficient of elasticity (also known as Young's Modulus). As shown earlier in the post about what FEM is, a spring with it's spring constant (k) can accurately model a material with it's modulus of elasticity (E).
When modeling a rigid chain, the only way to adjust how far the chain sags is to add or remove links to change the length of the chain itself. This changing mechanic is limited by the length of each change, as the chain has to have a whole number of links added or removed. A spring chain, however, can adjust the spring constant of each link to manipulate how far down the chain hangs, allowing for much finer adjustments. Every time a spring chain is hung, it can be reset to the same level starting position, making the experiment repeatable:
Spring Chain Model, Starting Position |
And will end in a catenary curve, defined by the total length of the span and the spring constant:
Spring Chain Model, Catenary Curve |
The number of links, in this case, serves more to make a cleaner curve than to change its shape, with more links making a smoother curve, because each spring is assumed to be linear by itself. Each connection between the springs is simply a small cylinder, that is assumed to be a point. The mass of each member can possibly be changed to change the gravitational force, but the shape will most likely only be changed by manipulating the spring constant.
The next step is to read more about Solidworks' Motion Study, to find if/how the motion can be exported, preferably as a text file of element positions, so that I can turn that back into a curve shape to model and test. If I can get points that make the catenary curve, I can import the points back into Solidworks, then spline along the curve, then sweep a cross section along the curve, creating one catenary arch. This process, if possible, can be expanded across a network of springs to create a complete gridshell shape.
No comments:
Post a Comment