Like the others, I’m a little concerned about the proposed changes.
It took a while to dial it in, yes, but the Flex schedules were awesome. When we had a stretch of rain for 3 days in a row, for example, the schedule would adjust appropriately based on real data. Other than the need NOT to water at certain times (landscapers, school buses, restrictions, etc.), I totally fail to see the need for a fixed regular schedule. Is there really a use case for having a set schedule? If so, maybe you let the user select it. But I know that for me, letting the Rachio figure out the optimum time to water was ideal.
To make a fixed schedule with skips work, it would seem you would have to water much more frequently with shorter duration because a skip could otherwise leave you underwatered. And isn’t part of the point of the Rachio to calculate the best time to do a deep watering so roots grow to the right level (especially in a lawn)?
I’ll withhold judgment until seeing the final product, but I really hope this isn’t a step backward in terms of the advanced functionality that drove me to the Rachio in the first place.
@stonecliff, thanks for your concern. Please note, if you have a Flex schedule at the time of the 2.5 release, it will be grandfathered on your account – however, you won’t be able to create a new Flex schedule in 2.5 so make sure you don’t delete it.
Water As Needed schedules will still do this. The biggest difference is you’ll have extended visibly into what the schedule will do.
Water As Needed schedules use the same interval logic as Flex schedules. Since schedules are skipped due to rain and/or high moisture levels in your soil (greater than ET within the given interval), the skip logic mimics the delay logic of Flex – only the concept of the schedule adjustment changes from a rescheduling (delay) to a skip. Water As Needed schedules can be fine tuned to water more or less by zone, allowing greater control of your watering schedule.
MAD (management allowed depletion) is baked into the interval logic of Water As Needed schedules. Watering durations and intervals should align with the expected watering behavior of a Flex schedule when viewed at the month level.
Water As Needed schedules use more data points than Flex schedules and should save more water if the controller is configured correctly at the zone and weather station level. No functionality is being lost; just simplified. The new schedule logic is a hybrid of Fixed and Flex schedules. Like an ice cream cake – some things are better together
Thanks for the detailed response. I guess I’m still not convinced. If I have a zone that is “typically” watered every seven days, with Flex I could currently end up with an eight day interval because, say, there were a couple of cooler days where not enough water evaporated. But with Water as Needed, that would turn into a 14 day interval, correct? How is that an improvement?
Is it possible the solution to our concerns is in letting the user set the frequency that the intervals adjust? Set is a default to monthly if you really think that is what users want (does anyone really care what the future schedule is?), but let the user change it to daily if they like the way Flex worked (like me)?
@franz@emil: Thanks for all the information you’ve shared. I’ve been thinking and following the conversation for the last week and also believe you’re taking a step back. I understand the core principle of MAD scheduling is to perform a full watering on the lawn when needed (thus the variable duration and fixed application amount). By following this method, the promise is that one will achieve a healthier lawn with deeper roots in the long run.
With your new scheduling, people will gain the predictability of knowing what days watering will happen. However, it will come at the expense of not always having the schedule line up with days that require full watering. Thus you will lose some of the benefits to your lawn that full watering provides.
I also believe that this will lead to wasting water. Whenever you water, a certain amount is needed to coat the grass and thatch before water reaches the ground. That water will likely just evaporate off. Given your proposed scheduling, I would assume that watering will happen more often. Thus it’ll lead to more water waste in the long run.
I’d appreciate your thoughts back–especially if I’m misunderstanding something. The draw of the full MAD watering with variable schedule is why I purchased a ESP-SMTe, and also why I was strongly considering switching to a Rachio in the future. However, with your new version, I’m not sure it’s appealing to me to switch anymore. Thanks again.
Many of my neighbors let their lawns die last summer in SoCal. My lawn was green and we used about 40% less water than the previous years. This was all on Flex schedules that I never really had to babysit. Reading this post concerns me as well. When I first got the Rachio flex wasn’t out yet. I was disappointed in the fixed schedules as flex is how I imagined the system should work. I really hope this change does not remove the magic we experienced last summer with flex scheduling.
Thanks for all the feedback everyone, I love these discussions.
Please note that flex will continue for the rest of the watering season for folks that already have those schedules in place and we will be collecting data and doing analysis on the new simpler water as needed schedules to determine if there are any modifications we need to make for next year. One of our core tenets is continuing to look at improvements in our system in efficiency and app simplification. We believe water as needed in conjunction with our ability to skip intervals based on flex-like concepts using ET, precip, and minimum allowed depletion will be a great addition to our scheduling model. During the season we will be analyzing our data to determine if there are other optimizations that we need to refine or add.
@stonecliff, sorry for the delayed reply…ate too many chocolate bunnies on Sunday.
Your understanding of Flex versus Water as Needed schedules is correct. Flex schedules are delayed based, whereas Water as Needed schedules are skipped based.
Yes, potentially. We were hoping to explore this for the 2.5 release, but had to put it on hold. As Franz mentioned, we want to continually to improve our system’s efficiency and app simplification. User feedback is needed do this. As such, we’d love to get feedback from Flex Schedule users on the new schedules…just make sure you only disable your Flex schedule and don’t delete it if you choose to test out the new schedules.
@bzimmer, Water as Needed schedules use the same equations to calculate your watering duration by zone. We’ll re-crunch this number on a monthly basis, but the duration won’t change much much month to month (a few minutes +/-).
Water as Needed schedules will do their best to take advantage of what Mother Nature throws your lawn. We’ll skip using a number of triggers (rain, ET & temperature), but you’ll never go empty – MAD (management allowed depletion) is running in the background to ensure you have the reserves to skip. We’ll never allow the zone’s moisture balance to fall below the permanent wilting point. It’s really no different than waiting to fill up your half empty (or half full ;)) gas tank until the next exit in hopes of a lower price, or the following day when the weather is nicer. By default, MAD will always default to 50% and the related schedule interval is aligned to water accordingly. If MAD is changed on any zone, the interval will also change as the water requirement will be different.
In the comparisons ran during testing, the schedules balanced out pretty well. i.e. # of actual watering events over a given month on Flex schedules versus the number of watering events on Water as Needed schedules with skips removed.
Since Water as Needed schedules will not delay the schedule, but actually skip, the schedule is applying a managed stress principle to the zones, which will help encourage deeper root zone growth as anytime the moisture balance falls below MAD (i.e. 50%), the roots will search for water. Plants are very smart. If there’s water, they will find it.
@cwagz, quite the opposite – the magic will only get better! 2.5 schedules will be easier & smarter. I hope everyone will give them a test drive for a month or two and let us know what they think.
I look forward to the next generation of the software! Progress is a good thing! Since you’re working on scheduling software, it seems like the place to put my quarter on the table. I may bug support too.
Our lot is somewhat sloped and has dense clay soil. It doesn’t readily absorb water. With our old Rainbird sprinkler, we’d water each zone for 10 to 20 minutes, depending on the zone’s size. After all 6 zones had watered, we’d wait about 30 minutes and water them again. In very hot months (this is Texas, so it gets hot), we’d add a third cycle half an hour after the second one stopped.
It seems that your delay between cycles is more like, sprinkle zone 1 for 9 minutes, wait 9 minutes, sprinkle again.
Whatever the situation, we have water pouring off our yard and into the street. Luckily that’s not a ticketable offense, but it does get the local water militia after you.
A longer delay between cycles for dense soil would probably be a good thing™.
I went out today and measured our zones to get watering times closer to something reasonable. That said, it seemed that the delay between cycles was much shorter than the amount of time mentioned in the article you pointed me to.
Is there a way to download the watering history into a spreadsheet? Honestly, the low contrast on your website in the watering history area makes it hard to read for older eyes. (The Color Contrast Analyzer add-in/extension for Chrome makes it very easy to see what people with mild, or severe, visual impairment - or who are just getting on in years - will see at your site.)
Continuing the discussion from Simpler Scheduling for Spring:
The time honored method of highlight, copy and paste worked. Looking at the history, it looks as though the delays are shorter than you think. I can send you a spreadsheet if you cannot see my history.
However, it seems that it will cycle through other zones only if they are scheduled for that day. So, if only one zone is scheduled, there will some very short delays.
With reports from neighbors of heavy water flow, my wife has been aborting sprinkler runs, so our history may not be the best to look at.
Looking at the calendar the cumulative soak time is 30 minutes. Even though the cycle is 11 minutes you need to take the other zone watering times into account which will always equal 30 minutes delay from when the zone runs again.
Thanks for the clarification. Still, judging by the water flowing into the street, the soak period wasn’t long enough.
In the old regime we had two 10 minute cycles and three 15 minute cycles followed by about a 30 minute pause. So, when we ran two cycles there was an 85 minute delay between the two runs in the front yard. (10+15+15+15+30). Runoff was minimal.
I should probably move this over to support with a headline of, “why are we getting so much runoff?”
@mavery76266, it might help to edit your soil type to change the cycle/soak times. Do you know what type of soil you currently have selected? The support team (email@example.com) can help you review if you’re unsure (please refer to this post for reference).
I like the sound of bringing in history. My preference would be to apply it in terms of flex schedule prediction/estimation, but not to actual execution.
I agree forecasts are 50% a week out and less than useful so history helps.
But I’m getting more comfortable with flex the longer I use it. If I could get a report that shows historical use based on actual, next weeks estimates based on forecast, and the rest of the year based on historical, I’d be a happy camper. Esp if it was presented as a drill down on the water usage graphs, with one more depth added to the drill down - by zone.
By the way, did rain here this week on a day it was scheduled to water. Now I understand how it works