- This topic has 4 replies, 3 voices, and was last updated 1 week, 3 days ago by
.
-
Posts
-
I built an open-source service that turns a backyard weather station plus forecast data into OpenSprinkler runtimes, computed with FAO-56 Penman-Monteith ET, per-zone soil water buckets (species and soil-texture aware), and a transparent skip ladder (rain, wind, freeze, soil moisture probes, local restrictions). It talks to OpenSprinkler directly over HTTP on your LAN: scans stations during setup, schedules a sunrise-anchored morning run, and shows the complete math behind every zone’s duration so you can audit and tune it.
One Docker container, no cloud. Live demo: https://demo.localsky.io / OpenSprinkler docs: https://localsky.io/docs/controllers / Source: https://github.com/silenthooligan/localsky
Beta and OpenSprinkler users are exactly who I want feedback from.
Welcome to the OpenSprinkler forum. I just took a quick look at your links. Fantastic work! I am very impressed. I will give is a proper try asap. Again, thank you for sharing your work.
Thank you, Ray, that means a lot coming from you.
OpenSprinkler is what made this possible. I wanted irrigation I actually owned instead of something locked to a cloud, and your open hardware and local API let me build the weather brain and dashboard I always wished sat in front of it. If you end up giving it a try, I would love to hear any feedback to improve the engine/integration with OpenSprinkler or any other quirk or recommendation(s) you would have.
Thanks for the warm welcome.
This looks great. The fact that it uses local weather data and keeps the whole setup out of the cloud is a big plus. I also like that the calculations are visible instead of being hidden behind an automatic schedule.
The per-zone soil water buckets sound especially interesting. How much setup is involved to get those tuned for an existing yard with different soil and plant types?
Would love for you to test it out. Happy to hear feedback of any kind.
Setup on the LocalSky side is about five minutes if your OpenSprinkler is already up: the wizard scans the controller and finds your stations, and each zone is a short form of dropdowns (species, soil texture, sun, slope, area, sprinkler head type). It runs same-day on catalog defaults.
From there you can be as hands-off as you want. The app counts the rain that fell and the watering already done against each zone’s weekly target, and after a week or two of runs it hands you a tuning suggestion with a one-click apply when something is off. If you would rather tune by hand, everything it can suggest is also just a field in the zone editor, so a known soil type or a catch-cup measurement can go straight in. A soil probe adds two deeper checks: drying rate against the model, and your real sprinkler rate measured from probe readings during runs.
- You must be logged in to reply to this topic.