I’m confused about what drives the sprinklers. When you set up a zone it asks questions about vegetation type, soil, sun etc. Then you are asked to set up watering schedule. So if you set up a watering schedule what do all the factors about vegetation types, soil, sun etc. actually do if anything? Does the water timing drive it, or these other factors somehow. I can’t quite work out the relationships and what actually is significant in determining what happens. I realize this may be a really stupid question but it seems the two things may be conflict? Is see SteinyD asked similar question in July of 2014, but don’t see an answer.
If you configure a zone to run every day at 5:00 am (these are really allowed days) then:
the sun and vegetation is used to determine transpiration and evaporation rates of the water (also using local weather data) to determine a depletion rate and when to water next.
The soil type coupled with the grade helps the controller know how long to water/rest the zone in cycles per schedule to maximize absorption and reduce runoff. This feature works very well in my opinion. After 7 years on a pro c, I finally got it dialed in perfectly, out of the box and a few questions later, the suggested zone run time was 1 minute different and duration was 3 minutes different. I was blown away, and appreciated the 3 extra minutes.
Something I’m not sure, but I hope it works this way, vegitation should help the controller understand how much seasonal adjustment is required.
So you set the zone for 1/4 inch, and the controller runs the zone, then waits enough time for the water to deplete from the soil (days) the runs the next time the schedule is allowed.
For me, I found the device way too aggressive at depletion rates so I have worked around it for now by only allowing a zone to run 1 time a week. This is supposed to be improved in 2.0 so I’m very eager to reattempt another autopilot landing.
@plainsane Thanks a lot for your explanation. As someone who is ‘close to the code’, it’s refreshing to get another perspective on things. Your explanation is probably better than mine.
@garmanmd I’ll try to make this as simple as I can with a high-level matrix 
As you can see, there is no real simple answer. My best analogy is garbage data-in, garbage data-out. So the better our data is about your zones, the better our algorithms work. Some day I will get to posting all of the equations. I’d like to be as transparent as possible.
As @plainsane alluded to, our 2.0 app coming out in May will include flex schedules which will only apply water when we determine your zone is depleted of moisture. So, the watering frequency will dynamically change.
for me, the real money shot is when soil sensor integration matures and either i or a company build a sensor out of these chips http://www.wired.co.uk/news/archive/2015-01/07/intel-curie-quark-chip
i see the potential for wind blowing this around on a string to help recharge it, would be so cool.
then the controller can just run each zone when the soil moisture sensor reports low moisture.
that is my most ideal solution.
@plainsane Agreed. We are going to try to integrate with soil moisture sensors this year, but it doesn’t feel like there is a perfect solution yet.
not within a home owner’s price range.
honestly, i feel that you will end up building your own…but time will tell.
the button chip i posted seems to be a nice balance of power usage and processing capability but who knows about the price.
for now, when the api is available ill most llikely fall back to http://www.amazon.com/Aeon-Labs-Aeotec-Z-Wave-DSB45-ZWUS/dp/B00H3TJ3P4/ref=sr_1_1?ie=UTF8&qid=1430491387&sr=8-1&keywords=aeon+labs+water+sensor+dsb45-zwus
this way i can use my zway/raspi install to call into your api to indicate total depletion. i think this will be sufficient enough as i have tif 419 and it can handle a day with a totally dry root canopy.
plus, the zwave protocol i easier/cheaper to extend in my opinion than a wifi signal.
I may be dense, but I’m still confused. Let me see if I can state my question another way. I can easily understand two alternatives:
- automatic - the system uses all these factors and an algorithim to determine when and how much to water
- manual - the schedule you set for days and times determines much like a traditional timer
What I don’t understand is if this controller somehow mixes the two? I can’t see that there is a choice to clearly select automatic of manual so I’m assuming there might be some mix. If so what is the role of the manual settings, and does the algorithims override manual.
To ask a more specific question if I set the schedule to water on Tuesdays does this system look at all these factors and decide it really should be Wednesday or Monday. Or if the schedule is for 30 minutes does this system use the factors and decide it should really be 45. I understand the system splitting up the water time so it has time to soak in, and that all makes sense. But what I can’t figure out is what drives when and the total time the system is asked to put water to a zone. From an earlier question is seems that the historical data from the system is elapsed time, so if that is true then it seems impossible to figure out how much water it is actually putting out. Perhaps this is all meant to be automatic, but it seems there should be some understanding of what the settings actually do (primarily zone and schedule settings)
@garmanmd Good questions.
We currently allow you to create a watering time that can be seen as fixed, or running at the interval(s) you choose.
There are options like smart cycle (http://support.rachio.com/article/283-smart-cycle-overview), weather intelligence - or I like to call a virtual rain sensor - (http://support.rachio.com/article/284-weather-intelligence-watersmart), and water budgeting (http://support.rachio.com/article/282-water-budgeting-and-evapotranspiration-watersmart) that will make adjustments to your fixed schedule automatically (by either adjusting +/- zone durations, skipping schedules, and breaking up a zone run into smaller cycle/soak runs).
Water budgeting (week-to-week) adjustments are closest to what you are referring to for ‘schedule adjustment’ automation. Everything is transparent and shown in the app as to what your durations are so there are no surprises.
Flex schedules which we are releasing in our 2.0 app in May take automation much further. You will tell us what days we can water on, and then we will take over from there adjusting the irrigation frequency based on current weather. They will dynamically select watering days based on soil moisture needs. This promotes root growth, reduces watering frequency and results in a healthier, more drought tolerant landscape.
Hope that helps explains where we are, and where we are going.
1 Like
franz will need to clarrify some specifics.
to get manual behavior, you should disable the smart technologies (weather data, smart budgeting etc.). this will water exactly as your schedules are configured. throw on a rain sensor and you will skip a schedule if the rain sensor is in a triggered state.
to get automatic behaviour, enabled the smart technologies (weather data, smart budgeting, etc.) and that is the behaviour i will discuss.
im also going to speak with authority here, but franz, please correct me where im wrong, im making some strong assumptions here. ill also defer to franz on 2.0, there is verbage on this forum that i understand, but have not been able to make my own observations yet, so i will most likely be speaking out of my anus.
the most important thing i would draw attention to is this:
the only thing that square footage is used for is to construct the gallons used graph that you can access via the cloud or phone app. nothing else, i have not found any uga or tif compnay research concerning turf management that use footage to determine how to water. it is assumed that the irrigation system follows good head to head coverage principles because the only thing that really matters is the flow rate of the head itself. your gallons used graph will be as accurate as your square footage and head type are configured, but still an estimate. i have asked them to support a flow meter, that would be the tits.
so let me answer your specific question and lets assume you set the schedule for a zone to water on tuesday and weds at 1/4 inch with all the smart technologies enabled.
as the software stands right now the schedule would run on tuesday for sure, AND then wed BUT ONLY if the vegitation/sun/soil type/head type and their intergalactic algorithm, indicates that the water that has been applied on tuesday has been been depleted. BUT! if a weather station reports that 1/4 inch of accumulation fell out of the sky on monday, the controller would most likely skip tuesday and water on wed. franz can speak to the tolerances of their algorithm but i cant.
my main complaint right now (supposed to be fixed in 2.0) is that if on monday we get 1/8 accumulation of rain, the controller is going to fire off and deliver 1/4 inch of water on tuesday (because the zone didnt reach its peak of 1/4). what i really want is 1 of 2 things:
either the controller to wait for the 1/8 to be depleted (reaching a dry root canopy, fungus is a problem here in ga when you keep your turf at .16 inch mowing height), then drop 1/4 inch on my yard 1 day after total depletion but on the next scheduled day.
OR the controller delivers 1/8th plus the estimated ET (evapotranspiration) on tuesday which equals the configured 1/4 inch for the zone the following schedule (because 1/8th fell naturally) then wait again for the zone to be totally depleted for an entire day before running the next schedule.
for now i get 1/4 plus the ET rate from the controller and this causes my ground to become soggy and over watered (because in this month, we get rain for an entire week at a time). this is the “aggressive” behavior i was referring to in my previous post.
im not sure exactly how their algorithm works, that is their intellectual property. im making some big assumptions based on my own research/measurements from the last 7 years, which helped me dial in my hunter pro c.
dont feel stupid, this is reasonably difficult stuff. if it was easy, rachio could not provide value to the irrigation market thus they wouldnt exist.
2 Likes
Thanks for the reply. I don’t want to know their proprietary formulas, just trying to understand the general logic so I know what I need to to set and what the system does in the background. Maybe 2.0 will be clearer, not sure.
I think if I summarize your response regardless of what I set in the schedule, the zone factors, weather and their formulas may override the days it waters on as well as the amount. It sounds like in 2.0 they will at least add the ability to tell the system don’t ever water on Tuesday.
I suppose I really didn’t understand what I was buying, and the whole topic is fascinating. I live in AZ where we get very little rain, and the thing that most affects water usage is how hot and dry it is rather than how much rainfall we get. It sounds like their formulas would adjust how frequently the watering happened based on the weather (hotter and drier would be more frequent than cooler and/or more humid). Does that sound like a reasonable conclusion?
@garmanmd It’s great to have people interested in this topic!
Currently this is correct. We might skip a schedule due to rain (if you have any 
). We will also adjust weekly (if you have opted in for water budgeting) your duration based on current weather.
2.0 you will select which days you can water on, and we will only water when the zone needs it based on soil moisture levels.
Exactly.
Thanks, some of the fog is lifting. I’m still trying to understand what the schedule does so one more question. Does it essentially start with the schedule as the “average” and then adjust? And is that adjustment relative to what is in the schedule.
Again a specific example. Assuming the weather is the same in the following two cases will the watering time be the same or different? Case 1 schedule time set to 15 min and Case 2 schedule time is set to 30 min.
Yes, each schedule will be created with the “average” (it’s the actual average, not historical) for the last few weeks. From there, you can adjust manually or water budgeting will adjust the schedule weekly.
Flex schedules work different in that the run time for each zone will be the same throughout the year, it’s the watering frequency that will change.
Flex schedules will be replacing ‘water budgeting’. I’m expecting to retire automated water budgeting on fixed schedules in the next few months.
Here is some good content on the science behind flex schedules. As you can see after looking through these articles, no one would ever be expected to do this on their own. That’s the great thing about using technology for this problem.
http://www.ext.colostate.edu/pubs/crops/04707.html
This simulator will show you how often flex schedules will potentially run and for how long: