My long-running interest in trying to engineer myself a green thumb.

Photogrammetric reconstruction of seedling in hydroton pellets.

This is more of an 'origin story' than any sort of display of credentials or the like. So if the background you seek is one to showcase competence...I'm honestly just not sure how ya even got here.


Back around 2008, my dishwasher broke on me. So naturally I got to work disassembling to find out how it worked before my partner got home from work. I learned there’s not much to these pricey little boxes, but they do have 1) an integrated water pump, and 2) some bright ass white walls. So naturally my first thought was, “I wonder if I could use these fancy new Compact Flurescent lightbulbs inside of this thing to grow plants from seed to harvest?” Ok, so yeah, I honestly have no clue how the hell I made that logic leap, but it happened….here’s proof:

While this DID manage to get romaine heads to full ‘length’….it did so without much of that girth that a decent salad so desperately desires. But oh well, after seeing that this crude setup wasn’t going to work, I of course just pushed this silly thing aside and got back to my studies (I was in grad school at the time)….Just kidding, here’s Rev 2:

Yeah….don’t judge me…I would say I was ‘young and dumb’…but only half of that has changed much. If you’re admiring the kitty litter container hat, why thank you. That is, believe it or not, the air conditioning. Yeah, I for some reason thought a swamp cooler was a good idea and that Mylar Blankets were somehow also helpful. Believe it or not, this actually did work a bit better, but by this point I was starting to get curious about reducing the heat by improving the electrical efficiency side of the equation.

To start to get a bit of a more refined handle on how to optimize this setup, I decided to rig up a ‘cellular’ grow system (yeah, for some reason I was just fully ‘in it’ by this point…things don’t really get any LESS absurd from this point on, but I think there does start to be at least a smidge of technical competency showing through a bit later). You’ll notice a favorite construction material of mine at the time, cardboard panels, taking on a starring role! The objective was to be able to configure each of the 12 cells in the shelving unit to be an individual ‘experiment’.

And this is the ‘control center’ for these shelves about a year after the initial assembly one above. Sadly, I can’t find any of the videos of my ‘automated photography system’ that I had cobbled together on this thing…it wasn’t very effective, but I was VERY proud of it :)

And here you can see some oregano under some cardboard panels with blue and red LEDs. You can also see my NIST traceable light sensor :) It was actually just two flexible solar panels taped to some PVC, but it actually did give a nice proportional Voltage to Intensity output that at least give me SOME sort of quantified metric with which to compare setups!

Unfortunately, High Power LEDs were still FARRRR too expensive for my broke ass to buy for any of this. So I was buying 20mA LEDs (aka ones you’d use for a decently bright indicator light) and then patterning those bad boys allllll over some cardboard circuitry! Here are just a few of the cell designs I built and tested:

I even experimented with scaling this approach up to suit larger, fruiting crops like tomatoes and peppers:

...WIP... If you'd like me to prioritize finishing this, let me know on Instagram or Printables

No comments

Leave your comment

In reply to Some User