The PCB is made of three parts. The LED-PCBs are soldered at a right angle on top of the main PCB. The LEDs are charlieplexed. Two LEDs will light up one sheet of plexiglass. The glass itself is 0.5mm thick and 8x16mm wide. It's just a first proof of concept. I'll work on better light separation of the LEDs/glass plates by now.
Are you scanning the charlieplexed LEDs or are the constantly on? I ask because I notice that the higher the number, the less bright it is (maybe it's less obvious to the human eye than to the camera). If you can make the LEDs on the lowest acrylic sheet full brightness, and PWM the ones closer to the surface down you may be able to balance the appearance of all digits.
I love the look of the stack-up from the side. A great detail for the wearable piece; it's nice to show off what's special about the design.
If you know how much LEDs have to be one you can skip the other ones and increase the brightness. That's how I did it also with my other watch projects. The schematics behind each watch is nearly the same. Since the Atmega can provide only a certain current you can also skip any protective resistor for the LEDs.
interesting, I don't think the ATmegas are current limiting though, are they? I think the spec is there so that you design for that current or less.
I've always wondered about doing peak current on visible LEDs. I know that IR LEDs have a high peak current and that's how handheld remotes are so good. Can you get a spike of brightness from an LED for a short period of time to make up for the off-time during scanning?
WANT. How much fer a kit?