-
New boards! Better patterns!
04/15/2017 at 14:35 • 0 commentsAfter more user testing it turns out that the blink at the end of a cycle disrupts how the light is perceived. The word 'jarring' came up many times. So I changed it to a smooth fade, and it seems like people like it better!
The new boards from OSH Park came in and I've finally soldered them up. Since these new chips are the low voltage ATTINYs they should be able to get a longer run time on the same batteries. I've also added debugging outputs to the chip so that I can change some code if I really have to. It beats desoldering the ATTINY85s just to reprogram.
I figured out a nice way to hook the end is to use a craft earring and bend it into a looped shape.
I did some mechanical stress testing on some of the my older versions and I found that if I dropped it many times from chest level the battery pad has a tendency to come off. To fix this I have weakened the holders so that they don't push against the battery as hard.
-
More LED testing
04/06/2017 at 20:10 • 0 commentsI did more LED testing the other day and I concluded that the current supply can only reliably sustain 1 LED of draw current. Any higher than that the microcontroller is not able to boot. I have a few new boards that are coming in a few days and I look forward to putting them together.
I also put together a programming jig because I've had a few people who are interested in obtaining one. This will allow me to program the SMD chips easier than hooking the board up to test points, although I've included them to enable me to update the code.
-
Covers and Diffusers
03/31/2017 at 18:03 • 0 commentsWhile I like the board itself, some people might want a cover to diffuse the LED. After some experimentation, I found that white PLA plastic that has the LED flush to the surface actually provides the best diffusion. If there is a dome, the light doesn't diffuse that well, causing uneven brightness as seen in the gallery. Transparent plastic doesn't work that well either when it is printed. If you want to use a clear cover, try using a glass dome meant for circular DIY pendants.
I used 1mm thickness for the plastic.
-
Battery Stress Test Complete!
03/31/2017 at 01:26 • 0 commentsBattery stress tests currently show that the device is capable of up to 26 hours of continuous usage. A total of 4 reboots occurred during this period. Under normal usage, this is projected to last about a year.
Currently the device is still blinking. However, since the duration of the reboots was 2 hours, I decided to stop the tests there.
-
Version 1.1
03/29/2017 at 13:28 • 0 commentsSome potential updates to the algorithm used to simulate breathing:
- Fade in - blink - lit - blink - fade out
- Fade in - lit - fade out
- Fade in - lit - blink - fade out
Some changes to the design include adding breakouts for all pins so that the program can be updated, and testing with a second LED for alternative patterns of light. By adding breakouts I will have to rotate the ATTiny85 so that the leads don't cross. Because the battery is on the other side, I am unable to put vias in the center of the board because I don't want it to short.
-
Version 1.0
03/29/2017 at 01:38 • 1 commentAfter a few iterations the current version works. A battery stress test is being run to get an idea of the average battery life that this device will produce. Anticipated usage is about 15 minutes (personally I use it for about 4 cycles, which is 1 minute), so if the battery is able to last 8 hours this will give the device about 32 uses before the battery is flat.
- A month of hard use
- 15 months of light usage.
Future improvements will include diagrams on the board. Possibly an AUM in honor of this practice's origins.