This weekend we built a couple of resistive moisture sensors. While I cannot claim to have come up with the idea, I am nonetheless proud of building these moisture sensors from scratch. The basic principle is simple, embed two electrodes (wires) in Plaster of Paris which drys solid but porous. When dry the sensor will have infinite resistance (i.e an open circuit) but when placed in damp soil the plaster will absorb some of the water and the resistance will drop down into the 100 kilo-ohm level. By applying a voltage to the sensor and measuring the voltage drop across it you can get a pretty good idea of how moist it is.
Here is a series of pictures showing the construction process (which I will describe below)
We had some 20 gauge uninsulated wire laying around which we decide would make decent electrodes for these sensors. If I were to do it again I would try to find some insulated wire so that I would only need to strip the ends instead of heat-shrinking most of the wire.
The first step was to modify some electronic header material for use as spacers. We broke four two pin sections off of the header and pulled out all of the pins. Next we enlarged the holes for our 20 gauge wires. If I do this again I will also try to find 22 or 24 gauge wire so I don't need to enlarge the holes in the spacers.
Once the wires were inserted into the spacers and bent to hold the spacers in place, we added two types of heat-shrink a small diameter tube around each of the wires and a larger diameter to hold the two wires together.
Next we taped one end of our straw sections closed and filled them with Plaster of Paris. Into this mess we inserted the bare wires and spacer construction.
Then wait... and three hours later sensors are ready. I used an exacto knife to cut off the straws and then let the sensors dry for a couple more hours.
Ta-Da cheap moisture sensor.