Soil moisture meters for lawn irrigation

[garbled]

I'm starting a new project to use soil moisture sensors to control lawn irrigation.  I have installed some Davis tensiometer sensors in the ground, and have them hooked up to a hobby boards 1-wire system.
 
Right now, I'm just starting to get data from the sensors, so I don't know much yet.  However, I'm curious if maybe anyone else has fiddled with such things.  Most of my questions revolve around what the proper values should be.  The Davis manual seems to indicate that you should water when the readings get around 25-40cb. But then water until they reach what?  zero?
 
Zero seems pretty wet.. So I'm really trying to just get a handle on where the right butter zone is for a lawn.   Any thoughts would be appreciated.

 

[bbruck]

I'm just now installing a Spruce irrigation system that has four moisture sensors. If you look at spruceirrigation.com, there are some articles on the amount of moisture that you should set things for...

 

[pete_c]

Here for a bit did utilize 1-wire soil sensors for irrigation then went to using Evapotranspiration numbers based on numbers from the Davis weather station, 1-wire sensors, UV et al.
 
The variances of top soil et al would have me installing more than 50 sensors which probably wouldn't look that nice anyways.  It would probably be more than 50 sensors to get a good number anyways. 
 
I am currently using an old Seagate Dockstar running Dockstar Debian and Mono to run my irrigation mcsSprinklers software.  It is inside of the old Rainbird box with two 8 zone serial controllers. Been working fine now for a few years.
 
Originally my irrigation software program was running in Wintel on Homeseer (early 2000's). 
 
Just noticed my sprinklers set up is on the Home Automation Wiki.
 
 

 

[Swancoat]

I apologize in advance for the lack of precision in this post, but I'm at work without access to the links I've saved regarding this.
 
I kind of want to set this up too. Just have the sprinklers check every morning if the ground needs the water or not. I also couldn't find the right values (heck, I don't even understand the units) for when the lawn needs water or not.
 
Ultimately, the solution I found was more of a 'top down' solution. Something like, "once your sensor is buried and set up correctly, fill a 5-gallon pail with water and pour it over, and then take the measurement 24 hours later - and that value indicates sufficient moisture" (Note, that is NOT the exact procedure - I'm just illustrating that it's that type of method as opposed to just solving for the value with math).

 

[Swancoat]

Ok... so I forgot the links would be in my phone, so I totally found the link, and lo-and-behold, the procedure I listed above is roughly correct. Amazing.
 
http://homeguides.sfgate.com/control-home-irrigation-systems-soil-moisture-sensors-100694.html

I like the entire idea of this top-down style method since it will account for all sorts of things like a bias or inaccuracy in your sensor readings etc...

 

[garbled]

This is similar to what I'm currently playing with.  I've been mucking with values and numbers for a few weeks now.  Right now, I have a pretty simple setup:
 
1) Every 30 minutes, check all 4 sensors.  Find the highest sensor value.  If it's over a specific value (right now, using 25cb) water the nearby zones with the sprinklers heavily.
2) Wait 30 minutes, check again, see if a different sensor pops, or if that zone's overspray fixed it.
 
I found some really good articles online about this stuff.  Many university research papers actually.  What they all seemed to say, was to determine via obvservation of the grass itself, the "stress point".  This is where the grass is starting to react to a lack of water.  Find this stress point, back off it a little bit, and then trigger the water there, so it's just prior to stress.  I'm just starting to get results and measurements via observation now.  I think it's a long process, maybe taking about 2 months to really understand. I'm also trying to not accidentally kill the lawn while learning, so there's that.