I am building a moisture sensor with an AC current to prevent electrolysis.

The AC will be generated with a simple schmitt trigger oscillator with the midpoint reference set at 2.5V.

Water would bridge the gap between two tracks. It has an approximate resistance of 200K. The switch and the resistor represent the water.

When the oscillator is at 5V, (conventional) current flows from let to right. When the oscillator is at 0V, current flows from right to left.

When the switch is open, no current flows to the 2.5V.

I would like to detect whether or not moisture is present. This can be by detecting the reversal of current. How could this be implemented or is there a better solution? PS: this is a resistive measurement, not a capacitive measurement.

EDIT: I didn't realise that the answer was not 50/50 bias. Well spotted. I also noticed another issue. There will be multiple probes which will all have a common point (e.g. oscillator) and another side which can be probed.

enter image description here


1 Answer 1


Let me suggest a slightly different circuit:


simulate this circuit – Schematic created using CircuitLab

R1 is your moisture sensor. It's driven by AC, because the input and the output of the left inverter are L/H – H/L with a ratio of 1:1. C2 and R2 are a half-assed high pass (good enough). The more moisture, the lower the value of R1, the higher the frequency, the higher the voltage on the right inverter input.

You have to play with the values.

  • \$\begingroup\$ This will not give 50% duty cycle across the sensor probes. The current flow between the probes depends on electrolytic action so there is always an offset to contend with. This offset will also depend greatly on the depth and exposure of the probe tips and their composition. You would also need a resistor across the sensor probes (a real R1) to ensure a fixed maxima otherwise you won't produce a signal when not in the ground. \$\endgroup\$ Nov 25, 2018 at 19:36
  • \$\begingroup\$ True, but these are thing one can hardly come by with any simple approach. Analytic probes had to be calibrated before measurement, then inserted, removed and washed. There are solutions for this. Total overkill. \$\endgroup\$
    – Janka
    Nov 25, 2018 at 19:39
  • \$\begingroup\$ This circuit, just like the OPs original post will result in corrosion/plating of the probes and therefore the readings are always questionable. You have to drive the probes with capacitive coupling to ensure you minimize the corrosion and plating of the probes. If the OP is serious about understanding soil resistivity measurements there are endless articles out there. To understand how bad these circuits are simply place the probe in distilled water ….no reading, yet the probe is in fluid!!! .Start here: corrosion-doctors.org/Corrosion-Kinetics/Ohmic-drop-soil.htm \$\endgroup\$ Nov 25, 2018 at 19:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.