Forum Replies Created
-
Posts
-
Since I’m thinking about re-activating my opensprinkler, I thought I’d chime in.
I’ve been comparing my own weather data to OWM and Climacell for quite a while now and must say that both are equally bad. Big jumps in temperature (and especially humidity), mostly. Not comparable in quality to DarkSky by a mile (yes, I’m also sad that will go away).Rain is difficult to say as it’s usually thunderstorms around here, and those tend to be rather localized.
-Mike
@Samer: thanks, I didn’t know that. Is that in the manual for 2.1.7? If so, I must have missed it.
Quick question on the side: I used to be able to set the PWS of the station nearby to use (not all of them are equally good). Has that vanished in 2.1.7?
I can now only set the Location, but not the station anymore. At least I don’t find that option.In case those “20%”-stations should run the same duration, you could group the corresponding valves in hardware. Combine five 20%-stations into one.
Ray,
ah, I see. If I’m correct, that is a logic “and”?
if (wl < 20 && water_time < 10)
That would explain it, as when these waterings occur, they are usually below 20%, but certainly longer than 10 sec.
I’ll change it to “||” or remove the time limit, recompile and see what happens 🙂– Michael
Please note: the problem is solved. It was definitively not an user error, but the cause was with my system. I re-compiled the latest version of OsPi, installed it and now it seems to work as expected. At least the additional run times show up in the preview (it’s raining atm so I don’t want to actually water).
Ray,
indeed, that is very clear and exactly how I understood it. Still, I have selected “fixed”, entered the additional start times, checked the checkbox and saved the run, but the additional times still don’t show up on the preview schedule and are not executed, either.
I think I used that last year (before I updated the firmware) and I think I remember that it worked then. Does it work on your system with the 2.1.6(1) firmware or should I try downgrading? It does seem like a bug to me.
Mike.
Ray,
no, I didn’t because I want it to run three times during the day (10 am, 3 pm, 8 pm), but not at night. With “repeat” I can only set fixed times between the runs, but not the schedule I want. I could do it just to check it works, though.
Besides (I don’t have access to my sprinkler right now to check) I think you can only choose one option, repeat OR additional start times, right?
YT,
Michael@Ray:
I doubt that it’s a permission problem as the raspi only has one user (root) which is running the Opensprinkler via a startup script in /etc/init.d.I see it’s the wtopts.txt file that contains the values. If I change them manually with a text editor, they are used, so I can work around doing that. But it still doesn’t work via the web server.
Hi Samer,
just noticed that, but you were quicker 😉
I updated to Unified Firmware 2.1.5 minor revision 2 and will watch it further.Thanks!
Since a week or so I’m having the same difficulties. The water percentage is stuck at a certain value and no last successful weather call is shown on the weather diagnostics.
Restarting the OpenSprinkler service (I run OS on a Raspi) immediately fixes it and updates the water percentage, but it gets stuck again after a few days.
Sure, I could restart the service once a day via cron, but that is only a dirty workaround…
Cheers,
MichaelHi all,
I noticed a small bug with sprinklers_pi: when running a quick schedule, the first station in that schedule (whichever that may be) runs around 30 seconds shorter than it should be. The others run their assigned length.
I don’t know if sprinklers_pi is still actively maintained (would be a shame if it wasn’t) but I still thought I’d mention it.
I second this idea. The baseline values and factors probably work well only for certain climates, and won’t be correct for most users. It’d be great to have the script integrated into the local OsPI firmware (I actually was surprised not to find it there). I mean, even an early Raspberry Pi has the power to run a short python script before calculating the water cycle. For me, that would be enough as I then could alter the values in the python script itself. I wouldn’t mind doing that in imperial units.
Edit: I think landscaping experts agree that you shouldn’t water your garden a bit every day as the water then has no chance to reach the roots. It’s better to water only once or twice a week, but then thoroughly. In summer I usually water twice a week for one hour per zone. If this scheme is used, then the weather algorithm has to consider the weather of the past week and average it out to produce usable results.
Sincerely,
MichaelI’d like to say a really big “Thank You” to Richard for writing this (Where is the “donate a beer”-button?). I’m still waiting for my Ospi-Board, but the dry run seems to work fine. I had to change the values in Weather.cpp, though. Naturally the starting values (=100%) of 70°F and 30% humidity won’t apply for most places on earth, but you have to start somewhere. Using those, the sprinkler times went down to 17%, and it hasn’t rained in several days. But that is easily changed and I’ll experiment a bit with that.
