Excellent Soil Sensor

[czuvich]

I just wanted to say your product is great.  I've been using it for the last several weeks in my raised bed gardens, and it works as expected.  I have tried a few other soil sensors, and you can't really beat yours for the price.

Having said that, I am using your sensor in my product.  Basically, it's an intelligent, garden management platform.  I'm still working on my website; however, my product has been working great for several weeks!  I use the soil sensors to determine when to open the water valves.

Anyways... feel free to PM me if you want to check it out

[pinolec]

It's been a while since your post. I had no time to answer it straight away and later I forgot about it, sorry 

I am glad you like our sensor 
I have noticed you removed the link to your website the product. We don't mind links to products that use our sensors. Nice irrigation system!

Piotr

[czuvich]

After some extensive testing, here's my overall input.

The sensors (as stated) do vary a bit in terms of the 3V output.  They usually range anywhere between 2.75-2.9.  I have not seen any of them get to exactly 3V.

My application for these is in agriculture (outdoor gardens).

Since these sensors are FDR (I'm assuming), then you really need to be careful with the installation.  If any of the green stick is exposed to air, your results will be off (read very low).  My suggestion is to dig a small hole in the ground, bury the sensor (including the black part), and pack in the dirt.  Then I highly suggest watering your soil, and THEN taking a measurement.  You should see around 50% in mineral soil (composted soil in raised beds for gardening). 

I'm currently using the following formula for VWC (assuming 50% is the correct estimation... I am definitely looking for some help here with these sensors).

Code: [Select]

float VWC;
   
    // Calculate VWC
    if (sensorVoltage < 1.1) {
        VWC = (10 * sensorVoltage) - 1;
    }
    else if (sensorVoltage < 1.3) {
        VWC = (25 * sensorVoltage) - 17.5;
    }
    else if (sensorVoltage < 1.82) {
        VWC = (48.08 * sensorVoltage) - 47.5;
    }
    else if (sensorVoltage < 2.2) {
        VWC = (26.32 * sensorVoltage) - 7.89;
    }
    else {
        VWC = (62.5 * sensorVoltage) - 87.5;
    }
   
    if (VWC <= 0.0) {
        VWC = 0.0;
    } else if (VWC > 100.0) {
        VWC = 100.0;
    }
I actually disagree with the 43% VWC calculation mentioned on this forum, but please help me understand why 43% instead of 50%.  I saw 50% on the following white paper as well http://edis.ifas.ufl.edu/ae266.

In regards to soil specific calibration, I have not gone through the 24 hour calibration of my soil.  With that being said, I think I am seeing around 90% accuracy with these sensors uncalibrated for my needs if I follow the installation steps carefully. 

I've considered calibrating my soil and using that as a baseline for other gardeners, but please help me understand if that would be feasible.  Overall, for the money these sensors work very well.  It would be nice to not have to calibrate the sensors, but I think that would probably dramatically increase cost.  Ok... any input is greatly appreciated!

[pinolec]

Yes, the sensors vary in output but this is most apparent when readings are close to soil saturation. For lower VWC the error introduced by sensor output variance is usually less than 1% VWC.
The installation is very important. The sensor must be in contact with the soil. Any air gaps or difference in soil compaction will have an effect on readings.
Composted soil usually has a high content of organic matter and indeed may saturate at 50% or even higher.
Different soils will saturate at a different level:

Very sandy soils have SP values of less than 20 percent; sandy loam to loam soils have
SP values between 20 and 35 percent; and silt loam, clay loam and clay soils have SP values
from 35 to over 50 percent.

Our generic calibration (conversion curve) will not fit well for sandy, clay or high in organic matter soils. We recommend specific calibration for such soil. Once calibrations are done for such a soil it should be ok to reuse them.

So the 43% or thereabout is to fit as many soils and closely as possible.

Regards,
Piotr

[czuvich]

Thank you for clarifying.  If anyone has done a soil specific calibration + formula with general outdoor vegetable gardens it would be great to share.  Otherwise, I'll post here.

I would be curious though, is there any way to get an estimate on the level of accuracy of the sensors for high organic soils is without calibration?