Engineer & Dew
Hey Dew, I've been tinkering with a small automated irrigation system that could keep a micro-ecosystem alive with minimal energy. Think we could build one that also tracks plant growth?
That sounds like a lovely idea, the little garden will thrive with a gentle touch. A sensor array to watch leaf lift and moisture level could be neat, and a tiny log on a micro‑controller can keep track of growth without draining the battery. Just keep it quiet and low‑profile so the plants feel at home, and let me know if you need help wiring or choosing the right data‑logging software.
Thanks, that’s a solid plan. I’m leaning toward an ESP32‑C3 for its low‑power mode, a capacitive moisture probe, and a small Hall sensor for leaf lift. I can log data locally on an SD card or stream it over WiFi if the network is stable. Let me know which logging setup you prefer and any wiring tips you have.
I’d lean toward local logging on the SD card—no WiFi power draw and it keeps the system quiet, just like the plants. If the network is reliable and you can afford a little extra energy, streaming is fine, but keep the WiFi in deep‑sleep mode and wake only for a few seconds each hour.
For wiring, keep the capacitive probe at the low‑voltage side of the ESP32 and use a simple RC filter to smooth out the readings. The Hall sensor should go through a pull‑down resistor so the pin reads cleanly when nothing moves. Keep everything shielded from stray interference; a little metal plate over the probe can help. And don’t forget a small capacitor across the ESP32’s 3.3 V rail to snuff out any hiccups during deep sleep. Happy building!
Sounds good. I’ll get the ESP32‑C3, a 10‑µF electrolytic for the 3.3 V rail, and a 10 kΩ pull‑down for the Hall sensor. I’ll mount the probe on a 0.5 mm steel plate for shielding and add a 1 µF ceramic across the probe’s leads for extra noise rejection. I’ll run a quick test to confirm the RC filter holds up at 1 Hz sampling. Will ping you once the prototype is wired up.
That sounds like a solid setup, just make sure the probe stays dry and the 10 µF cap is right next to the 3.3 V pin so it really smooths the voltage. The 1 µF ceramic on the probe will give you a nice clean signal, and at 1 Hz the RC filter should do fine. Keep an eye on the humidity drift, the plants love consistency. Let me know how the test goes, I’ll be curious to see how the data looks. Good luck!
Test run done. The ESP32 wakes every hour, reads moisture and leaf lift, writes to SD card. Voltage stayed within ±10 mV with the 10 µF cap. The probe readings are stable; the 1 µF ceramic kept noise under 0.2 V ripple. Next step is to calibrate the moisture curve for the soil type and plot the leaf lift data to see if it correlates with watering intervals. Will upload a sample log in a few hours.
That’s wonderful news! The stable voltage and low ripple mean the plants will get a steady signal—great job. Calibration will be a bit of a dance; compare the probe readings to a known moisture range and maybe use a simple linear fit for the soil type. When you plot the leaf lift, watch for any lag between water added and the lift spike—it can give you clues about root pressure and soil tension. I’ll be excited to see the log once you upload it. Good luck, and thanks for sharing the progress!