Shadow Pavillion Models

Shadow Pavilion paper studies from Karl Daubmann on Vimeo.

Shadow Pavilion prototype from Karl Daubmann on Vimeo.

Video

Project Re-cap

The loftiness of goals in the initial weeks was incredibly hard to give up. Our project had begun at a project about "evil green" in which cigarettes and thermal electrics would unite to provide power. We saw this as noble enough simply because we were eliminating waste as well as providing energy. I still believe this is a good idea for a "smart surface", however it had no real place in the course at hand and probably would not go over well inside a gallery. I think the issue that really overcame our group was that there was never an idea on the table after this that any of us could get excited about. We tried and tried to come up with something new and exciting, all without making anything. Thinking by making is a tried and true method, and I continue to wonder why this did not happen any earlier. I feel that there were also quite a few issues in trying to please professors, all of whom seemed to want different things. Finding an idea that would fly with all professors seemed an impossible task, and I don't believe it was ever reconciled.

Editing: I think editing was something that our group had to do nearly every meeting, and I think that this was in large part the reason why our group had a successfully moving surface. Our lofty goals were continuously edited down until we reached something that we didn't just think, but rather knew that we could accomplish. I think that this method of editing is quite a contrast to the methods of other groups including the Simon robot, who seemed to really let editing happen by chance of mechanics not working out. However, I believe that their project did a much better job than ours of allowing group members to test out different territories, Mark learned to TIG weld for example. I think that this cannot be said about our group, as people somewhat stuck to respective disciplines during the entirety of the project.

Motivation came back to our team in the end when we began to see results. The project is full of small steps, and unfortunately we left most of these small steps until the last two weeks, and when people began to see the small steps materializing everyone seemed much more aboard the project than ever. I have to admit the idea was weak and I resisted it pretty much the entire time-I disliked the intention and really tried to deviate the project into other directions without much result. In terms of intention I thought the project was not a complete success, but in terms of getting a team together making a surface that could move and do something functional, as well as be a aesthetically pleasing thing, I believe it was a complete success.

Final Documentation

The Final Review went relatively well, and our surface was moving and responding so our team was reasonably happy with the outcome. The reviewers seemed to like the physical surface, but really had nothing good to say about our intentions, which were incredibly loose and never really materialized into anything other than a heliotropic snow-flinging smart surface. Here are some good snaps of the surface.

Week of...

The final fabrication of the surface was incredibly difficult to plan out mostly due the dependancy of certain tasks on the completion of other tasks. The coordination between group members, more cnc router time, and certain jobs proved to be the most arduous of the tasks. After Tuesday, and after the formica was finished routing we knew that time was incredibly limited so we worked all night to get the cells glued and soldered together. The teamwork really came together this week, and although there had been a few disputes between the work of a few members of the team, I feel that basically everyone was just ready to work. The night prior to the project was a tense one, with only a few of the cells working consistently. We had anticipated a few set-backs, and of course a few servos burnt out that were already screwed into the structure. We adapted and decided to have 1/2 of the cells move, which actually began to work out nicely provided a visual contrast of the motionless cells to the moving ones.

The architects had a structures exam at 830 the morning of our review, so we had to retreat a little early, but luckily Yuming stayed longer to get the code working really well and Mat came in quite early the next day to finish the structure.

I must tip my hat to Yuming, who was probably one of the hardest workers on the team-he pretty much put his head down and worked for the entirety of this project without any qualms or complaints. He tried to reach out and help with the fabrication and got to drill his first hole and pick up a power tool, which was awesome.

The routing nightmare continues

Routing is still a nightmare, and the only way to counteract it is to sign up for about 4x the amount of time that should be needed to complete any given file. In efforts to highly decrease the time necessary for routing use, I was able to ask Maciej, professor of Digi-fab and highly skilled with the router and materials, to help deal with the plastic and set up the correct files/bits to cut with. The most recent cell was not a cut-and-dry mill session, but rather Mat witnessed one of the fiascos mentioned in a pervious post and watched the bit go the wrong way around the material to produce terrible cuts and ultimately break and ruin part of the material. We were excited when Maciej agreed to help and spent the first portion of the week tweaking the design and setting up the files to be routed.

Much of the time you will be getting a graduate student to help route, and generally it is someone inexperienced, frazzled, timid, or overly ambitious. This is always worrying given the small amount of time given to routing and the large possibility for error as well as the large amount of money spent on the material. This particular Friday we had a grad student named Gretchen who really worked out great and was interested in getting our material routed in a quick and effective way, essentially the group really got lucky getting such a cool person for router help.

We were off to a relatively quick start and Maciej and Wes were able to watch the initial cells being cut and micro-adjust the speed of the bit to get the correct edges for the cells. Each row was routed into the .5" plastic and we were able to get all of our rows routed on Friday. I was incredibly excited that all of the rows were routed because I knew that it would permit us to do more week on the weekend as well as get everyone excited to keep moving on the project.

Routing 11/21

After the initial routing for the foam cells, the group decided that we badly needed an exploration of materials. Mat and Michele were able to run to the store and find a .5 inch black plastic that seemed to fit in with what we were looking for. It was durable and it was able to resist most of the bend that we were fearing after the decision to get rid of the lone cell and form the surface into rows. We were slightly worried that .5 would be too much for the servos to turn, but it seemed to pan out in a few simple tests that the group conducted.

Routing=Nightmare

Routing is always an issue when a tight schedule is to be adhered to. First the file must be made into a file that is easily transferred into mastercam, meaning that the rhino file must be a series of curves and surfaces that are to be mapped into routing paths.

The mastercam file then needs to be set up by a worker at the fab lab, and I have tried a couple times at making my own file only to be told that it has to be changed dramatically or redone- which is to be expected without really ever having any tutorials or training. The curves can be made into various paths for the router including contours (our cutlines are contours) and pockets (where the subtraction of material occurs below the solar cell.

The correct bit must be chosen between a myriad of different bits with various numbers of flutes, and we must input this into mastercam.

The largest and final step is that the routing must go smoothly, because quite often bad things happen when routing including broken bits, rouge pieces of material moving, incorrect calibrations, and overall human and machine error.

Mat was able to route these gems before thanksgiving, allowing the group to get a good grasp on what we needed to aspire to.