I'm hard at work on the second/final version of the business card. Currently I'm figuring out some of the creative elements, basically how the final version should be layed out such that the front of the card looks clean and professional with my information printed on it.
I'm also refining the circuit design just a little bit. I've made the decision to remove the two slides switches as they are horribly unreliable. The little tab on the switch is difficult to move with a fingertip, and the tab is prone to breaking off. I've also had some of them rip up from the board. I'm reworking the power management a bit to use a single pushbutton switch as a wake button, with the IC itself being able to put itself into a low power state when it detects inactivity (both by monitoring the accelerometer and by the light level on the ALS). This should also make for a nicer "user experience" as the card can be turned on by merely shaking it.
While working on the new design I wanted to test the photogeneration hypothesis, so I carefully soldered one of the white
0402 LEDs onto a little piece of prototyping board I had. This proved really difficult, since I'm back at school now and didn't have immediate access to a hot air station. I recreated the circuit I have on the card, just a series current limiting resistor, and then measured the voltage across the two components with my Salae Logic as I shined a very bright flashlight onto the LED. Here's a video of the process.
The video is awful, but it's the only documentation I have unfortunately. You care barely see the tiny LED between the two blobs of solder next to the relatively huge
1/2W resistor. Near the end you can see that over a half a volt was being generated by the LED. I'm not exactly sure this is enough to trip the low voltage programming mode, but I can't think of anything else that's probable. Theoretically the photogeneration rate should increase as the amount of incident light is increased, and it's possible that the ring light on the microscope was either brighter, or of a spctral content that was more closely matched to the bandgap of the LEDs semiconductor compostion. -- Note also in the voltage readings the flashlight pulse width modulating it's LEDs.