Electronics > Projects, Designs, and Technical Stuff
Soil moisture sensors and ground loop issues
ogden:
--- Quote from: perieanuo on March 05, 2019, 05:44:28 pm ---The dcdc solution is the ONLY one.or rethinking the moisture detection...
--- End quote ---
It's not. There's wireless solution as well.
aberco:
Than you all, this is very helpful.
mikerj, I'd like to keep the sensors I have already purchased. I have used watermark sensors which are the standard in irrigation measurement: they have a granular material between the electrode that has been factory calibrated, so it is easy to convert sensor resistance into soil tension values (in -hPa or cBar).
JVR, as noted above, even when not using an SMPS (system is powered from a battery), I have problem when more than one sensor is in the ground.
ajb, I will give it a try... that's all I can think of right now and I can implement the DC/DC + optocoupler relatively easily: I am designing a little bodge PCB for that purpose. I have good experience in PCB design so I will certainly separate the sensor front end to the rest of the circuit and minimise coupling. The sensors are all physically separated by a couple meters from one another, which I hope will be enough. If not I will write a software routine to sequentially turn them on one at a time.
ogden, wireless could have been the solution, but I am also running other sensors like a CO2 sensor that consume 200mA. Having 16 battery powered units with that much power draw is just not very practical for my application.
I have attached a photo of the sensor units, the humidity sensing front end is on the upper right corner of the PCB.
Will send an update of the results as well!
aberco:
So I finally got the parts and got my little fix PCB manufactured.
I have used one of the common 1W 5V-5V isolated DC/DC converters. The load (555 and IR LED) is about 3.5-5.0mA, on the DC/DC input side it consumes about 25-30mA (efficiency is terrible for light loads, but of course it's the isolation that matters).
Also, at low loads, the converter is not delivering 5V but a higher voltage, so I rewrote my sensor calibration (resistance to frequency conversion by the 555 is dependent over supply voltage).
Electrically everything does work, now I have to plant it in the ground and see if the isolation does the trick!
aberco:
And it all work!
I planted 3 sensors in wet soil, and they all give super steady readings.
Thanks for the help and I hope this can serve someone interested in designing a similar application.
Navigation
[0] Message Index
[*] Previous page
Go to full version