Soil moisture sensors

I'm trying to use a Watermark soil moisture sensor in the my home automation project. I believe this sensor cannot be read as a standard resistive sensor using a DC voltage across the sensor, the voltage across the sensor must be A/C.

Also, Irrometer who make the sensor don't seem to want to giveout details of the sensor's response curve, and temperature correction details.

So,

Does anyone have, or know where i can get an interface circuit schematic diagram for the Watermark sensor.

Does anyone have, or know where i can get details of the response curve, and temperature compensation details.

Thanks in advance

Keith.

Irrometer doesn't want to be responsible for calibration of your instrument or the likelihood that the data in the table they provide is not close enough to the performance of your soil-sensor system for your purposes.

Irrometer also doesn't conform to a retail market expectations because they don't cater to the retail market directly.. In a previous thread in this newsgroup another poster got all bent out of shape because Irrometer supplied some of their data in AutoCAD format instead of whateveritwas that the poster wanted. But they've always been helpful to me and other researchers. I've used their watermarkers off and on for ~20 years.

Third-party interface modules for waterwmarks are available, but traditionally watermarkers have been hooked up directly to data loggers with the data loggers providing the needed pulse excitation. I have some tables that it may be OK for me to pass on to you. Also see the linear temperature correction suggested below.

However it is best to calibrate your sensors yourself in _your_ porous medium (typically soil). The sensors are semi-quantitative at best but are well suited for binary decsion-making in home automation setups (water/don't water).

And here are some general instructions quoting from a previous post of mine (Marc Hult) in this newsgroup (comp.home.automation) 4-5 years ago:

Watermark Model 200 sensors can be treated as a resistance that varies non-linearly between approximately 550 ohms ("wet" = 0 centibars) to

28,000 ohms (200 centibars = very dry). However, the length of time that a DC excitation voltage is applied to the sensor must be short enough to prevent significant hydrolysis of water and consequent formation of gas bubbles on the sensor's internal electrodes.

The manufacturer,

, recommends 5 VAC or a pulse between 5 and 20 volts of 40 millisecond duration at a current of 200uA or less. If repeated measurements are needed, the leads should be shorted together or the powered side shorted to ground for 30 seconds to remove any accumulated DC potential in the sensor. The sensor will self-discharge without shorting in about 10 minutes so several measurements per hour are practical without shorting. The sensor resistance is inversely proportional to temperature (correction of 1 percent / degreeF from 75 degreeF)

Also, in order to avoid ground currents that can permanently damage the sensor, neither lead should be connected to earth ground for extended periods.

These requirements can be met without adding series capacitors if the sensor is connected through a CMOS analog IC switch.

HTH ... Marc Marc_F_Hult snipped-for-privacy@nothydrologistnot.com

Aside from minor technical quibbles, this is a very good write-up. The ever-handy 555 timer circuit is easily built and presumably works.

Note that the tables presented are not the tables provided by Irrometer with their stated/recommended constant-frequency excitation but rather the response when using emesystems' 555-based circuit.

Consider 'calibrating' the sensors phenomenologically ("Turn on the water when the reading reaches x whatevers because at x whatevers, the plants look thirsty".

HTH ... Marc Marc_F_Hult

This page duplicates the PDF and also has links to other sources of information on the Watermark sensors.

The Watermark sensor varies resistance inversely with soil moisture content. It should work fine with a DC analog sensing circuit.

There's a UC website which includes a downloadable spreadsheet you can use to graph sensor readings. The URL follows:

Enjoy.

That's not unreasonable considering widely varying soil properties and plant water requirements among users. Anything they say about performance in one environment might be off considerably in another.

Note: Upon further study I see my original comments about DC sensing circuits were wrong.

Thanks Marc and Dave,

The links you provided seem to show everything i need to know.

I'm going to be using the sensor in a lawn irrigation system - do you happen to know if these sensors can be left in the soil all year round ?

Marc - I did see the post you mentioned whilst searching through google groups - i can't beleive that geezer, obviously got out of the wrong side of the bed that day.

Thanks again,

Keith.

I have some more questions about the interface circuit given,

The circuit shows an LT1761 IES-33 3.3V regulator - this seems to obsolete with my suppliers, can anyone suggest an alternative.

Also - the circuit shows two 4.7uF ceramic caps, to isolate the sensor from the circuit. I'm having difficulties finding suitable ceramic cap of this value - can anyone one suggest an alternative ?

Thanks,

I doubt the regulator is critical or even that it needs to be 3.3V although a different voltage will change the curves. Look at the LP2950/2951 family. Search on LP2950 at Mouser or Digikey.

Mouser 80-C340C475M5U is a 4.7µF multilayer ceramic capacitor. They are pricey. I'd try n>I have some more questions about the interface circuit given,

The LT1761 is/was a low noise, low power device with good transient performance. emesystems docs suggest that this device was chosen for known-good performance under "pulse" conditions. If you have a scope, you could look at the power rail (output) of a substitute device (most any 3.3 regulator -- or higher with commensurate signal output voltage).

But if you are running wires to the watermark, rather than using RF to send the data, and can provide good DC source, power consumption should not be an issue. For belt and suspenders engineering , simply add more capacitance to the input and output terminals of the regulator with another several ufds of electrolytic and a small cap with good high frequency characteristics (eg ceramic 0.1 ) Depends in part on connection to (remote?) power source.

You could try audio-grade mylar or non-polar electrolytic capacitors. Depending on how cold it gets in the winter where you are, you might want to confirm suitable storage temperature range of electrolytics if you are going to over-winter the probes and signal conditioner in the ground. I suspect emssystems chose ceramic over electrolytic owing to long-term storage issues and high voltage (think lightning) and ceramic over mylar owing to size.

Another (better?) way is to omit the input caps in favor of switching between sensors using (eg) small DIP signal relays. Leave all sensors unconnected (not grounded) while not measuring and so avoiding long-term DC offsets that could lead to change in sensor characteristics over time. Again, this presumes that you are running wires from the sensor to _nearby_ logger or signal input multiplexor.Think lightening (worse case) and ground currents (ubiquitous).

Or (best) build one 555-input circuit with input cap and switches for different inputs.

Or just buy their module(s) in appreciation of the nice DIY write-up that they provide ;-)

Where and how you install the sensor is important. Unless you are doing this for a PhD dissertation, place the probe(s) in the _root zone_ at a location where the plants of concern dry out first. If you place the probe too high in the soil column, it may record precipitation/irrigation that never makes it to the roots. Too low and it may not reflect near-surface moisture variations. One sensor is not very many for this sort of work ..

HTH ... Marc Marc F Hult

As I tried to make clear at the time, there weren't many (any?) useful and affordable alternatives to the Watermark for soil moisture measurement by DIYers. Davis weather systems subsequent adoption -- and this thread -- would seem to indicate that this is still true ~5 years later ;-)

Some folks jist reeely reeely wanted to convince themselves that dependable and meaningful soil moisture measurement is easier than it really is ....

HTH ... Marc Marc F Hult