Kate Hollenbach / MAS.863 How to Make (almost) Anything

I'm creating a "digital rubber stamp" as part of a project for another class. The stamp has a simple circuit that connects an infared LED to a battery, switch, and resistor. When the switch is pressed, the LED turns on, indicating that the stamp has been pressed. The switch is on the bottom of the stamp, while the LED is on top so that it can be seen by an overhead camera. I thought it would be interesting to try to design a mold that would help this device "feel" like a stamp.

I first tried designing the casing for the stamp using cad.py. I think the design turned out okay, but I had a really difficult time using the Modela to cut the mold.

There were two parts to the mold. One side was round, so that it would make the stamp easy to hold. The other side of the mold was to extend the handle and attach to the top of the battery casing. Here is a picture of that design in cad.py:

I built this model by starting with mold.cad and modifying it as needed. I had some difficulty getting cad.py to produce good contours for the Modela, and didn't realize until too late (too close to class time) what was causing the problem. The sample code had a square model, that used the same "nx" and "ny" values in cad. But because my mold was rectangular, this resulted in gaps between the y contours. For reference, my modified version of mold.cad produced this sort of output, where there is too much space between the vertical contours:

I had to settle for a less precise model by building the casing out of moldable wax:

I then used Insta-Mold to create a quick mold of the handle:

Determined not to let the Modela get the best of me, I designed another mold with cad.py to create the bottom of the stamper. It's a rectangle a little wider than the bottom of the battery case, plus it has little "feet" to keep it up off the ground so the stamp can rest on the table without activating the switch on the bottom:

Although the casing on top wasn't as clean as I would have liked, it using rubber for the bottom of the stamp worked beautifully. The "feet" were strong enough to hold the button off the table unless the user applied a force to the device. Plus, the squishiness of the rubber made the device feel more like a stamp, and made it easier to press the button. I'd like to try it out using capacitive sensing instead of a switch, too, like Neil suggested -- hope I can find the time.