Sophia's Soda Cup
may or may not be for cat-people, gamers, and basketball players… Stay posted, there is more to come
Design Drivers
-My design was driven by length of fingers (level of dexterity), outside temperature, a constant (minimum) volume, and how that effects your beverage container needs.
-I assumed that to hold the container comfortably, each person’s fingers would have to wrap a certain distance around the container without overlapping.
-I also assumed that that those with the longest fingers would have the most control over the glass (because though the grip scales according to need, the volume does not) and there fore determined that those with the longest fingers could have the roundest opening on top.
Finally I assumed that each consumer wanted to be able to fit a full can of soda in their glass at a given time (which gave me a min volume of 12 fl. Oz.)
Logic Diagram
 |
CAD model v.1
This is the original model I made, still with some major kinks in it. Feel free to play with it. If you are having trouble the only real help I can give is that to change the views you have to go through manually and change the bottom ellipse, and the middle ellipse. There are also a few problems with the surfaces which you will find if you try to change it to much.
sophia_v1.catpart
Incarnations for v.1
It started off well. Notice the gaps that start to appear. The round shapes turned out nicely because of the way I had to create the surface. Finally my very favorite model of the entire project. It shows a complete failure of efficeint parameterization, but is still both beautiful and interesting. It was also completely unexpected.
 |  |  |  |  |
 |  |  |
CAD model v.2
THis is my second model. To manipulate this one, play with “Base Aspect Ratio” and “Ellipticality”.
siawkittydance_the_one_on_tuesday.catpart
Incarnations v.2
I could still get that Ufo shape. Here I tried to see how skinny I could make that champagne flute and turned it more or less into a go-gurt container. I just wanted to see if I could get my model to explode the way my first one did.
 |  |  |  |  |
 |
Lessons Learned (or so I hope)
One of the first things I learned while working on this project is that it is impossible to design something you don't understand.
Originally I tried to fit my three container models into one over-arching parametric design which seemed impossible at first. Once I identified what features defined each model, I was able to choose the primary feature of each.
It is very easy to over-complicate a project if you do not know how to use the program to your advantage
As I got a little more comfortable with digital project, I realized that sometimes the program has a different logic than the two cad programs I had have played with in my life. Being unfamiliar with the program (and it not having the best oversight), I assumed that it could do everything I wanted it to if I could only find the right button. This was true in theory, but it involved not only looking for the elusive icon or command, but also thinking about the whole problem differently.
Starting over is often the best way to move forward (though Pride may get in the way)
Late one Tuesday night I decided I had had enough and decided to call it a night. Even though I had a product to show in class, I was convinced there was some way to convince digital project to do what I wanted it to. I realized that working with the old model would be difficult since I could not remember what exactly I had constrained to what to get it to limp along. Starting over allowed me not only to have a clearer idea of what I was doing from the beginning (because to be honest I just jumped in the first time), but it also allowed to me stay a little more organized. It is also somehow easier to recreate something than to destroy something that took you so long to build in the first place, no matter how bad the first incarnation is. Finally, it helps solidify and streamline you model since you put everything in with intent rather than just hoping that it will work out in the end.
Major Problems and their solutions
Quite epically, I had the problem of trying to put a surface on my elliptical based (changing parametrically of course) cup with sides that change from concave to convex. I pieced together a skin using the multi-surface tool for wednesday, but knew there must be a better way to solve the problem. It looked alright for some of my manifestations of the cup, but it became abundantly clear that I had not figured it out in a view I call failure. In the end, the multi-surface was the right tool, but to get it to work had to make the other surfaces guiding ellipses around the z-axis (on the xy plane) rather than trying to use splines and an ellipse as a more literal skeleton the way we did in class as an example. This did the trick and lead to a much smoother Soda glass v. 2.
Of course this led to a new unforeseen problem…
Constraining Ellipses
I am still working on this, any tips are appreciated…
Things that made life easier while working
-Projected points.
-Named Real parameters.
-Adding Planes.
How to post on a wiki
By the way, the links will hopefully all work as soon as I am on a computer that can upload what I want to put there.
Discussion
I want to say that I absolutely agree with your comment on “Starting over is often the best way to move forward.” I also had to restart creating my model several times just to try a different strategy to get the model working as I wanted it to work. The good thing about this is that the second or third time you recreate the model it doesn't just become significantly faster but usually you also find several thing to improve from the previous model.
…and you don't have all that clutter in your head of what worked and what didn't mixing so that you are not sure what you have actually built.
It's interesting to see what process you gone through with the design of your “Soda Cup”. The ellipse form at the latest result is a good shape for a hand. I met the same result when I designed a welding handle and the ellipse form came out as a good solution for a good grip handle (se Gustav's page).
About the CAD-software Catia I realize that it is not just like “walk in the park” to understand the logic, but by practicing it helped you. :)
I'm fascinated by your goals and drivers. Looking at your logic diagram, I'm interested to learn more about the relationship between hand size and temperature. I was with you in the computer lab and got to see some of your “surprising” designs as they came up and I have to say, they are by far some of the most interesting ones of the class - particularly incarnation v.1, 2nd from the top right.
Most of all, I really appreciate the extreme nature of the shapes you came up with. I think this made for a maximum exploration of your parameters which seems to be an ideal exercise for the first CAD assignment.
Also, I'm interested in learning more about your use of skinning with the multi-surface tool. This sounds like it could be really helpful in the final project on architecture so it will be interesting to see how those concepts could be applied.
Very impressive work, Sophia! I think it is a great idea to have the CAD files available for anyone interested to explore them on their own. I immediately opened your version 1 and started to alternate it. I also agree with Fernando, that the advice to “start all over again” is a great one. I'll will surely return to this page when we Swedish PhD-students are about to do our CAD assignments! Thank you.
Erik.
It sounds like you really went on a journey with the CAD software. As Claire said in class I think the fact that your ideas were so unconventional really helped you to be creative with the CAD, despite the failed attempts. You taught me that I need to think a little crazier next time, and maybe something more interesting will come out of it. I also think that the first method of creating the container was still pretty ingenious even if it didn't work out completely.
It is surprising you can get three totally different design by controlling parameters. Also I quite impressed about analyzing the undeserved models and getting deeper undestanding about your model.