[Thur May 3] LAST CLASS

Today is our last class. Some groups still needed some time to finish up their projects so we waited until 3 to do our final demo.

In the meantime, Oscar asked Essie and I to duplicate our previous designs for future reference. Therefore I found my final Inscape design for my box project and started cutting.

I remembered that I used two cuts on the 3/16 Delrin: first cut speed 0.6 and the second cut 0.9. However, it didn't cut through. After ten minutes of poking, only one piece out of three of my design came out. Therefore, I tried first cut speed 0.5 and second cut 0.9... AND I SET THE MACHINE ON FIRE... 

That was that. Oscar helped me clean up and said he could cut it with me. We tried speed 0.7 for three cuts and it cut through but was very melted... Oscar suggested me to wait for a little while before I do the third cut next time. I was really confused why two cuts (first time 0.5 second time 0.9) worked last time but not any more. Oscar said because 3/16 is unnecessarily thick for this project and the thicker the material is, the less accurate it gets. That made lots of sense because I didn't spend any time exploring different materials. I just grabbed something and it turned out to be 3/16 Delrin. After looking at similar designs such as Kelsey's box, I thought it would have been a lot easier if I used thinner materials. Lesson learned.

Finally it's time to present our project!

Essie and I went first. Our design had a really complex structure and really easy circuits. It was very different from what we thought it would be. Here are just some examples of the ideas that we gave up on:

1. Screaming. I couldn't even get the beeper working when I had time to develop on this idea. When I had the resource of asking how to integrate the beeper, I was more occupied of getting the basic features working.

2. LEDs light up on top. We wanted it to be a "dice", so ideally when it lands at the bottom, the LEDs on top should light up. But by Wednesday evening, we had our design of LEDs on the bottom light up because it's a much direct and easier contact to make, which involves no other metal plates and free fall issues. We also only had four button batteries, all of which were used in our current prototype, so we didn't want to risk breaking the old design and build a new one that might not work given the limited time we had.

3. Change batteries. Essie was thinking really hard on making a battery case that enable users to change batteries if needed. But given the complexity of the inside structure, that was not our priority.

4. Six faces have LEDs light up. That would involve a very complex channel design if channels work at all. 

Some of the little designs that we had to make our prototype more functional:

1. Sliding channels. We wanted to make sure that the circuit have really good connections so the inner structure needed to be tight. The best solution we came up with was to build a channel just big enough so that the battery case can slide freely. That actually didn't turn out to be as hard as we thought. Of course, that was mostly due to Essie's accurate calculations.

2. "Springy" LED legs. In order to make sure the connections are good, we bent the legs of the LEDs and they acted as springs to make better connection. (More detail from last entry)

3. No screws! That helped to save space. Although our design did become more fragile (not necessarily because of screws I don't think...)

We didn't get much feedback from the audience. Hopefully people liked it.

Then we went to look at Erin, Hannah and Kelsey's project - a Harry Potter book.

I loved it. It had a fairly complicated circuit and lots of interaction. It is probably the most entertaining project of the class. I liked how they used their bi-LEDs by having different colors light up according to different plots (controlled by switches). One thing they could work on, as they mentioned themselves, is to have a separate power supply so this book can be "read" anywhere. I liked how they used paper and cardboard as part of their design. It gave them more space to be creative than working with wood, Delrin or acrylic.

Caily and Maria made a sound detector that a red light would light up when it's too loud. I loved the idea and I've seen some applications of that somewhere in China. When it's too loud, a red light will light up to warn people. The circuit is fairly complicated with a beeper. They haven't had the time to make a outside box for it, but the general idea was there and it was cool to see something I saw in real life done by fellow students.

That was the end of the class... We wrapped up by talking about different branches of engineering and future exploration of engineering in general. I truly learned so much from this class in many ways so different from what I imagined:

1. Engineering is not a mysterious subject that only really science intense people can do. Actually one thing cool about engineering at this level is that I don't have to know why something works. I just need to learn how to make it work and apply it. I was not very comfortable with this in the beginning but I learned to go with it and explored.

2. Engineering is very practical. I had some electrics background before coming to this class (although I started to get lost from the third week of circuits...) However, what I learned might not be applicable at all if a tiny thing went wrong. For example, wires have resistance, which might mess up the calculation of the circuit. Using Inkscape to cut materials can be equally frustrating because the smallest mistake will simply screw up the whole design.

3. In the beginning of the semester, I felt every design was harder than I thought, especially after starting our first box project. The simplest joints can take up three weeks to figure out. Towards the end of the semester though, many things seem simpler than I thought they are. For example, there is no way that I think I could have done something like our final project at the beginning of the class. The other two groups design would also have seemed impossible for me. Now I am more confident about what I am capable of doing and learning to do.

I have been asking around for advice for my future exploration of engineering. I would like to start with a more traditional way, but I am certainly hoping to learn more about engineering.

[Mon Apr 30] Last Class Before Demo

Today was our last class before the final presentation. That was intimidating... Essie worked on the design a little more during the weekend and came up with a channel inside the dice so that the battery case can slide. The channel was necessary because otherwise the battery case could fall all over and jam the LEDs and other circuit parts. With the channel, we were able to control the path of the battery case. Although we were still having problem with lighting up the light on top when the connections were made at the bottom.

Given the one way channel we had, we could make our dice light up when landed on two sides but not the other two. Essie was saying that we could simplify our design and make a "shaky" dice that would light up different LEDs when you shake it in one of the two directions, although that would not be a dice any more...

Therefore, we asked for Oscar for his expectation of the complexity of our design because if we still want to roll our dice, then we would need to have two channels, which will make the structure extremely difficult (at least that's what I thought). Oscar had a discussion with everyone to see where every group was at. We also had a discussion of user interaction because that has been a little vague... Oscar said the user should be part of the process. By the user doing something to a machine and the machine does something back, the user should be able to react to the results of the machine. An example of a good user interaction is Maria and Cailey's project. They use sound sensor to detect the volume of the sound. If it's too loud, the red light would light up and the user will lower his/her voice accordingly. If the volume is within the standard, no light would light up so the user knows that he/she is not too loud.

Therefore we concluded that we should make our dice roll instead of just "shaky" to make it slightly more interactive although we are missing the piece where the user can use the results coming out of the machine. I guess it could possibly motivate you to roll again and keep playing with it. I did think again about our "screaming" idea - the dice will scream if it is lifted up so the user would know to put it down. But at this point, I am not sure whether we will have the time and energy to integrate sound in our design.

Anyways, in order to make it roll, we need to have two channels and two sets of battery cases. That was completely out of my capability of mapping and Essie successfully designed the two channels in a relatively short time.

We also had a revised version of our battery case since our channels were smaller than the battery case we built last week. It was still made of small identical pieces wired together.

Taken from Essie's blog.

Drawn by Essie. We were going to make a prototype of the battery case but we used all four button batteries that we had. You can see the battery case in our final project pictures.

Here is a picture that has a slightly better view of the batteries:

We cut the pieces and assembled the battery case. We experimented it in the channels that we have. The battery case would not slide very smoothly but the general intuition was right. Therefore, we just needed to make the battery case pieces were even smaller.

While Essie was working on Inkscape, I thought more about the connections and kept bringing up problems so that Essie could take them into account and tried to fix them with the structure design. I was concerned whether we were in the right direction to spend so much time on such a complicated structure design like our lantern project, but Oscar said the structure is always harder than the circuits...

Meanwhile I tried to find a way to connect metal pieces with acrylic to make connections. I tried Glue-All, but it didn't work. Hannah trained me to use hot glue guns. It was really fun to use but it still didn't really work. Double-side tapes would work fine but we don't seem to have them in the lab so we will have to think about some other methods somehow...

Then class ended... Next time we met up, we already had our basic structure done. Well, basically all done... We added some features to our structure to integrate the LEDs in our design:

Each small piece from the left is used to "close up" the openings of the channel. Actually, they are used as a platform for LEDs. The two legs of LEDs go through the holes and were bent back inward in a hook shape: 

Taken from Essie's blog.

Look very carefully on the top of the inner structure. You can see the hook shape LED legs.

We (sort of) solved the problem of "battery case can't free fall" and "can't stick metal piece to acrylic" by simply lighting up the lights on the bottom. In that case, we can simply complete the circuit inside the battery case so when it touches the LEDs, the LED lights up.

Here is the design layout from Inkscape:

Final project:

(I was going to cut it but the first part was kind of funny... Anyways, starting from 0:22 is the real demo.)

Please refer to the zip. on the class website for detailed Inkscape design and instructions.