- Robotic Plant 0 – Introduction
- Robotic Plant 1 – Solar Engine Design
- Robotic Plant 2 – Gallery
- Robotic Plant 3 – Final Report
- Robotic Plant 4 – Followup (Gallery)
- Robotic Plant 5 – Really Final Report
I’m getting close to wrapping up this project, and I was home during daylight, so I thought I’d put up some photos.
As I made it through the shoot, the daylight really left, so I had to adjust the exposure in post processing, so if they look a little odd to you, that’s why.
This is one of the SolarBotics cells I was using, with my own power and ground wires soldered on. This is the reverse side, and you can see the built-in Miller SolarEngine board that introduced me to the concept.
A marginally better closeup of the solar engine board, but a tad blurry.
This is the business side of the panel.
Here we have all five panels arranged as the “leaves” of the roboplant, mounted on fabric-covered cardboard to be placed in the flowerpot. The empty hole in the center, of course, is for the stem.
The underside of the leaf assembly, with a mess of panel wires, and the fabric taped on.
This is the board I put together for the solar panels, with a terminal connector for the positive leads, a bunch of 1N5817 diodes (see the Solar Engine Design for details), an additional 1N5817 for a bit more voltage drop, and the connector to the rest of the system, separated, to make it harder to mess up. Note the high-quality polarity labels on the terminals.
This is a bad photo of the solar engine circuit, on the left (ignore the junk on the right under the tape, it was my last piece of perfboard). On the bottom, you can see 4 terminal connections for the solar panel and the supercapacitors. Above that is the solar engine 1N5817 diode, and the 10uF capacitor, for increasing the hysteresis (though I’m not convinced this really does a lot). Above that is the voltage detector, and at the top is the 2N7000 MOSFET (I’m not sure this was the right choice, see the post soon to follow). The left-most 2-pin terminal is for the motor connection, on one side, VCC, and the other, the drain of the 2N7000.
Another view, from the side, showing more high quality labels (SP for “solar panel” and M for “Motor.”
Another, separate board for the supercapacitors, to make dealing with them (and reusing them for other projects) easier. Note the 100K balancing resistors in blue passing under the capacitor leads, and the jumpers on the outside to enable or disable them. There are two sets of wires going out from the terminals, one to go to the solar engine board, and one to stick out of the plant to monitor the voltage across the capacitors with a voltage meter.
Here’s one side of the stem assembly constructed from the remnants of my childhood Meccano erector set, saved in an ancient toolbox in my folk’s attic. On the right is the motor, on the left is the post holding the retracted stem (a red-striped drinking straw). At the top, in yellow, is the pulley guiding the string that you can see attached at the bottom of the stem and going around the pulley and under the horizontal brace to the gears, in back.
Here’s a view from the other side, you can see the worm gear that drives the large yellow one, and the string attached to the gear, which passes under the crossbar, up and around the pulley, and down to the bottom of the stem. When the motor is engaged (in the right direction), the gear moves, from this angle, counter-clockwise, slowing winding the string around the axle and drawing the stem up from inside the support structure.
A crummy close-up shot of the gears. Note the main round gear offset a bit to the right, it turned out to be really tricky getting everything lined up just right. Note also the ancient parts. Some of this erector set came down to me from my father’s, I think.
This is the flower pot enclosure that will hold it all, though it gets a bit tight. The square base assembly juuust fits in the bottom, and the circuit boards somehow fit all around without getting in the way. I think I got it at Ikea, I haven’t been able to use it for plants, because it doesn’t have any drain holes. I’m glad I finally had something to do with it.
See here the solar panels hooked up to their circuit board. What a mess.
This is a view of everything, all wired up and ready to be crammed into the enclosure.
This is a poor attempt at showing what it all looks like crammed into the enclosure. It turns out you can’t really take a photo in low light with a quick shutter speed pointing into a dark cylinder mostly covered by an opaque lid, and expect it to turn out. But it’s the only shot I have of the innards in place, so here it is.
The assembly, with the panels in place, with the stem partially extended and the voltage meter test leads hanging out.
This is a poor shot of the blossom in place on the retracted stem. The blossom is cut from a sheet of magazine ad, held together with tape, and stuck in place (i.e., not blooming) with sticky tack putty. I have some thread hanging out to tie down somewhere, to provide the tension to pull the blossom open when the stem is extended.
The stem extended and the blossom open. This was a rigged demo, since the device isn’t quite working properly yet.
Here we have everything sitting out on my dining room table, doing a timing test of how long it takes to charge up, and whether the motor can drive the stem completely up on one charge. As you can see, the voltage across the capacitors is 2.94V, so the device won’t be triggered until this gets up around 3.08 or 3.09V, which will take a while (if ever), in this weak indirect light.
Coming soon, the final (ish) write-up, and analysis of the design and implementation.