As part of the Numerika2Neets projects the participating labs made Slow motion frames like https://www.instructables.com/id/IKEA-Frame-Hack-SLOMO-Slow-Motion-Frame/ to compare different strategies for making things. Which was nice, because i always wanted to have one :-) Of course, a simple rebuilding would be boring, therefore each of the labs tried something different. For me, main ideas was RGB leds for color changing lights (not the best idea), and a frame which could be used for watered plants.
Teachers! Did you use this instructable in your classroom?
Add a Teacher Note to share how you incorporated it into your lesson.
Step 1: Design and Lasercutting
For the water bottle i choses a small chocolate cream bottle from Cusco (from https://www.chocomuseo.com/ - it was very good) - sadly i later lost the bottle but could restore the bottle sticker from the pictures of the frame... but that's not important ^^.
More importand was designing the frame around it. For a more dynamic frame the concept was using two trapezoids, one for the bottle, one for the plant, with the vibration system and driver in the connecting middle and LEDS on the outer sides, covered with diffuse acrylic.
For easier design, i first made a sketch with sqares in inkscape, then converted it to the trapezoid structure. For fingerjoints (because of the non-rectangular edges), the sketch was then imported in cutcad (our own designtool for these things) and fingerjoints made with this tool. After a bit of clean-up work in inkscape i could the cut the frame with a laser cutter and assemble it with wood glue.
The frame is then stained for a nicer look.
Step 2: Electronics
LED stripes are the glued on the edges. For the first electromagnet version i used 12V stripes, later i reverted back to 5V RGB stripes to have a USB-powerbank-powerable system.
A small breakoutbord for the electronic was deigned which fits nicely into the middle beam. On the bottom beam additional holes are added for variable resistors, a button and power supply. LEDs are the covered with 3mm diffuse plexiglas sheets, hold in place with small pieces of transparent acrylic, glued with double sided glue tape on both sides of each beam (3mm distance to the open side where the diffuse acrylic will cover the rest).
And then the problems begun :-) The first thing is of course that the PWM-library don't allow unlimited PWM pins with variable frequency on the Arduino Nano which i used - 3 worked, enough for the RGB colors but the one was missing for the magnet. Also, the magnet holder wasn't that easy to construct, if it should partly go inside the water bottle, dissassembly would be therefore a nightmare.
Luckily, at our first meeting comparing ideas Charles-Albert de Medeiros, the founder and lab manager from Fab Lab Lille had the idea that vibration motors should also work for such a system. Since frequency depents on applied voltage, i could then just switch to this idea and use a small vibration motor, with a resistor for voltage settings. Of course we can then only control frequency and not that much the movement length compared to the electro magnet, but for small movements of leaves it's enough.
The vibration motor is then taped onto the plant, hold in distance to the wooden frame with a piece of foam rubber (which also dampends the vibration to be only happening on motor and plant, therefore almost without sound). As additional benefit, the vibration motor allows to easily stay within the 5V range.
Step 3: Final Thoughts
Of course, the RGB color still needs some fine tuning: Since now all three colors are switched after another, the resulting image becomes blurry: Since a small timeshift exist between each time one led color is on, different positions are enlighted at the - for the human eye - same time. Thats the reason why on pictures instead of "gaps" as for unicolor LEDs i got the rainbow effect on the second picture - for human eye it's cold white (maybe a little bit blue-ish).
Looks surrisingly funny (like the old two-color-3D-images), but of course not intended. Using just one base color eliminates the problem, but thats of course a boring solution.
So in the next steps i would have to dig into register manipulation, basically using the same timer and switching the colors in parallel with a bit mask or something like this.
The easier solution is of course either revert back to white leds, or live with the system as it is :-)
Participated in the