17 December 2007

"Lessons Learned...the hard way" :: Advanced Structures :: Project #2

My second large project of this past semester was for my Advanced Structures class. Despite my initial concern about what exactly "advanced structures" meant, I really liked the class. The stated goal of the course was to develop an intuition about structural design, not focusing so much on the equations but what they represent and what they mean in the grand scheme of things. In this respect, I think the course succeeded. Also I was glad not to get so caught up in the math, because while it generally doesn't cause me too much trouble, I think the graphic analysis methods we practiced in the class were much more effective.

Ok, so the project, as I stated earlier in the semester, was to take a train station and redesign some piece of the structure of it. We chose Glagow Central Train Station in Glasgow, Scotland, because while it's a really beautiful station, it's also quite old, and in the theoretical scheme of things, it could use a structural update. Here, again, is what the original station looks like:

My partner for the project and I decided to go with a much lighter structural scheme....the original station has an all-glass roof, but it's supported by a lot of bulky structure. Since it was built in 1879, bulky structure was the best they had available at the time. So, we picked a particular type of cable truss which is both functional and attractive and we decided to adapt it to our conditions. We liked the cable truss both because it brought the structure outside of the space, and because the curve of it related to the curve of the existing (to remain) structure in the other half of the terminal.

Our final presentation consisted of our original board, which displayed historical information on the station; a new board, which had schematic design drawings and details that we designed; and our design and testing model, which I will talk about in a second. This is what the overall presentation looked like:

Now, about that model....my partner and I decided to use piano wire, since we were representing a cable truss, and metal tubes for the posts, because we figured we should keep the material consistent. This meant that we needed to solder pieces of metal together, which neither of us had done before. First we tried aluminum posts....and it turned out that you can't solder to aluminum at all. On to plan B, which involved soldering to brass posts. This worked just fine, and we finally managed to solder the piano wire together as well, but we didn't trust any of the connections, so once the solder had cooled completely, we covered all of the joints with 2-ton epoxy. Which took care of the joints nicely. On to the tensioning of the trusses.....throughout the modeling process, we discovered quite a bit about the design that we chose, and it turns out that it works better when the whole thing is pre-stressed. Which we didn't really do. So we tried to make up for it by putting a lot of tension in the anchor cables....with thread. Strike 2! Thread, even heavy duty thread, doesn't work in this application, so we had to use nylon instead. And this worked as well.

When it finally came time to test the model using weights on the testing stand, the model didn't break (thank goodness), but it deflected a LOT. But since most of the members in the model weren't acting as they would in real life, we expected this. It was a slightly stressful exercise watching such a frail-looking model deflect so enormously, but I learned a heck of a lot from the whole ordeal and I now know how to put together a model using metal a little bit better.

our testing model with 8 pounds of assymmetrical loading

In my last installment of the series, which could be called "My 3 Projects at USC FA'07", I will talk about frustration and the trials and tribulations of lighting design, and why I didn't sleep much in the last week. Stay tuned!

No comments: