April 22, 2015 at 10:21 am #37003
Has anyone else been able to get a station to properly turn on using sunrise/sunset timing on an OSPi with the unified firmware? It works if I use a specific time, but when I try “sunrise -30m” for example the station doesn’t turn on (and the app doesn’t say its running either) even though the preview schedule page says it should.April 22, 2015 at 10:32 am #37007
The first thing is to check whether the controller is getting the correct sunrise/sunset times. To do so, click on the weather icon at the homepage, and the sunrise/sunset times displayed for today is the one that the controller returns. If that does not look right, that means the controller is not able to query the sunrise/sunset times correctly.April 22, 2015 at 10:35 am #37008
Just checked, the sunrise and sunset times are accurate. The preview page shows the appropriate timing, but the station doesn’t actually turn on.April 22, 2015 at 6:30 pm #37020
Ok I figured this one out. Station blocks show up 5 mins early on the Preview Programs screen. That is, the display block for a program will show up 5 mins earlier on the timeline than it should. I created a 5 min long run of one of my stations to confirm this and watched the time bar on the Preview Program screen. The station did in fact turn on at the very end of the 5 min time block.April 22, 2015 at 6:48 pm #37021
The program preview works for every day of the week and not just the current day. As a result, it has to calculate it’s own sunrise and sunset time for each day it is rendering. Therefore, the times might vary for programs using sunrise or sunset due to the source of the data.
Basically, the controller get’s the sunrise/sunset data every day from the OpenSprinkler weather service whereas the program preview on the app is calculating the sunrise and sunset within the app without requesting this data.
I do try to minimize this difference by comparing the current controller’s sunrise compared to the calculated sunrise and use that difference as an offset. This brings the predicted times fairly close and 5 minutes is a bit surprising for the current day. Therefore, I will take another look at this to make sure it is accurate however I hope this explains the discrepancy.
Just to make sure, when using a specific start time, this 5 minute delay does not occur?April 22, 2015 at 7:38 pm #37026
That makes sense, thanks for the explanation.
You are correct in that this discrepancy does not occur when using specific start times.
- You must be logged in to reply to this topic.