RainMachine Sprinkler Irrigation Controller Driver

[gReatAutomation]

For those with a DarkSky API account who want to add it as a data source:

https://support.rainmachine.com/hc/en-us...orecast-io-

[kfly]

  In the condo next door, in rolls 4 guys from Boston, these guys had the full Boston accent.  What was so funny about it was this was when the movie Urban Cowboy was all the rage, and these Boston guys show up in Hats, boots, jeans, and western belt buckle.  They were shocked that none of us Texans had boots or hats, nor did we own them.  I wished I had taken a picture of them.

I hear you. I live in NY/NJ and I have trouble sometimes with the thick boston accent. Hope you have a fast recovery.
Kevin

[kblagron]

I updated the RainMachine Driver to V1.96 in the first post as follows:

• Changed the way the Start Time for the Snooze operation was obtained so that it now reads the current time on the RainMachine, vs the Server Time, which could vary and cause the Snooze operation to take effect at a later time if not synced properly.
  
• Added the RainLastStartTime Field, which will show the time that the last rain event started if a Rain Sensor Installed.  Because the RainMachine JSON does not report that date/time until after the rain sensor is no longer detecting rain, the driver will automatically record the date/time that the rain sensor becomes active, and store that date/time. When the rain sensor turns off, and the RainMachine JSON is updated, the date/time stored there will now be used by the driver.  Because it is captured differently depending on whether the sensor is active or not, it could be 1-4 minutes off, since the rain sensor is only queried every 5 minutes. 
  

[kfly]

OK just turned on my HD touch Rain Machine. Hard to believe it was snowing 2 weeks ago.
Seems there is an issue as driver won't come online.

Seems my RM Unit returns a float  in the JSON item "scheduledWateringTime" for a few of my zones.

Although my HD is at firmware 4.0.1129. and my  PRO(dev test unit) is at 4.0.1148 I think it has something to do with the automatic "Weather-Adaptive Watering"  calculation.  Or something to do with just turning it on and having 1 less day of weather.

Question:
In your watering program do you use "Weather-Adaptive Watering" or "Fixed".
I use "Weather-Adaptive Watering".
Will do more investigaing later today after I fix the Bellagio fountain in my front yard created by the landscapers...

-----------------------------------------------cut----------------------------

Line where it bombs out:
UserDuration := m_JSON.FindCardValue(DailyStatsZoneListAnchor,"scheduledWateringTime",False,0);


      "programs": [
        {
          "id": 1,
          "zones": [
            {
              "id": 2,
              "scheduledWateringTime": 1808,
              "computedWateringTime": 1216,
              "availableWater": 0,
              "coefficient": 0.67,
              "percentage": 67.27,
              "wateringFlag": 0
            },
            {
              "id": 3,
              "scheduledWateringTime": 1762.8,
              "computedWateringTime": 1186,
              "availableWater": 0,
              "coefficient": 0.67,
              "percentage": 67.27,
              "wateringFlag": 0

[kblagron]

That's an easy one,  RM returned a float value when the driver was expecting a card.  I will get an update out that checks for this and adjusts back to a card.

[kfly]

ok and one other in the 30 day history method. Then I think we are good.

UserDuration returns the same type of float.      "userDuration": 1762.8,

Here is the problematic line around line 1154:
UserDuration := m_JSON.FindCardValue(WateringCycleListAnchor,"userDuration",False,0);


                "uid": 3,
                "flag": 0,
                "cycles": [
                  {
                    "id": 1,
                    "startTime": "2020-05-17 04:56:19",
                    "startTimestamp": 1589705779,
                    "userDuration": 1762.8,
                    "machineDuration": 1186,
                    "realDuration": 1186,
                    "flowclicks": 0
                  }
                ]
              },

[kfly]

One last small thing to fix.

at line 757 you  do:
                WriteBoolFldByName(FldActive, ProgActive);

but you never set ProgActive..

Need one of these around line 738:
                ProgActive := m_JSON.FindBoolValue(ProgramListAnchor,"active",False,False);

Thanks again.
Kevin

[kblagron]

I updated the first post with an update that reads the numbers that were being sent to the driver as a float and converts it to a card so those errors won't  appear.  I also updated the ProgActive variable as you suggested. 

Let me know if this fixes the problems - Its working fine on my Pro-16.

Blake

[kfly]

I updated the first post with an update that reads the numbers that were being sent to the driver as a float and converts it to a card so those errors won't  appear.  I also updated the ProgActive variable as you suggested. 

Let me know if this fixes the problems - Its working fine on my Pro-16.

Blake

Thanks!  Both issues resolved and working great on my TouchHD
Kevin

[kblagron]

Whenever I see that the RainMachine has gone offline, which seems to be everynight around 5-6 am here, I was thinking that the driver should bypass this event if it is for a short period of time.  I am not sure why it does it, but maybe it's recalculating RainMachine data to determine watering amounts.  It seems to always be an HTTP 503 error on the daily stats that are polled, but have seen it at times on other items.

What I was thinking about doing was not throwing a fatal error if the data was only missing for the last 30 minutes or so.  This would keep the driver from restarting, and most of the data that was gathered previously would be suitable for display without having to send the driver offline and restart.

Any thoughts?