Since a flashlight was described loosely, as something that is portable and that creates light, I wanted to move beyond a cylindrical flashlight-looking object. As our first fabrication class covered drilling and making holes in different materials, I wanted to incorporate this in the flashlight in some way.

[ and since out first physical computing assignment was to make a simple switch, I'm killing two birds with one stone.]

The initial idea was to create a portable hand-held cube, that had a light inside, and holes on the surface of the cube for the lights to come through. Since portability was important, the material needed to be light-weight enough to carry around with ease. I thought thin ply or balsa wood would work, though I wondered about how neat the holes would be with such thin material.

The other question at this phase of ideation was around the circuit. Should there be a mechanism to switch the flashlight on and off? Could I use conductive paint or copper tape to make a circuit? If I left a part of the circuit incomplete, could I close the circuit by touching it with my finger? (I assumed not, that our bodies would have too much resistance). I was thinking of the paper-led circuits I've seen, and how that could work well for this – lightweight and simple, and powered with coin cell batteries. Working with this, I though a simple mechanism could be to have a loosely attached part on the outside of the box, with copper-tape on it, that can be flipped to complete the circuit (sketch above) – so the majority of the circuit would be inside the box, with a little bit on the outside, coming through the holes, or a gap in the plywood. Instead of holding the flap closed to put the light on, one could simply use the weight of the box, and rest the light on a surface instead. (I also contemplated a sliding mechanism, thought it seemed far more complex). It then occurred to me that this mechanism could work just as well inside the box, and it would probably be nicer to have it concealed.

This seemed like a good starting point, so I went and got some cheap craft plywood, and got to work. I wanted to try drilling some holes first to see if I could get them neat enough, or if I would need to reconsider the material. I'm glad I tested this – as it turned out, I had the wrong drillbit.

Clamped up and ready to drill
The 2 holes (and 1 semi-hole) were with the wrong drill-bit. The hole on the right was with a wood drill bit, and did a better job

Once I had a wood bit, the hole was better, though still not very neat – probably because of how thin the ply was.

The other side of the ply. Left hole with the correct bit (for wood), but still not the neatest.

I decided to move on, and focus on the circuit part, and come back to this later. If time permitted, perhaps a laser-cutter would have been a good solution.  

Ready to saw.
First part of first prototype
Added the circuit

The first prototype was a bit finicky. The connection at the flap and at the top of the battery were sometimes too weak.

If I pressed down, the LED would work.

I then tried taping down the copper tape-battery connection, and adding some solder to the connection points of the flap and tape, hoping to make it easier for the flap to make good contact with the rest of the circuit.

Flickering light, not the most reliable or stable connection

This worked a little better, but was still not the most reliable solution. A heavier flap might have helped, though I wasn't sure what material I could use. Since I had taped the two pieces of ply together, I also noticed that the slight movement of the 'standing wall' made the connection unreliable. So I decided to test again by trying to cut the wood as neat as possible, and use wood glue to stick the walls.

Trying to create a firmly attached 90deg angle

This still didn't make for the most reliable connection. I asked Ben for some help at this point. I learnt that what I was trying to make was essentially a tilt-switch (same as a mercury switch?) – something to keep in mind if I develop this idea further. Ben suggested putting a weight on the flap, and elongating it, with the weight at the end. High-school physics finally came to some use – the force exerted by a weight is greater if its further away from the fulcrum.

Putting a weight on the flap helped

I made the flap longer, and glued the nut onto it. I also wasn't happy with the brightness of the LEDs, so tried putting 2 coin-cells together to allow me to use 2 lights. While testing, instead of moving the flap around, I put a bit of solder in place of the flap to complete the circuit, and realized that I incorporate this onto the flap.

Longer flap, weight on the end, and some solder on the flap to help complete the circuit. 2 coin cells for 2 LEDs
The completed circuit

Ben also suggested leaving the back of the cube open incase I needed to get in, and thinking about just one surface as the 'front' of the cube, so I could just focus on that side for the more 'decorative' element. I decided to create another glued wall, and tape the the 'ceiling' of the cube so I could remove it if I needed to make adjustments, and use paper infront.

Finished cube
It works :D

The next iteration of this could probably use a real tilt switch,  laser cut pieces of wood, brighter lights/more lights, a battery holder for the coin cells (ideally in a way that they could be replaced), perhaps a mechanism to open and close one side of the cube to get to the batteries, and diffused acrylic on all sides.