Soil moisture sensors better than ET for irrigation

[NeverDie]

Seems like measuring soil moisture is the most direct way to manage irrigation. There's ugmo (http://www.ugmo.com/) for numerous wirelessly monitored soil sensors (one per zone), and I'm not sure what else other than some weather stations (e.g. Davis) can do some soil moisture monitoring and various home made devices and systems. The Lowes IRIS pulse moisture meter is possibly a new option also (although it seems at risk for getting mowed down to ruin, whereas the ugmo is totally buried).




ET is open loop, and it's also generally just one value per site rather than per zone, which means you'll need to tweak a multiplier for each zone via trial and error. I'm not saying it can't be done, but likewise hitting the mark may not be easy and isn't assured.

On the other hand, I'm unsure as to whether soil moisture sensors remain accurate over time. Some designs do seem to degrade. What are some that don't?

Am wondering what those who have tried both ET and soil sensor end up preferring?

[Michael McSharry]

For the general user and for larger spaces the ET will provide a more satisfying solution. There are a few discussions in the mcsSprinklers forum some years back http://board.homeseer.com/showthread...light=moisture
http://board.homeseer.com/showthread...light=moisture #14
http://board.homeseer.com/showthread...light=moisture #3
In the later two Jim disclose where he did the actual comparison with results that were the same from both methods. He also references a comparison study. In the first Marc goes more into the physics.

Soils and the moisture patterns under the soil at various levels are not uniform. Closing an irrigation loop on a point sensor may be good for that point, but not for the point that is a foot away. Adding sensors adds complexity, maintenance and failure modes. There are situations where this is the best choice, but for the general user the ET will provide what is necessary to account for environmental variations. If you do not have weekly/daily environmental variation then the even more reliable timer will give the most dependable result.

[NeverDie]

Thanks for the links.

Because the first link circa 2007, I am curious about the soon-to-be available moisture sensors that Hult alluded to. For instance, I notice that in http://board.homeseer.com/showpost.p...31&postcount=7, Hult states:

Getting calculated "ET" methods to agree with reality was still a daunting task. That hasn't changed fundamentally, but better moisture sensors may soon be a reality for DIY HA.

Also, in his very next post on the same thread Hult states:

There are may also be capacitative soil-moisure methods affordable/in the offing for DIY (Methinks ;-)

It's now almost 7 years later, and so I'm wondering if the sensors ever became available and/or what he was referring to. Anyone know or care to guess? Unfortunately, Hult hasn't logged into this board since July, 2013, and his website (the one in his signature) has gone dead, so I'm unsure as to whether he will ever return.

[NeverDie]

By the way, I just noticed there's also an IRIS irrigation controller (and even an IRIS hose tap), and it appears there's a way for irrigating based on IRIS soil moisture sensors:

http://www.youtube.com/watch?v=kq6PQNoyF8Y

My local Lowes has the IRIS soil moisture sensors in stock (not sure about the IRIS irrigation controllers, as I hadn't even thought to look for them). If I recall correctly, the price was $29.95. I was going to confirm the price on the Lowes website, but I'm not finding it there. There are IRIS repeaters in case range is an issue.





I guess I (or my gardener) could just pull the sensor out of the ground before mowing and then re-insert afterward. i.e. Some hassle, but not a lot when compared to the total effort required for mowing the lawn.

[NeverDie]

I'm guessing Hult may have been alluding to probes similar to the following:


http://www.vegetronix.com/Products/VG400/VG400.phtml, which (unlike the gypsum sensor blocks available and commonly used circa 2007) are allegedly waterproof, don't corrode, and "can take an accurate reading in under 1 second." The one pictured here has a list price of $37.95.

I've read that the gympsum block sensors needed to be periodically dug up and re-planted to remain accurate (although I never understood why, except maybe the gypsum dissolved away, impairing the soil contact?), which would be a major hassle. Not sure if the same would be true for these newer sensors that are based on different technology, but I'm guessing their characteristics avoid the need for periodic sensor replanting. Anyone happen to know?

Looks as though very inexpensive (<$10) capacitive soil moisture sensors are also available and deliberately left open so as to be easily customized using Arduino IDE: http://wemakethings.net/chirp/

Anyone here using (or tried using) capacitive soil moisture sensors?

[Michael McSharry]

I have used both the Vegitronix and gypsum sensors. In the desert climate the vegitronix did well at showing the moisture decrease between the drip system irrigation. It was monitoring one drip head. It was sensitive to depth of install, but on a relative basis it was doing what I wanted. I have it interfaced via analog through a WebControl. http://board.homeseer.com/showthread.php?t=149865

The same sensor in the Northwest was useless as the overall volume of water did not change that much while the surface inch or two did fluctuate and the lawn was sensitive to this surface level water.

I have also used gypsum via 1-wire interface. If I installed it deep it never changed. If I installed it closer the surface then it fluctuated what looked like surface temperature.

My main point about using moisture sensors for general irrigation is that you are doing a point measurement and assuming that all soil receives the same water patterns and is uniform in moisture content. There is much more maintenance and install costs with moisture sensors.

[Uncle Michael]

I second Michael's point about the local nature of a physical moisture sensor. I've been using ET calculations for several years, and my major heartburn is generated not by the inaccuracy of the estimate (you can dial that in pretty well with some experience and trial and error), but the fact that for many of my zones the estimate is either a good approximation for part of the zone, but not so good for another, or a tortured "average" of areas that simply have different soil moisture contents.

Making matters worse is that the differences change through the course of the year. In the spring, when the water table is typically high, some zones have persistent wet areas. In the late summer, those same areas are often very dry. Yet, the rest of the zone may exhibit consistent behavior all year. Short of subdividing the zones, the basic problem is not the technology used to estimate soil moisture, it's the simple reality that the soil conditions, even within one irrigation zone, are the major source of variability, not the methods available to gauge moisture content.

[NeverDie]

My main point about using moisture sensors for general irrigation is that you are doing a point measurement and assuming that all soil receives the same water patterns and is uniform in moisture content. ere is much more maintenance and install costs with moisture sensors.

I thought the strategy wasn't that but rather to install the moisture sensor at the driest point of the zone. For instance, that's what UgMO recommends.

I can imagine there might be two different driest points: the driest point if you're irrigating only, and the driest point if getting rainfall only. Here, I would think the driest irrigation point is more relevant and also easiest to find. I only mention it because rainfall (because of a different distribution pattern) might somehow interfere with the reckoning.

[CharlieWayne]

Have you seen the moisture sensors from hobbyboards?

They work off 1wire and I think can populate data into mcsirrigation.

Sent from my htc Inspire 4G using Tapatalk

[NeverDie]

Have you seen the moisture sensors from hobbyboards?

They work off 1wire and I think can populate data into mcsirrigation.

Sent from my htc Inspire 4G using Tapatalk

Do you mean this one ($44)?



http://www.hobby-boards.com/store/pr...5%27-lead.html

It does say " This is the same sensor as the Davis 6440 Soil Moisture Sensor."

However, it uses the older resistive technology. Maybe it has been improved (?), but it's from the lineage of sensors that reportedly might need be dug up and replanted periodically to remain accurate--although I don't know whether that's true of this model in particular. Since it has no price advantage, would there be a reason to prefer it over the newer, capacitive sensors (such as the vegetronix above)?

[CharlieWayne]

Yes that's the sensor. I have no experience with this but that was something that I seen while looking into this topic.

Let us know what you find.

On another note, I seen you were talking about rotary heads in another thread. For the past year, we have been using K2 PRO from K-Rain. Tons of benefits using this Head such as True Flow Control and Easy Nozzle Changes to achieve proper water coverage. Lots more but I wont cover that on this thread.

Mixing types of sprinkler heads ( rotary vs. fixed spray) is truly a challenge to achieve uniform water coverage. If mixing heads while using Soil moisture sensors, it would seem that might not give you accurate representation of your zones needs. Without Accuracy your kind of negating the original intent of using such a system.

[NeverDie]

As it happens, just yesterday I finished changing over all my fixed spray to Hunter MP Rotators. The motivation was to apply at a slower precipitation rate (0.4" per hour) and also get a more uniform distribution. I doubt whether getting a truly uniform distribution is possible. Even if you did, then variations in soil composition and elevation within the zone would likely distort the intended effect anyway.

That's why I like the idea of managing irrigation to the driest point in the zone. It's seemingly simple, and it doesn't rely on idealized assumptions. Because of the changes I made, my driest points might be different now than before.

I have more work to do: now that I upgraded the spray heads, I was able to measure my head pressure in situ with the irrigation running, and measured head pressure is too high (outside the MP rotator spec) in a number of zones, so I'll need to address that. When I do, the driest points may shift again. If I do end up taking the soil moisture approach to irrigation (which I've only just begun to assess), it wouldn't be until after I'm done upgrading and tweaking the irrigation, because all that could affect the location of the driest point in a given zone.

So, now that I think about it, a possible failure mode with the soil moisture approach would be a change (such as a clog) in a spray head that covers the driest point. Such a failure would result in wasted watering throughout the zone. I think having ET as a check against that sort of failure (or a way of detecting/signaling the failure) might be useful. The answer might not be either/or, but both. An alternate check/alarm might be comparing against historical flow rates using a flow meter, which might be better and good to do anyway. As probably obvious, I'm still thinking this through.

[CharlieWayne]

Obviously there are several factors that effect watering needs such as soil composition, elevation, slope, sunlight or shade, actual plants consumption, and many others.

It would seem for most cases it would be best to run multiple sensors and then avg. the readings.

Now the issue becomes cost.. At $44 per sensor plus controller your looking upward of $200 per zone just in sensors and controllers. Factor in Labor & Equipment and the price will double or triple parts for a retrofit on existing construction.

So the question is now, Why Are we taking on such a project?

[NeverDie]

Obviously there are several factors that effect watering needs such as soil composition, elevation, slope, sunlight or shade, actual plants consumption, and many others.

It would seem for most cases it would be best to run multiple sensors and then avg. the readings.

Now the issue becomes cost.. At $44 per sensor plus controller your looking upward of $200 per zone just in sensors and controllers. Factor in Labor & Equipment and the price will double or triple parts for a retrofit on existing construction.

So the question is now, Why Are we taking on such a project?

If the IRIS thing works (for example), then it might be $30/zone, plus some system costs like the $99 wireless receiver-controller. I'm not convinced multiple redundant sensors are needed if there are ways to catch and flag errors.

Before getting into all that, though, I'd first like to decide whether it's worth doing in principle.

[CharlieWayne]

As it happens, just yesterday I finished changing over all my fixed spray to Hunter MP Rotators. The motivation was to apply at a slower rate (0.4" per hour) and also get a more uniform distribution. I doubt whether getting a truly uniform distribution is possible. Even if you did, then variations in soil composition and elevation within the zone would likely distort the intended effect anyway.

I have more work to do: now that I upgraded the spray heads, I was able to measure my head pressure in situ with the irrigation running, and measured head pressure is too high (outside the MP rotator spec) in a number of zones, so I'll need to address that.

To reduce pressure and achieve better wind resistance for Turf Rotary Nozzles, one should run the largest nozzles as possible. Nozzle Section is Crucial for uniform coverage IF one installed their Heads for "Head to Head" coverage. The simple idea is that you want to lay the water down on the turf as fast as the soil will absorb it. If you have slopes or clays then still run high discharge nozzles but start to utilize Round Robin applications (Michaels software does this)

Do some Google searches on nozzle selections but below are the basic idea based upon the heads rotation.

90 Degree = 1.5 Gpm
180 Degree = 3.0 Gpm
270 Degree = 4.5 Gpm
360 Degree = 6.0 Gpm

The idea above is that if a sprinkler only turns 90 degrees it will only cover 1/4 of the area of a sprinkler that turns 360 degree during the same period of time. SO the 90 degree head should have an output of 1/4th of the 360 degree nozzle.

Here, we combat 2 types of fungus in our turf, Grey Leaf Spot and Take All Root Rot. Both are promoted by over watering and/or to much nitrogen. This fungus can severely impact a lawn and cause thinning followed by weed invasion. Personal, I would prefer under-watering then over-watering.