Anyone do any graphing?

This is where I've been impressed by conversations with folks from Rachio.  They're really into automating residential irrigation and using a lot of fresh approaches to it.  There's definitely something to be said for on-site monitoring and data collection.  But that brings along a lot of on-going maintenance.  Whereas using internet/cloud services (wunderground, etc) lets you get "close enough" to do nearly as a good a job without all the on-site gear.
 
 
Will you be watering all year round or do you have a winter where you live?
I live in the Pocono Mountains of Northeastern Pa.  We definitely do have a winter.  My watering schedule would begin around mid April and run thru mid September. I use the most water in the June and July months.
 
 
 
Are you utilizing your own water source (well / lake?) or do you pay for your water?
Although I live in a wooded area, "city" water is available which I use and I pay for.  The water pressure is a constant 65 PSI. 
 
Thanx for posting that article.  It's very informative.  At this point, I will probably set up a pilot project for the 2016 watering season.  That is, I'll implement my project using only three watering zones.  One "sunny" zone, one "shady" zone and one "semi sunny" zone.  Capturing soil moisture data and water consumption for each of the three zones and pushing it to ThingSpeak for analysis.  If the pilot project proves to be successful, I'll completely implement it for the 2017 watering season.
 
Yeah where I was going with this was relating to maximizing your own data numbers with your hardware.
 
I live in the midwest and the great lakes have much to do with the weather.  You can have snow or rain in one area and go 1/4 mile over and it can be sunny.  The great lakes alone do some willy nilly stuff with the local weather here. 
 
I do like looking at the new hi resolution almost 3 D weather maps these days with historical snapshots of moving weather pattens.
 
I have one installed water meter dedicated to the irrigation system.  It is next to the shut off switch.  The manifold box / manifold / solenoids / wires are sort of elevated in a box in a berm some 40 feet from the water source.  From there the last 3 zones are on the other side of the house running some 300 feet or more.  I did originally adjust the rotary heads for the use of a Davey pump/increased pressure.
 
Capturing soil moisture data...is difficult sometimes.  Here originally played some with gypsum sensors and very wacky numbers.
 
Old fashioned way to get some indication of how much water time and utilization relating to your sprinkler heads (flow etc) is a sort of manual test.
 
Use a can or a bucket and a ruler.  Turn on your irrigation for one zone.  Put the bucket in the middle of a zone.  Use a ruler to figure out the time required for 1" of watering while watching how many gallons are utilized per zone of water using your water meters.  So say it is 30 minutes of sprinkling that provides 1" of water that creates a 100% soil moisture. 
 
Yeah where I was going with this was relating to maximizing your own data numbers with your hardware.
Well, from a bean counter's POV, one could develop a statistical correlation between the changes in watering schemes and variations and time lengths of weather patterns. In my situation, this may require a few years of data acquisition since I'm only watering at most five months out of the year.
 
 
Capturing soil moisture data...is difficult sometimes.  Here originally played some with gypsum sensors and very wacky numbers.
Yea, I'm sort of expecting some wacky numbers etc.  Thus, the reason I'm starting with a pilot project of only  three zones.  As you would say "baby steps".  The upside is that there are quite a few soil moisture sensor modules on the market and they're all relatively inexpensive.  Maybe, just maybe I might be able to find a very reliable sensor.  Every project sounds great in theory but in my experience, actual implementation often provides a lot of unexpected gotcha's.
 
 
 
Well, from a bean counter's POV, one could develop a statistical correlation between the changes in watering schemes and variations and time lengths of weather patterns. In my situation, this may require a few years of data acquisition since I'm only watering at most five months out of the year.
 
Here saw my progress over the years based on old and new satellite pictures.  IE: I have Terra server pictures which I utilized to see.
 
Initially here did much infrastructure stuff relating to run off et al and eating top soil stuff.  Concurrently also added mechanisms of recording stuff.  Primarily it was all 1-Wire stuff. 
 
The software evolved a bit originally running it as a plugin for Homeseer.  I had so many connections to the software went to running it independantly from Homeseer.  Mostly though with all the doo whats of analog connectivity mixed in with historical values and internet live values the irrigation is based on ET values.
 
Yea, I'm sort of expecting some wacky numbers etc.  Thus, the reason I'm starting with a pilot project of only  three zones.  As you would say "baby steps".  The upside is that there are quite a few soil moisture sensor modules on the market and they're all relatively inexpensive.  Maybe, just maybe I might be able to find a very reliable sensor.  Every project sounds great in theory but in my experience, actual implementation often provides a lot of unexpected gotcha's.
 
Really you would have to have hundreds of reliable sensors maybe to come up with some good numbers unless your topology is all the same, top soil is the same, et al. 
 
Have a look:
 
Chapter 5 - Introduction to crop evapotranspiration (ETc)
 
Following soil wetting, the vapour transfer rate from the soil is high, especially for crops having incomplete ground cover. The combined surface resistance of the canopy and of the soil determines the (bulk) surface resistance, rs. The surface resistance term in the Penman-Monteith equation represents the resistance to vapour flow from within plant leaves and from beneath the soil surface.
 
Ideally then if you take your local analog values, mixed historicals, area weather values, UV / solar et al you can guesstimate how much water you need to have a nice lawn.
 
So far what I am seeing with the internet connected irrigation controllers is mostly the ability to remote control your irrigate with a smart phone which really doesn't automate your irrigation.  They haven't figured out the smart pieces of the irrigation and just attract customers with the eye candy. 
 
Old irrigation companies are now offering smart irrigation controllers which sort of work but still depend mostly on external resources which at best will give some very round numbers.  Here started by installing a Rainbird system years ago.
 
I did totally take apart and had a look at the original analog timer.  I didn't bother with it and replaced the the solenoid control with a couple of Rain8Nets in the early 2000's.  From here went to software serial control of said valves.  Basically the more sensors I added and the more historicals and internet data I added the better the irrigation got.
 
Rain Bird ESP-SMTe Series
 
Other Smart Sprinkler Timers claim simplicity and precision; the Rain Bird® ESP-SMTe Smart Control System delivers. Combining the proven simplicity of the Extra Simple Programming (ESP) controller family with the accuracy of weather-based control, this innovative system can now expand up to 22 zones.
 
With mounting regulations and severe drought in many regions of the country, the days when water inefficiency was overlooked are gone. The future promises even more restrictions for the irrigation industry. But along with these challenges come opportunities. Pioneering smart irrigation technology, Rain Bird was the first full-line irrigation manufacturer with a controller to earn the EPA’s WaterSense label, ensuring it is at least 20 percent more efficient than non-labeled products.
 
All models of the Rain Bird ESP-SMTe controller are EPA WaterSense approved.

What is the WaterSense Label
The U.S. Environmental Protection Agency developed a WaterSense program to make it easier for consumers to:
  • Recognize water saving products by labeling products that meet water conservation requirements. 
  • Gain access to new water saving innovations.
  • Understand the value of water efficiency.
  • Reduce strain on water resources.
How does the ESP-SMTe Save Water
The ESP-SMTe integrates technology used by top golf courses for decades and puts it into a user friendly controller.
  • On-site rainfall data - The ESP-SMTe collects site rainfall amounts and uses this information to re-calculate run times for each zone.
  • Historical weather data – The internal memory of the ESP-SMTe contains over 8 years of historical weather data.  This data, along with the onsite rainfall and temperature information is used to adjust the watering schedule on a daily basis. 
  • Reduced water run-off –The ESP-SMTe Cycle and Soak™ feature allows the controller to adjust zone run times based on soil conditions and slope to reduce/eliminate water run-off. 
 
Really you would have to have hundreds of reliable sensors maybe to come up with some good numbers unless your topology is all the same, top soil is the same, et al.
 
Nope.  I actually meant different brands of sensors.  I only foresee using one possible two moisture sensors in each zone. There seems to be a fair number of soil moisture detection sensors.  So, maybe, I have to test possible  10-12 different manufacturers sensors.  Since most of these sensors are in the 2 to 5 dollar cost range, it gives me the impression that they may not be heavy duty commercial quality.  The projects I've seen built with these sensors are indoor projects such as the testing the moisture level of a potted plant in a well controlled environment.  So, it's not a question of how many sensors to use but the accuracy of the selected sensors that may provide wacky numbers.
 
I'll probably have to determine the best way to make these sensors weather resistant. Weather may be one of many factors influencing the reliabilty of the sensor. Questions such as how heavy torrential rains will affect the sensor.  If the sensor is installed in a depressed lawn area, it may very well be saturated with water.  Not to mention how will close proximity lightning stricks affect the sensor. Will chemicals affect the sensors?  If so, how will they affect the sensors.  How will soil salinity affect the accuracy of the sensor?  Also, have to factor in the accuracy rating of the sensor.  Most sensors seem to be in the +- 2% range at 25 Celcius.  Another question, how does temperature swings affect the accuracy.
 
This is an ongoing learning experience for me.  It seems the more I learn about the sensors in particular and lawn irrigation in general, the more questions I have.  Here is a link to an interesting video  http://www.decagon.com/en/education/virtual-seminars/virtual-seminars/why-your-soil-moisture-sensor-may-read-negative/
 
For me it's all about maintenance.  I'll trade inaccuracy based on infrequent spot-checks (be they merely by plant health or handheld probes) over maintaining (and winterizing) a network of sensors.  Perfection is a fine idea until it consumes all your damned time trying to keep it running.  
 
wkearney99 said:
For me it's all about maintenance.  I'll trade inaccuracy based on infrequent spot-checks (be they merely by plant health or handheld probes) over maintaining (and winterizing) a network of sensors.  Perfection is a fine idea until it consumes all your damned time trying to keep it running.  
I agree.  I'm really beginning to think that Pete's suggestion of using a bucket and a ruler is probably the simplest and easiest solution after all.
 
I used the bucket test to just do a visual on the documented flow rates (and time), gallon usage, et al to provide optimal saturation of the lawn. 
 
I have over the years added two water meters to get a bit more granular with water usage.  That said I did do a bucket test to confirm the accuracy of my new city water meter / older installed water meters.  It is easy and fast to do.
 
This morning looking at CWOP weather station stuff noticed that I fat fingered the software configuration of the location coordinates.  That put my weather station 28 miles from it's actual location.  CWOP didn't send me an email relating to the discrepancy but did post a message on my CWOP weather page.  I didn't get any messages or see anything relating to the discrepancy on my WUN stuff.  Typically I do get emails from WUN if something is amiss.
 
Back
Top