- This topic has 10 replies, 3 voices, and was last updated 6 months ago by
.
-
Posts
-
Hi Guys,
I have installed docker version of “opensprinkler-weather” / local weather service.
I can send data there from my local weather station. One question I have and haven’t been able to find is how often that should happen for OpenSprinkler to be happy. My weather station sends data like 4 times a minute, but I think OpenSprinkler just checks every couple of hours. So how often should I send new data to the local weather service?
Cheers,
// Jonas
Another thing that is not completely clear is if temp and humidity should be live values or average over some time. The dailyrainin and rainin properties are well defined, but for tempf it just says “Outdoor temperature in fahrenheit”. It is somewhat confusing to have to average out measurements for two properties but not for the rest. Should tempf be the value that comes in from the weather station at that time, or should it be the average value over 24 hours? Same question for humidity
It seems there is not a lot of knowledge in this area and I am still scratching my head. I am now updating every 5 minutes and that is wrong! The issue I am getting is that when I send along rain for 24 hours and 1 hour every 5 minutes the weather-service is somehow adding it up incorrectly. I get 3.84 inches in the Opensprinkler GUI (System Status) when in real life I just had 0.685 inches over the last 24 hours. I will switch to updating once an hour and report back
Hi Jonas, did you already get an answer about your question of the update frequency intervals of weather data? I want to shorten the update frequency because of making the weather adjustment more accurate. But I didn’t find any value. Greetings
Hi Walderama,
No answer. I switched to update every hour and it seems to have made things better. The thing is that I live in Colorado and the way we typically get rain here is like events. We don’t get the trickling rain for days that you can see in other places. It seems to me that if we have multiple rain events the same day it becomes inaccurate, but that’s more of a gut feeling than anything else. I have tried to look at the code but it is solving a lot of stuff for different situations and configurations, which makes it a bit complicated to follow. I am thinking about simply implementing Zimmerman in my own python script that currently is extracting the data from Influx and feeding it to the weather service. Then I can set the watering level manually via APIs.
I ended up getting rid of the weather-service. The weather station sends stuff out on a radio frequency that can be intercepted. I then dump it on an MQTT bus. Then I use some Telegraf service that dumps it into InfluxDB. Then I have a Python script that:
1. Queries InfluxDB for 24h data
2. Calculate Zimmerman
3. Use the API to set the watering levelI run this twice a day. All good, problem solved.
@ Jonas,
What API command do you use to set the watering level? Could not find anyone specific. The closest I found is in section 13. Manually Start a Program [Keyword /mp]
/mp?pw=xxx&pid=xxx&uwt=x , but this will start the program as well.http://<ip/host>/co?pw=<md5hashed password>&wl=<integer>
Jonas,
I used the command but did not work.On my program I have unchecked “Use weather adjustments” and in options I have weather adjustment method as manual.
Tried 150 , to get 50% more time on the watering time, but that did not happened. BTW, did get “result : 1” on response with the http message, so it is correct.
Also, when changing the WL % via the API, should the GUI reflect that as well?
Any idea what might be missing?Read the user manual and it states ” % Watering : global scaling factor applied to station water times. For example, if set to 150%, all station water times will be
multiplied by 150% (except in programs that do not use weather adjustment).” So I checked on the program but still no change . Tried other commands and API is working fine.
- You must be logged in to reply to this topic.