The 3D Midterm (Re-Planning and Model)

As I mentioned in this previous blog entry, I've been working on my mid-term assignment. The general idea is that we would make an arrangement of geometric shapes that uses at least 3 of the rules of proximity and involves at least 2 basic pure geometric shapes. After the approval of our 2D layout, we are to make a 3D model of the result that explains the parti. In the last entry, my 2D model looked something like this...

Yeah... that didn't happen, though the Professor liked the concept. She liked it a lot actually, but had a few cool ideas to toss in, like slicing the circles and honeycombs and rotating them 90 degrees. We also discussed perhaps making a beehive effect out of it. In the end, however, I realized there was no way I could make a good honeycomb model on my own, and I would need to simplify the concept considerably. But I still really liked the slicing effect.

As a result, this is what I turned in for the "final draft." Keep in mind that we can alter the draft slightly during the modeling phase, but for the most part, this is what I committed to. To make it clearer, here's the same layout, but with the guiding lines and grid added...
I call this the "gears" concept. Imagine 3 gears have rotated. The first gear is in the northwest portion of the parti. The large western hex has a slice rotated 90 counter-clockwise towards the north. The next gear, in the middle, slices a golden scale smaller hex, and smaller circle, and rotates them 90degrees clockwise (as if turned by the larger gear). The third gear is in the southeast, where the larger circle has been sliced and rotated 90 degrees counter-clockwise, as if turned by the smaller gear. For you gearnuts out there, no, I almost certainly did not get the correct gear ratio on these turns.

So now, grandious schemes develop as to how I'm going to translate this thing into 3D. The concept I came up with, much like the cube project earlier this month, was overly complex. I planned on having crystal pillars with lights inserted into them, and the two bottom bases elevated, a mirror. The structures would be plaster of paris, the central pillars would be clear plastic (or dyed). This would be an experiement in light along with the assignment. The professor nixed the lights idea. Thank g-d.

The final concept would look something like this. Each crystal pillar would support two floors at differing elevations. The elevations would increase by 2" for each floor, starting with the western structure and going up clockwise from there. Each floor would be a rigid, clear plastic sheet, so that one could look down from a top view and still see the 2D effect of the slice. The elevations tell the story of the slice directions. West, then North, then East, then South. The whole thing seemed quite beatiful in my mind, and I think had I a bit more time, patience, money, and skill, I could have pulled it off, even with the lights, and made something truly cool out of it. But no matter... leaving out the lights would simply mean more time to work on the actual meat of the project. Again, thank g-d that the professor nixed the lights idea.

After the first day, I had basically managed to get all my materials together, make a huge mess, and a really, truly, horrible attempt at a hexagonal structure. See that white thing that looks like a crumpled ball of paper in the upper left? That was the hexagon. It gets worse, but I'll spare you the details there. Suffice it to say I only had one day left, and all my previous work had done was to tell me how NOT to build a model. This would take a lot more simplification...and measurements.

The first thing I did was go back to the idea of using foamboard. Good old reliable foamboard, how I love thee. Only, building a complex shape out of foamboard required a good deal of math and also a lot of carving foam away from the corners so they would join properly. I also figured I'd shorten everything by an inch. I'm not sure how I forgot about that second part, because in the end, the floors ended up being 2", 4", 6", 8", with all the bases at effectively zero-elevation.

The 3/4 dome and 1/4 dome structures were actually far easier to develop, thanks largely in part to the hollow cristmas tree ornament I'd bought and cut for the job. Of course, I'd have to cut my PVC pipe a lot more precisely, so no more hand-saw. Instead, it was time to bring out the big guns.

This is a compound miter-saw, capable of cutting at nearly any angle needed. It's used to cut molding. In this case, I just bought a thin plywood-blade (~130 teeth) and that worked fine for cutting PVC pipe. However, a word of advice to those of you looking to but this wonderful little timesaver for your own work: get the 10" saw, not the 7.x". There's almost nothing you can fully cut through with a 7.x" miter saw. Not even a 2x4. I had to play all sorts of crazy-dangerous finger-arobics in order to cut what I needed, then flip it around and do the exact same thing, only in reverse. For every single piece.

Once my shapes were properly assembled, I gave them a quick layer of white undercoat to help hide all the scotch tape I used putting them together. If I'd had another day, I'd have painted the model very carefully to improve upon this. As it is, I only finished the thing an hour or two ago.

The table, with my tools, my drinks, and my replans. Before I continue, I must take a moment to thank my wife and son. I thank my wife, because she not only gave me the entire weekend to work on this project, she took care of all the chores, the boy, and and fed me as I toiled over this thing. She did so with a smile and grace I never would have been able to manage, even though she deserved so much more than what I was able to give her this weekend. I also thank my son, who occasionally forced my hand in taking a break and remembering the reason I was doing this in the first place. I feel terrible about missing out on time with him so much lately, but summer semesters are just like that. I thank him for trading his hours with Daddy for hours with Mommy. I just wasn't able to be a very good father this weekend.

So, here's my "Model Draft." I'm not really sure what we were supposed to call it. But you'll notice a few changes, such as the loss of the extraneous bases, the blatant circle to emphasize central slices, and the removal of the invisible lines. Here are the final results in 3D.





The western view is actually quite similar to the one in the original 3D planning phase. It's just lacking the extra bases and elevation differences on the bases. The final result is a model that has a 13"x13" datum, and golden height of 8". The crystal pillars in the center were a bit of a paradox. In some ways, they came out better than I'd feared, but worse than I'd hoped. I learned a number of valuable lessons.

The first is that if you are going to use polyester resign to form something, you need a leak-proof, non-pourous, easy to open and remove mold. One of my molds was made from aluminum cans and duct tape. The aluminum did peel off revealing beautifully smooth clean surfaces underneath, but the duct tape was a mess. Even worse, though, the "slice" had been created with a cut piece of paper towel tubing (actually model rocket tubing, but the point remains the same). Both leaked terribly, losing me about 3/4 the resign through the bottom and sides of the containers. After curing, when I removed them, not only did I end up having to cause a lot of damage to the exterior part of the crystal pillars to separate them (the carboard tube divided had leaked, remember, but the cardboard the tube was made of didn't come off. The stuff is bonded to it like expoxy. Goo-B-Gone, sand-paper, whittling with a razor...none of these things could fully remove the stuff. I resigned myself to having bad crystal pillars. There was no way to re-create them in my timeframe and budget. Which brings me to my second lesson...

If I think I'm going to use polyester resin again, I need to order it bulk, online, and not from the local hobby store. I also need to design smaller things for it. That stuff is expensive! Anyway, the actual strength and relative shape of the cured resin was pretty decent. It does not exactly match the circle they've been aligned with, but it's a close enough fit that it gets the spirit of the parti across.

Of course, I fully expect that the professor will want some things changed with it, though honestly I'm not sure how much I could actually change at this point. In any event, I look forward to the class discussion tomorrow...er...today... in a few hours.

Good lord, I need to go to bed!!!