Software Bug on "Fill" and Zone Over 100%?


#1

I may have found a bug in the software or at least it did not act as I would have thought.

Here is the scenario

All zones on daily flexible schedules (which I love by the way)

April 29

  • 2.5 inches of rain
  • My PWS went down so was not recording rain amounts
  • Once I had my PWS up and running went into my web app to “Fill” the zones to ensure the missed rainfall was captured (at least that was my thought process).
  • The first zone I went too had a forecasted 107%.
  • I manually filled the zone and the forecasted amount went “down” to 100%. Hmmm. decided to not do any others.

Here is where the problem began.

  • On April 29 (the date that I adjusted the zone to full) the software entered a -3.09" of irrigation.
  • For the remainder of the day only recorded 0.8" of rain.
  • Result was at the end of the day the zone was shown as empty, 0%
    April 30 watered 1" in just that zone.

Seems that when the zone was above 100% and I adjusted to “Fill”, something in the software forced a -3.09" into the irrigation.

Will try to upload screen shot below.

User error - probably, should not have tried to fill a zone already over 100%.
Software error - I think so, should not have entered a large negative rainfall?

Thanks


#2

Hey @tomlay-

Thank you for bringing this to our attention! One thing you will notice under the Irrigation senction of more detail is that the negative precip that was recorded was from user edit (fill/empty). It looks to me like for some reason it actually emptied instead of filling. Do you mind telling me what browser you were using when this occurred? I am going to try and replicate it on my end!

PS- love to hear flex has been treating you well :blush:

McKynzee :rachio:


#3

Google Chrome.
Yea I thought the same thing on empty vs fill. However right after I did the “fill” the new forecasted amount was 100%. I “think” if I had hit empty by mistake would have gone to 0% instead of 100%? In addtion showed “forecasted” rain amount equal but oppisite the adjust amount i.e. +3". So the forecasted amount was 100%. But then when the rain was only 0.8" vs +3 (forecasted) took the zone to 0% at midnight and thus the water cycle.

Hope the extra detail helps.

Tom


#4

I’m pretty sure this is not a bug, but working as designed. When you tell Rachio that you want to mark the zone as “Filled”, it has to get the zone to 100%. It does that by adjusting the irrigation amount of water. If you had been less than 100%, it would have added inches of water to the irrigation amount, even though you didn’t actually put water down. In your case, because you were over 100% (or already showing saturated), it had to put in negative inches to get you down to 100%.

@mckynzee, check with the team, but I’m pretty certain this is not a bug.


#5

HI Linn, What happened was that it for some reason when I went to fill it showed forecasted +3" of rain and then adjust to -3" of rain. So when midnight hit and only had 0.8" of rain, that new zone moisture content was 0%. Not sure that would be the correct behavior. May have something to do with the forecasted rain amount be so very high as well.

Agree should not have tried to “fill” a saturated zone. Was “hoping” that since I had had so much rain that when I hit “Fill” it would go back to 110%. But that is not what happened.


#6

Here’s my guess of what happened and why:

The day that you did the FILL, April 29 was still a “forecast” – things remain “as this may happen” until the end of the day. So at the time you clicked on FILL, there was a forecast of 2.5" of rain. Because you were already over 100%, AND the system thought it was going to get 2.5 more inches of rain, in order to compensate, it had to show a -3.08 inches of watering to bring you DOWN to the 100% level. That number ACTUALLY gets entered in, just as if you had manual watered, that number would get entered in, but everything else (that pesky precipitation) remained a forecast. Then at the end of the day, the system recorded everything that actually happened. And you got 0.8" of rain instead of the 2.5". So now, based on “reality” figures, you ended up at 0% moisture balance.

Hopefully it makes sense when you think about everything for the day being a forecast, except for the watering that actually occurred. (and in this case it was a negative watering).

(Don’t worry – you’re not the first one to do this – I did it myself a week or so ago – saw my ground was saturated, wanted to keep it from running, I was at 112%, did the fill and it went to 100% – now it thought it was LESS saturated!! So I learned my lesson!!!)


#7

Yep agree that is what probably happened. Just don’t think the software should work that way. If forecasted and actual end up being different after a manual adjustment seems to me when the calc. was done at midnight would reset everything to 100%. Maybe there is a use case where doing this creates another problem. Don’t know. Will let @mckynzee and company figure it out.

Thanks for your help.


#8

I think they used to do it this way, and it confused people even more. You would set to FILL or EMPTY and it would look like nothing happened. It didn’t show until the next day. So I was pleased when I saw it doing the FILL immediately. Until I realized that it really did happen right then. So it may be a catch 22 – or maybe a little more tweaking. And for all we know, it may be changing (or not) with V3. In the meantime, I think that just understanding how it’s working now helps alot.


#9

Hmm interesting. Seem like the logic could be

When Fill set irrigation to Forecasted moisture level -100% ( I think this is what it does currently)

At midnight
IF manual adjust level <> 0 then set manual adjust = moisture level - 100% (would need to be a different manual adjust register so doesn’t loop on this every night)

Or something like that.