User Tools

Site Tools


projects:blinker:work_logs:1_beginnings

This is an old revision of the document!


The beginnings

I stalled out on the glass sphere project. It was too hard to make the glass nice, and get the surfaces wringably flat, and I had serious doubts about the strength of glass I was going to be able to produce.

Several years passed and I scaled back my ambitions. I decided I would make the enclosure out of metal, with a glass window for light to get in and hit the solar panel to provide energy for the system. Output would e via a single LED that would blink Morse coded messages.

I want it to be very reliable, which means as few points of failure as possible. The initial versions will have a very simple, inefficient circuit. I plan to use supercapacitors for the storage, for their long life. Ideally the solar panel would produce a voltage no higher than the rated voltage of the supercaps.

I'm planning to use an attiny10 to blink the LED. It can run happily down to 1.7v, and only uses a tiny amount of power.

The housing was designed to hold the electronics, potted into a cavity, and the glass window. The enclosure applies pressure to only the front and back flat faces of the window, avoiding any pressure on the sides, or any contact with the edges. A gasket seals the glass against the metal at the front. The gasket must be ductile and made of a noble (unreactive with the environment and the housing). Gold would be a good choice for this. Copper may work too. The enclosure will probably be made of Aluminium due to its chemical resilience.

Once the enclosure is clamped shut, using two bars and two bolts, or a vice, it will be welded closed along the seam.

The code

The attiny10 is very tiny, only 1kB of flash and no EEPROM. The code that blinks the LED and the pattern for the blinking must share the same 1kB of space.

projects/blinker/work_logs/1_beginnings.1642896565.txt.gz · Last modified: 2022/01/23 00:09 by tjhowse