Hello there. Soil moisture sensor idea.



  • Hi guys, thought Id better jump in and say gidday as have been lurking/researching for a wee while now.
    Big thanks to you developers first of all, talented, helpful, generous folk that you are.
    I am an avid reader of HAD, where I first heard about mysensors. Have been using mediaportal since 0.3c (I think).
    I am undertaking my first attempt at automating anything, and am planning to 'hook up' my tunnelhouse.
    So far I have an ethernet gateway and a single temp sensor. Took my a while and a couple of delete/reinstalls but got it working. I really like domoticz, gave up on openHab as I coudnt work out the graphing/permanence stuff. Just spotted mycontroller tonight and am keen to try this also.
    I would like to have soil moisture sensors but they are either expensive, tie up too many pins, or tend to drift/corrode from what I have read. So I thought to myself why not dry a DHT11 under the soul in a wee inverted dome, with some cloth over the mouth to keep the bugs/soil out? Would be cheap, reliable, also measure soil temp (another goal) and use one pin/sensor. Couldnt find any other discussions on this idea on the net, clever me! Or my google-fu is crap.
    So I set up a wee experiment after reading a bit with a pre weighed tin of soil with my sensor in it. Started out okay but the DHT11 just crapped itself after sitting at 97% for the last 24 hours or so. In its defence Ive had/negelcted it for five years and flambeed it a bit last night (heatshrink/lighter).
    So yeah, thought Id ask here if I am being thick somewhere before buying another one/trying again.
    After my tunnel house and garden are fully wired the kids bedrooms/heat distribution system is next.
    Cheers,
    Matt


  • Hero Member

    there are things around here, see http://forum.mysensors.org/topic/990/soil-tensiometer-sensor-network/20 and search the forum



  • Yep, did see that thread and am aware of the options in there. Basically I am looking at spending around $5/sensor, want something reliable and easy to implement. Was after thoughts on the DHT really and keen to know if anyone else has tried this. Cheers.



  • Huh, seems to be going again. Must be my dodgy soldering or something. We are down 11 grams and 3% humidity in 24 hours. I am hoping that the humidity of the trapped air in the wee inverted plastic jar is directly related to the moisture content of the surrounding soil.


  • Hero Member

    Hi @Matt. I think you are mixing things up. There is a diffference between humidity and moisture...

    some definitions:
    Humidity, is a measure of the actual amount of water vapour in a particular sample of air: measured as a partial pressure.
    Relative Humidity (RH), is the ratio of the actual amount of water vapour present in a sample (the Absolute Humidity) to that amount that would be needed to saturate that particular sample.
    Moisture is not as standard , but defined as diffuse wetness that can be felt as vapor in the atmosphere or condensed liquid on the surfaces of objects; dampness.

    The DHT11 is a (very inaccurate) capacitive humidity sensor which measures water vapour in the air and not suited for soil moisture. although there will be a certain correlation. Certainly the DHT will not be happy with being "soaked" for a long time and will corrode fast internally. Better to use a few corrosion resistant pins and measure electric resistance while changing polarity of the measurement current. (sorry for the complicated explanation).



  • Thanks for the that, I do get what you mean. But I don't think you get what I am doing. I have a small plastic container, open only at the bottom so water cannot get in. There is some holey type cloth covering this opening. Inside the container is the dht, cable is sealed where it comes out the top. So the dht is not in soil or water, it is in air only. I am hoping that this airs humidity and therefore my readings will have a direct relationship with the surrounding soils moisture levels....



  • To add, if I can establish a useful correlation, I will splash out and buy dht22s.



  • @AWI Sorry are you saying that the dht won't tolerate being in a high moisture environment for long, even if it is in (buried) air? Thanks again.



  • @Matt said:

    So the dht is not in soil or water, it is in air only. I am hoping that this airs humidity and therefore my readings will have a direct relationship with the surrounding soils moisture levels....

    I actually thought your idea was really good one until I did some research.

    The main problem with this thinking (I'm not an expert, so don't quote me on this :)) is that Relative humidity is relative. Meaning that for the same amount of moisture in the air, you will get a different humidity reading depending on the temperature of the air.

    Imagine that there is 16 grains* of moisture in your plastic container and it's 27 degrees celsius outside, the relative humidity will be around 10%.

    If the temperature drops to 4 degrees with no change in the moisture level, the relative humidity will have jumped to about 45%.


  • Hero Member

    @Matt I think I understand what you are doing. But like @mvdarend mentioned Relative air humitidy is not the same as Soil moisture. Although I think you will find some kind of correlation there are other parameters involved. As for the DHT11 (or the more precise DHT22, or SHT21, etc) these are sensitive capacitive air humitidy measurement devices which don't like a lot of direct wetness. They probably will last a while burried under ground in a plastic container.

    Electrodes will give you better information for a much lower price. There is a lot of info on the web... and on MySensors


  • Hero Member

    @mvdarend said:

    @Matt said:

    So the dht is not in soil or water, it is in air only. I am hoping that this airs humidity and therefore my readings will have a direct relationship with the surrounding soils moisture levels....

    I actually thought your idea was really good one until I did some research.

    The main problem with this thinking (I'm not an expert, so don't quote me on this :)) is that Relative humidity is relative. Meaning that for the same amount of moisture in the air, you will get a different humidity reading depending on the temperature of the air.

    Imagine that there is 16 grains* of moisture in your plastic container and it's 27 degrees celsius outside, the relative humidity will be around 10%.

    If the temperature drops to 4 degrees with no change in the moisture level, the relative humidity will have jumped to about 45%.

    That problem goes away if he converts from RH to dew point.

    What he's proposing is similar to this: http://www.adafruit.com/products/1298?gclid=Cj0KEQjwsb-vBRCLj7TvqpGx_MoBEiQALgFGnq9Gn8OFyRZu9j8FPptXyfjZMXGdPEkiry7SWkRE_20aAtjA8P8HAQ
    except that his cleverly stays above ground surface, so it shouldn't run the risk of getting soaked in water.

    Eventually, though, he may find that he cares more about the soil moisture level 6 or 8 inches down, and that will be different than what's measured just at the soil surface.

    One issue I see is that it's going to trap moisture in the dome, whereas the surrounding soil it's going to evaporate. So, before very long, your humidity is going to skew high. Also, if you have grass or other green plants under the dome, they're going to evapotranspirate, also artificially raising humidity in the dome while pumping the soil dry. That would be during the daytime though, not nighttime. So, if he were to do measurements at night and remove during daytime, maybe it would work better. Either that, or scrape the grass (or whatever green plants) off the area where the dome will sit on the soil so that they don't interfere.



  • @NeverDie Thanks for the explanation, I hadn't thought of dew point. The SHT10 you linked to looks interesting too, although it's way over the OP's budget.

    I don;t know if this has already been linked to, but this seems a nice cheap DIY solution:
    http://www.cheapvegetablegardener.com/how-to-make-cheap-soil-moisture-sensor-2/

    Edit: How about this one? It's still over budget but not too expensive:
    http://www.aliexpress.com/item/Protection-type-temperature-and-humidity-sensor-temperature-and-humidity-sensor-built-in-sht10/1650578403.html?spm=2114.031010208.3.76.Kuscd1&ws_ab_test=201407_2,201444_5,201409_3



  • Edit: How about this one? It's still over budget but not too expensive:
    http://www.aliexpress.com/item/Protection-type-temperature-and-humidity-sensor-temperature-and-humidity-sensor-built-in-sht10/1650578403.html?spm=2114.031010208.3.76.Kuscd1&ws_ab_test=201407_2,201444_5,201409_3

    This one seems to be just what Im trying to achieve I think? Inside the DHT is a capacitive moisture sensor, and I think this turnkey one does it the same way, so I gotta be barking up the right tree. Or maybe right paddock, wrong tree. Neverdie I think you are right I have been reading 95-93% since I started the experiment three nights back and there has always been condensation inside the plastic container. The weight of the tin can/soil is going down. I guess when the soil gets 100% dry I will notice it, but will be too late for my plants by then. Will give it a few more days though, its cold here and you wont get much evaporation from a tin can full of dirt with a plastic cup occluding 50% of the opening. I will try again with a can made out of wire mesh or similar, covered in cloth. Just a wee impermeable disk above the sensor. I dont have the facilities to make a sintered steel waterproof vapour permeable enclosure.
    From what I have read gypsum sensors are dirt cheap, but dont last and tend to skew when you fertilise, which I plan to do. Also they use up four pins each and require a number of passives. Or am I just being lazy here? To be honest I am lazy and want to find a 1 pin cheap solution and am still hoping the DHT will work if I get the enclosure right. Will also have to do some reading on dew point, thanks for that 😃
    My blimmin solenoid valve hasn't arrived yet, ordered it on monday =( Ex washing machine.
    Have a good weekend.
    Matt


  • Hero Member

    Ultimately you'll most care about measuring soil moisture at a particular depth below the surface. I suggest trying to build a cap sense soil moisture sensor that operates around 80Mhz.


  • Hero Member

    On the other hand, if you were determined to stay out of the soil, then IRT of canopy temperature as a proxy for plant water stress looks like it might be pretty easy to put together: file:///C:/Users/LR/Downloads/sensors-14-19639-v2.pdf



  • I am doing exactly this, and found this thread trying to work the problem.
    I have an Acurite temperature/humidity sensor that transmits on 433Mhz. With RTL-SDR and OpenHab, I can connect the sensor to my sprinkler schedule.

    I am converting the RH values to gH2O/m^3. In direct sunlight, I'm seeing 140F, and the response of the sensor doesn't seem to be very reliable.

    I'm not interested in accuracy, mainly repeatability. I can find the threshold myself to determine when the lawn needs watering. In my case, I'm using the sensor fasted to the bottom of a plastic show box. It's in the yard open side down.

    I'm toying with the idea of a 2D array with threshold values for watering. humidity by temperature.


  • Hero Member

    @brianmichalk Do tell us more. Sounds interesting.


 

Suggested Topics

191
Online

8.8k
Users

9.6k
Topics

100.4k
Posts