Originally shared by Fredrik Wiberg
Soil moisture fail. The soil moisture sensor I used for my Plant Monitoring was a huge fail. I used a cheap sensor from china that didn’t seem to handle water so well! ! The conductive foil has deteriated! The new solution is a lot simpler. 2 wirer soldered to two nails. Seem to work just fine.
You could generate AC current with your microcontroller. Then both sites would deteriate but slowly.
Solution: capacitives sensors
http://www.dfrobot.com/index.php?route=product/product&product_id=1385#.VzCC16ghXqA
How old is the probe, and how many powered hours has it logged? My plant water probes only power up the sensor for 0.5 seconds per six hours… just long enough for the op-amp on the sensor to stabilize and the adc to capture a value. Then the entire thing is turned off.
schematic for an i2c based module here:
https://drive.google.com/open?id=0B8I8gtvTqCqrRnRzUjg3Ynp0dUU
I used an mcp23008 to power the soil sensors, and an ads1115 module (from ebay) as the adc.
With electrolysis the metals move from one pole to the other .
If you use alternating current almost no deteriaration will occur.
@hagay_godovanik do you have a link to share, how to sample ac voltage with one of these conductance probes? oh, how will the ESP8266 generate the ac voltage?
If you have 2 adc available you can conect the probe to 2 output pins via resistors whose values are soil dependent. You then alternate 10 01 on those outputs and monitor the diffrence on the probe point’s.
Once upon a time i used a sine wave generator and opamps but that might be overdoing it.
Mind you that to realy adjust these probes is very soil dependent and even roughfly accurate takes a long time.
