Skip to main content

Mark Schmidt

It began with a simple question: How difficult is it to make the <redacted name of a famous fictional armored vehicle made a long time ago in a studio far far away> walk correctly with Thermal Desktop® articulators? I felt this was a very good question, and I was lucky to be the chosen one for it. I set aside my task of herding nerfs to give it a try. That didn’t work, so I decided to either do or do not, and that’s when things really came together.

I used TD Direct® to model everything except a few simple shapes that were suitable as finite difference cylinders in Thermal Desktop. I used swept meshes where possible to limit the node count. While I would usually recommend simplifying as much as possible, in this case I only removed features that were not visibly significant from a version I found on GrabCAD. After all, the goal was to be faithful to a fictional concept rather than to produce a good thermal model. It took some time, but my creation was ready for thunderous applause.

It was a trap! The question had been rhetorical, and things were not proceeding as I had foreseen. We discussed copyright laws, public domain, how much I really wanted to keep it, how much that didn’t matter to an angry mouse, and more. Somebody had a bad feeling about this. I insisted it would be fine, but my boss informed me that I was mistaken about a great many things. It was ultimately decided that we, as a company, would prefer to avoid any “cease and desist” sort of nastiness and weren’t interested in hearing the odds.

It almost went into the garbage chute, but instead I explored the many features of TD Direct that allow for rapidly changing geometry. The body and head were easy to push and pull into new shapes using SpaceClaim's intuitive controls. The leg segments all referenced a single part, so changing one foot changed all of the feet. The update to this new geometry was relatively painless (aside from a great disturbance of destroying my creation).

In short, I crossbred the unidentified fictional vehicle with a steampunk camel, visited, and made a parallel use of short prepositions. And now it is with somewhat less pride than I had previously hoped that I present to you the Regal Stepper ON-ON. It may not look like much, but it’s got it where it counts.

The ON-ON is an exercise in nested articulators rather than a complete thermal model. There are no material properties, contactors, boundary conditions, nor heat loads as we do not have enough information about the conditions on (…sigh…) “Hoch” nor the energy delivered by the weapons of an (…ahem…) “Iceracer.”

Each leg consists of 4 nested assemblies that rotate. The nesting starts at the top, since any motion from there would change the rest of the leg. The rotations are controlled by interpolating singlet symbol arrays over time. The legs cycle back into their original position every 8 seconds. The first leg to shoot out is the front on the left Han side.

The angles were calculated somewhat manually using a secondary AutoCAD document to sketch out the legs. Two basic positions were used. The first is with legs on one side forward and legs on the other side backwards (as shown in the image above). The second is with the legs on one side vertical, and the legs on the other side outward. The motion between these positions was mostly trial and error. The hip and knee joints worked together to keep the feet from crossing the ground plane, and I tried to make it look realistic.

The ON-ON translates through use of a 17th articulator that includes everything except the ground plane. The walking motion is 4 cycles through 32 seconds. Now witness the power of this fully legged and operational articulating model.

You might be wondering why the ground plane is orange. Quite honestly, it’s because AutoCAD displays white as more of a grey, and it was hard to distinguish the colors (not that there was any reason to assume white, of course). Or perhaps it’s a nod to the steampunk camel.

Depending on the use, the translation may not be necessary. The ON-ON could be run through repeating 8 second orbits in a fixed position. The only problem is that’s not any fun at all. Maybe it is on the way to a shield generator that looks like a prime candidate for finite difference objects, or maybe it is just going to pick up some power converters.

There is a lot of room to improve the automation. I am certain that adjustments could be made to the time arrays to make the motion a little smoother and realistic. I’d suggest exporting the symbol arrays to a text file, working with that, and then reimporting them. Perhaps you could find something in a science-fiction movie of some sort that might mimic these movements to use as a frame of reference if you were so inclined. Of course, if you are working on a model of something you are actually building, it seems someone might provide you with the exact angles.

Please see the forum post to read more and download the model. 

May the Kinetics be with you.