Programming that spans 2 days

[Andrew]

This is a new installation so I’m not sure if this is just a gui issue or will affect functionality. My standard programming is that I start my run at 10pm and the program will run well past midnight.When I look at preview program it ends at midnight and nothing shows up the next day. Is there a way to accomplish this in a single program?

Attachments:

1. 

   prgpreview.jpg
   

2. 

   prgpreview2.jpg
   

[Ray]

The preview currently does not show time span over midnight, but the controller will run programs over night. More technically, as long as the program has started, it will finish and will not cut off at midnight.

[Andrew]

“sh0uld be able to” 🙂

Can I open a gui defect against this? If so, where?

Also, will logs show that all of the programmed zones ran?

[Ray]

OK, to avoid confusion, I changed my wording to: but the controller will run programs over night. As I said: as long as the program has started, it will finish and will not cut off at midnight.

The preview can be easily fixed to show an overnight-run program on the day that the program starts. For example, if the program starts at 11pm on Oct 24 and runs over to 2am on Oct 25, the preview can be fixed to show this for Oct 24’s preview. But if you check Oct 25’s preview, it won’t be able to show there is an overnight-run program from Oct 24. This is because the preview is calculated on per-day basis, and it doesn’t consider programs whose start times are on the previous day.

[Andrew]

I was reading in the forum that you posted. This thread
 “Because the program’s end time has to be prior to start time, what you need to do is to split the program to two the first program will start somewhere before midnight (say 9pm), and end at 11:59pm. The second program will start at 00:00am (or perhaps later if you don’t want it to start right at 00:00), and end at, say 6am. Both programs will be set to water the selected station(s) for 15 minutes, and repeat every hour. This should achieve what you need.”
 Is the above no longer accurate? Since it conflicts with what was written in this current post.

[Andrew]

Sorry about the garbage in the prev post. Cut and paste was the culprit.

[Ray]

Sure, you can split the program into 2 (one finishes before midnight, the other starts after midnight).

Just so you know, the post you were reading was referring to an older version of firmware (prior to 2.1.0) — in firmware 2.1.0, there is no ‘end time’ anymore. There are now only the first start time, and two choices of additional start times.

[Andrew]

Ah, ok. That’s why you previously HAD to split it because you had to program and END time. Now that’s not the case and you can run it as one contiguous program.

[Ray]

That’s correct. The ‘end time’ has caused much confusion in the past and it’s not intuitive. So in the new program settings, it has been removed.

[ottorino]

Hi all,
I think that this is the right place to report a behavior.

One of my programs starts 2h after sunset (23.01 local time) and then it have a repetition after 2 hours for two times
Only the first event is run, the one at 23.01.
I can tell it for sure because I was awake at 01.01 and nothing happened.

Am I forced to split the program in two ?

Moreover, when I started the program manually, the main screen of app was reporting 2h and 49 minutes to be run rather then 30 sec scheduled.
Also two other channels were turned yellow and some strange figures appeared. I
Then I stopped everything and went to bed !!

app 1.4.5, firmware 2.1.5(1), HW 2.1(AC)

[Ray]

As the user manual describes: ‘additional start times that run over to the next day will be ignored’. What this means is that if you have a program that starts at 23:00pm, and repeat every 2 hours twice, then only the start time at 23:00pm will run. The additional two start times (which would have occurred at 1am the next day and 3am the next day) will both be ignored. The reason these are ignored has to do with how the scheduler works — it will only check programs whose first start time falls on today. Ideally it should also check programs whose first start time falls on yesterday, but this is more complex to check and hence it’s not implemented yet.

“Also two other channels were turned yellow and some strange figures appeared.” — can you be more specific? Do you have a screenshot?

[ottorino]

“Also two other channels were turned yellow and some strange figures appeared.”
can you be more specific? Do you have a screenshot?

Thanks for the reply, Ray.
No unfortunately I did not take any screenshot.

I have the screenshot of the log:

The program was started manually:
The channel “Prato,sole” started and stopped regularly after a minute, the scheduled time (+ 8 sec from evapotranspiration adjustment)
then the channel “Prato,ombra” started and was expected to run for 30 sec + extra time from evapotranspiration adjustment

The sprinklers were watering, and the dot for “Prato,sole” on the main screen turned from red to green.
I cannot remember if the yellow dots appeared at the very same time to the green one, but “Prato,ombra” was yellow after the green/red change on “Prato,sole”.
Within the “Prato,ombra” frame there was a yellow dot on the left, the sprinklers were watering and a sentence appeared under the name of the channel.
The sentence was something about time and I remember it was about 2:49 hours.
At the same time, on the main screen two channels (“Vasi” and “Fasce laterali”) were in the very same conditions of “Prato,ombra” (yellow dot and all the rest), but the sprinklers were not watering and the times were about 16h:something and 29h:something.
I stopped the whole thing by “Stop all stations”.

Attachments:

1. 

   Schermata-1.png
   

[Ray]

Are you sure it was 2:49 hours? If it shows 2:49 that actually means 2 minutes and 49 seconds, because OpenSprinkler allows watering down to 1 second. If it’s 2 hours it would appear as 2:49:00.

[ottorino]

No, I’m not sure. And I think that you’re right in pointing that actually it was 2 min and 49 sec.

Nevertheless, the scheduled time was 30 sec. Also allowing a plus for the Zimmermann method I would expect a figure round 33 seconds
(30(sec) * 1.08 (%) = 33 sec)

OT
It seems that seconds are considered on base 10, instead than on base 60.
I see that the other channel (1 min scheduled time) with a correction of 108% results into a 1 min + 8 sec

Is that right ?

[Ray]

All timing values are converted to units of seconds before they are multiplied with the watering level, so there shouldn’t be any issue with base 10 or base 60.

[Author]

Posts