4 ms·

Ben eater on YT has a lot of cool projects: https://eater.net https://eater.net I'm currently working on a project to do soil analysis in realtime via a LoraWa

Ben eater on YT has a lot of cool projects: https://eater.net https://eater.net I'm currently working on a project to do soil analysis in realtime via a LoraWan network (kinda like https://youtu.be/iN6j1AbUbYo https://youtu.be/iN6j1AbUbYo). Also looking at doing hydroponics, but that isn't very microcontroller heavy.

bush-bby 5y ago
agreed ha, coincidentally I just started on a similar project this weekend. One thing I've found interesting is the accessibility and accuracy found in such small modern sensors. Hydroponics may not be "microcontroller heavy", but can definitely be applied.
loxias 5y ago
Coincidentally, I also, also, started a project to do soil analysis remotely! :D Tho mine is just a few crude proofs of concept and a lot of dreaming. The eventual goal is an inexpensive automated black box box that provides me with fresh lettuce daily, with no human intervention. I'm curious, what route are you taking for sensors? Off the shelf moisture? NPK? DIY? I love the idea of logging everything that's possible to log :>
- theossuary 5y ago
  That's cool! Honestly right now mine is too :) But I'm getting the parts together now to build the first box, the plan is for it to be outside in the garden so it needs to be pretty weather proof. I went with temperature + moisture sensors [0]. I was seriously considering NPK, but I didn't like most of the sensor options I saw. I can't find the original white paper I was reading, but this is a good one on the subject too [1]. Basically it seems like all the NPK sensors you can easily find are three-pronged and based on electrical-conductivity. These sensors don't have a super long lifespan, and can be pretty inaccurate (when soil moisture changes for example). There may be some good options out there, but I haven't found it yet. I might be overestimating how inaccuract the electrical-conductivity sensors are too, need to do more research. I also want to log temperature and sun exposure; as well as place a rainfall sensor somewhere. I'm really looking forward to a Grafana dashboard of my garden As for your "black box that makes lettuce," I'd highly recommend you look into hydroponics (specifically NFT). That's what I'm most interested in, I keep looking at this blog post[2] as inspiration. [0]: https://www.tindie.com/products/miceuz/modbus-rs485-soil-moisture-sensor-2/ https://www.tindie.com/products/miceuz/modbus-rs485-soil-moi... [1]: https://sci-hub.se/https://www.sciencedirect.com/science/article/abs/pii/S0168169909002087?via%3Dihub https://sci-hub.se/https://www.sciencedirect.com/science/art... [2]: https://kylegabriel.com/projects/2020/06/automated-hydroponic-system-build.html https://kylegabriel.com/projects/2020/06/automated-hydroponi...
  - loxias 5y ago
    Thanks for all the links and pointers! And yeah, now that I know the term "NFT hydroponics" that is exactly what I was imagining, just didn't know what it was called. "i want plants that are plugged into the matrix. living in a fake reality, the organic component of a larger machine that nourishes me.". As for NPK, I (obviously) don't know what I'm talking about, but I'd really want to do something optical. For one, I think you're right, it strikes me as bad to use electrical conductivity to measure both hydration, and relative concentrations of ions, and two, optical just seems ... cooler... to me. In theory, you can "just" use 3 tuned LEDs and a photodiode... Wow, that last link is... now my standard to strive towards for project writeups.

bush-bby 5y ago

agreed ha, coincidentally I just started on a similar project this weekend. One thing I've found interesting is the accessibility and accuracy found in such small modern sensors. Hydroponics may not be "microcontroller heavy", but can definitely be applied.

loxias 5y ago

Coincidentally, I also, also, started a project to do soil analysis remotely! :D Tho mine is just a few crude proofs of concept and a lot of dreaming. The eventual goal is an inexpensive automated black box box that provides me with fresh lettuce daily, with no human intervention. I'm curious, what route are you taking for sensors? Off the shelf moisture? NPK? DIY? I love the idea of logging everything that's possible to log :>

theossuary 5y ago
That's cool! Honestly right now mine is too :) But I'm getting the parts together now to build the first box, the plan is for it to be outside in the garden so it needs to be pretty weather proof. I went with temperature + moisture sensors [0]. I was seriously considering NPK, but I didn't like most of the sensor options I saw. I can't find the original white paper I was reading, but this is a good one on the subject too [1]. Basically it seems like all the NPK sensors you can easily find are three-pronged and based on electrical-conductivity. These sensors don't have a super long lifespan, and can be pretty inaccurate (when soil moisture changes for example). There may be some good options out there, but I haven't found it yet. I might be overestimating how inaccuract the electrical-conductivity sensors are too, need to do more research. I also want to log temperature and sun exposure; as well as place a rainfall sensor somewhere. I'm really looking forward to a Grafana dashboard of my garden As for your "black box that makes lettuce," I'd highly recommend you look into hydroponics (specifically NFT). That's what I'm most interested in, I keep looking at this blog post[2] as inspiration. [0]: https://www.tindie.com/products/miceuz/modbus-rs485-soil-moisture-sensor-2/ https://www.tindie.com/products/miceuz/modbus-rs485-soil-moi... [1]: https://sci-hub.se/https://www.sciencedirect.com/science/article/abs/pii/S0168169909002087?via%3Dihub https://sci-hub.se/https://www.sciencedirect.com/science/art... [2]: https://kylegabriel.com/projects/2020/06/automated-hydroponic-system-build.html https://kylegabriel.com/projects/2020/06/automated-hydroponi...
- loxias 5y ago
  Thanks for all the links and pointers! And yeah, now that I know the term "NFT hydroponics" that is exactly what I was imagining, just didn't know what it was called. "i want plants that are plugged into the matrix. living in a fake reality, the organic component of a larger machine that nourishes me.". As for NPK, I (obviously) don't know what I'm talking about, but I'd really want to do something optical. For one, I think you're right, it strikes me as bad to use electrical conductivity to measure both hydration, and relative concentrations of ions, and two, optical just seems ... cooler... to me. In theory, you can "just" use 3 tuned LEDs and a photodiode... Wow, that last link is... now my standard to strive towards for project writeups.