To follow up on the last post, we’ve done two more labs in 245X this last week with some more good results. Yesterday we did the infamous “12-node” project that Kevin Dong has developed at Cal Poly. Students are given 12 “nodes” (actually Skittles, though I’m sure at MIT they would use stainless steel ball bearings), and they have to stabilize them 3″ above a fixed plane using wooden sticks. The key is that the math behind translational stability says that no matter how many nodes you have, and no matter what arrangement they’re in, you will always need three restraints (sticks) to stabilize each node. So for this, you should be able to do it using 36 sticks. Students have to figure out, though, how to arrange the sticks so that each plane (count them–there are seven) is adequately stabilized.
One of the most interesting parts of the last couple of labs has been what happens when you get many, many iterations of the same model. What we’ve noticed is that, despite what we think are carefully written ‘rules’ for each lab, there are always a few projects that think a bit more laterally than we thought possible. We’ve taken to calling these “cheats,” but we’ve rewarded them with some special shout-outs and some explanations about why they might work better, or differently, than the “textbook” models. In this lab, the most common cheats were to use bracing in the top, horizontal plane (which allows distribution of in-plane stability among more than one column line) and bracing that occupied a diagonal in plan (thus bracing multiple planes at once). Both of these anticipate the next exercises in the 12-node series, so it gave us a good segueway into the work we’ll do after Thanksgiving break.
As one student put it (and this might be the new STP motto), “if you ain’t cheatin’, you ain’t tryin’…”