Soil moisture sensor cable guide/holder/clip

This design allows for a waterproof full burial of the AITRIP 10 Pack Capacitive Soil Moisture Sensor Corrosion Resistant for Arduino Moisture Detection Garden Watering DIY Electronic for Arduino and Raspberry Pi.

Dec 2025 update - V3.3 posted

This version of the cable guide housing allows for reliable potting of the electronics without leaking epoxy all over.  Use a small amount of silicone between the front edge of the guide block and the sensor to seal the underside, and then a small amount of silicone to seal the cap on the topside.  This creates a nicely sealed cavity for filling with epoxy.  I use https://www.amazon.com/dp/B0B5DJJZT3?ref_=ppx_hzsearch_conn_dt_b_fed_asin_title_1&th=1.

I've perfected waterproofing these sensors for permanent burial:

• Spray with MG Chemicals 4223F (https://www.amazon.com/dp/B06XC3JBVJ?ref_=ppx_hzsearch_conn_dt_b_fed_asin_title_1) and bake at 100C for 2 hours. 

• Remove the 3-pin connector and attach a cable.

• Insert into the cable guide and seal per above.  Fill with epoxy.

These sensors should be good for a few years with this treatment.  Without the conformal coating the black paint (solder mask) will delaminate within a few weeks.

Note that I've had several sensor electronics failures probably due to really cheap components:

• The R4 1M ohm resistor was open on several new sensors this results is slow response and offset analog output voltage.  Otherwise functional.  I check this resistor before the conformal coating step.

• (I also change out R1 from 10k to 1k Ohms for a more analog response.  The output voltage range is ~2.85V dry to ~1V fully immersed in water.  I generate a calibration curve from output voltage (its not linear) to moisture level.)

• A few sensors have just quit working (0V output) after months of usage.  No delamination on the probe surface or corrosion in the potted electronics.  Probably electronic component failure.

Original post:

I printed this guide in PETG with no supports.  There is an overhang that may have to be cleaned out for the sensor board to fully slip in.  

Remove the 3-pin connector and directly solder an appropriate cable to the sensor.  Goop the whole thing up in Liquid Electrical Tape, including between the sensor board and the guide/clip.  Also carefully coat all edges of the sensor board exposed FR4 to protect against possible long term water incursion, and fully goop the cable wires.  Be careful to get NO coating over the sensor metal (under the solder mask) beyond the depth line.  Add a zip-tie to keep everything together.

The v2.0 sensor is a bit narrower than the v1.2 sensor.  I'm using modified v2.0 sensors to achieve AOUT values across the full range of moisture (water immersion). My Amazon review for the sensor link above, including my circuit modification, is also attached to this posting.

Fusion 360 model included.

Amazon review of the sensor:
They work.  Two issues, plus I made it waterproof for full burial

Issue 1:  The schematic in the listing is not accurate - part IDs are wrong, there are not 8 caps on the board, etc.  I traced the schematic, attached.  I did not measure the capacitors, and assume they track the posted schematic.  Also, my pin numbers on the VR are incorrect (I grabbed the wrong VR model).
Issue 2:  These are set up to have almost a digital response, which may be fine for some applications.  With R1 = 10k Ohms, AOUT is at about full scale (fully saturated, 1V) when the sensor is only dipped into water to about the "i" in "Moisture", which is less than 10mm.  I needed a useful reading across the full range of moisture, so I changed R1 from 10k Ohms to 1k Ohms.  See the attached graph.

Theory of operation - the 555 timer is set up as a free running / astable / multi-vibrator oscillator.  Assuming C3 is 470pF (per the schematic posted on Amazon) then the frequency should be ~870 Hz.  The duty cycle should be roughly 50%.  The output is fed thru R1 to the sensor capacitor, creating a AC voltage on the sensor capacitor that varies based on the moisture on the sensor - the higher the moisture (higher capacitance) the lower the amplitude of the 870 Hz on the sensor capacitor.  This waveform is then rectified and filtered and sent to AOUT (there is a good amount of ripple on AOUT).  AOUT swings between ~3V when dry to ~1V when fully wet.

Making the sensor fully immersible - I 3D printed a cable guide [Printables]() and gooped the electronics and cable connection up with Liquid Electrical Tape.