I've used the Vegetronix VG400 soil moisture sensor for several Arduino Fio projects (transmitting data back to a receiver) and the results have always been great. However at $30 per unit those suckers are costly. Can anyone suggest a cheaper sensor?
Thanks,
Andrew
Brainstorm modus:
maybe you can try to make one yourself. The idea is that dry ground doesn't conduct and wet ground does. So if you have two pieces of resistance wire plugged in the ground you get a different resistance depending on the depth of the water. So it behaves like a variable resistor. Make it part of a voltage divider and connect it to analog in.
Something like this
+ AnalogIn
|
+-------+ |
+5V +-| |-+-----------+ +------------+ GND
+-------+ | |
| |
~~~~~~|~~|~~~~~~
| |
| |
UNDER | | GROUND
| |
| |
+ +
You can make a sort of resistor wire by putting multiple resitors in series e.g 10 x 1K per wire. The resistor between 5V and the analogIn could be 10K . Now depending on the depth the total resistance of the "sensor" is approx 0K (high water) 2,4,8,10,12,14,16,18,20K (dry) . The resistors-wires can be taped on a wooden pin or so.
Total costs less than 5$ some tinkertime and some time to calibrate the application
Hope this helps,
Rob
It's hard to beat the Vegetronix sensor. It uses RF to measure moisture in a region around the sensor.
If you use resistance, you need to use an AC signal to measure the resistance. There are various polarization artifacts with DC resistance measurements in soil.
I have used gypsum blocks with the circuit in this article http://www.emesystems.com/pdfs/SMX.pdf.
I made my own gypsum blocks, which is cheap but takes effort.
A capacitive sensor might suit your needs as well.
http://www.dietmar-weisser.de/analogtechnik/BodenFeuchte/FeuchteSensor_1.php.
Advantages, it isn't influenced by the amount of nutrients, no electrolysis and it doesn't influence the soil.
You may need to translate the article though.
Use Rob's design, but use a digital output to supply the 5V, but only for a short period while you take a measurement, then turn it off again.
A lot easier than AC and it will be enough to prevent any electro-chemical effects.
I am also going around soil moisture measurement. Here you have other ideas for soil moisture sensing that i located on internet:
http://gardenbot.org/howTo/soilMoisture/
http://www.deepdarc.com/
I hope it could help you.
madepablo
Please help me stating the calibration methode and operating principle of vegetronix vg400 soil moisture sensor.
Although this is an old thread - this site gives some interesting ideas.
http://tuxgraphics.org/electronics/200908/eth-flower-watering.shtml#0lfindex2
Andy
Very informative thread Andy! bookmarked!!
I've made a sensor based on the tuxgraphics sensor tuxgraphics.org: Remote flower watering and monitoring so far I've only used it to see how wet my soldering iron sponge was 
Results are good. With nothing between the plates I get an output of 0.4v, with a wet sponge I get 1.8v (with MK 0.0.1 I got 3v but not reliably).
I found the sensor was intolerant of long leads. Long leads twisted together form a capacitor swamping the sensor plate capacitance and long leads not twisted together form a loop antenna picking up power frequency interference.
My sensor is driven by a 555 running at 5kHz, I've used the detector board from tuxgraphics Add-on parts Eth-flower-care. That lot is mounted in a 92x38x27mm diecast box. It is supplied with 5v.
This is a view of the 555 astable board. The plates are made from copper clad board painted with 3 coats of Hammerite paint.
View of the detector board and whole of plates.
How not to connect the plates. Those twists cost 50% of the signal.
How to connect the plates.
Andy