Measuring ground moisture level via conductivity

Well firstly, it's likely that the electrodes will become corroded/oxidised over time, so the resistance you measure can be expected to change dramatically as a function of time since installation.

Secondly, if you're measuring the resistance with D.C., which you are, then you're going to get an electrolysis effect, whereby each electrode will behave differently; one will attract positive ions, and one negative ions. I wouldn't be at all surprised to hear that if you swap thr DMM leads over, the measured resistance would be different.

You need to measure using low-voltage A.C. if this bothers you, and to use electrodes of an inert metal, or at least plated in such, like gold, platinum, rhodium (?), etc.. Well, that's my guess, anyway. No doubt others will have more knowledgable insights.

Martin

--
M.A.Poyser                                                  Tel.: 07967 110890
Manchester, U.K.                                http://www.fleetie.demon.co.uk

Not that weird - you made a battery - you can always call it a soil pH meter :)

Commercial soil moisture measurements are made with probes inside (I think) a gypsum block. They give a resistance reading which must be measured with ac excitation (to avoid electrolytic action damaging the probes).

Search the web you will find information and suppliers.

I

direction

chemical

think)

Yeah, the circuit I was sent uses AC for that very reason. My thought was that if I only measure the soil resistance once a day, and only for a few seconds, the oxidizing/electrolysis effect would be negligible. Certainly if I left the meter on 24 hours a day with a small DC voltage going, I'd have all kinds of interesting things happen. I could even alternate polarity from day to day.

So I guess there are two forces at work here. #1 is that there is some sort of chemical reaction occurring between my probes and the soil. This is why, on the first day, I measured resistances around 1k, and am now measuring resistances around 1M. Force #2 is the electrolysis effect, which is why my garden has no hair. No wait, wrong effect. The DC of my ohmmeter caused some movement of the ions in the soil, creating a battery, and that is causing the perceived difference in resistance going in the two directions.

Does that sound right?

That is so cool! I just went out & measured, and it's got a nice steady 220 mv DC! (and -220 mv DC measured with probes reversed, just for double-checking). It turned my wife's rose garden into a battery! I wonder what this will do to the roses ...

I'm going to grab a small resistor and try to drain the rose garden battery.

My gut feeling is that measuring the resistance with AC (which is what the Whetstone bridge circuit does) "averages out" the effect of the DC, which is another reason to use AC? Even turning the meter on for just a short time each day won't help, if I measure the resistance using DC I will get this battery effect which will throw off my measurements.

Thanks! Chris

then I

direction

chemical

corroded/oxidised

dramatically

others

110890

An interesting idea ... what would cause these "other currents"?

Any idea on a probe that will become corroded/oxidized more slowly? That's why I picked galvanized spikes, I thought they would be able to handle the corrosion better ... I'm not sure that gold-plated platinum spikes are in the budget (grin) ...

then I

direction

chemical

corroded/oxidised

dramatically

others

110890

I was not suggesting you buy a complete meter. I was suggesting you find out what a decent probe is like here

or here

that one is $30.

You have polarized electrodes, and electrolysis is going on. You need to use an AC Wheatsone bridge with totally inert electrodes. You might try carbon electrodes, from an old zinc-carbon battery. No metal showing, cover all conductive connections with epoxy. Initial connection to carbon via conductive epoxy. Cover this also with non-conductive epoxy. Bury them, don't smash them in. Use a sine wave at about 1000 Hz and earphones for the bridge null measurement. There is also a capacitive component to think about, since you're using AC. It's effect will be to give you a "fuzzy" null zone, unless you can sharpen it up with a compensating capacitor in the reference side of the bridge, parallel to the reference resistor. Cheers! Charlie

Aha! That's much more within the budget!

It looks like I'm dealing with a slightly different process here? What I have right now, is two "probes" with 3' of earth between them, and I was trying to measure the resistance between them. The probes you pointed out both have two wires coming out, which makes me think that I only need one probe (not two) per area to be measured? And so instead of measuring the ground moisture, I'm actually measuring the moisture in the probe? Which should be "closely related" to the moisture level in the surrounding soil? That's fine, by the way, I just want to make sure I understand what I'm doing!

Thanks, Chris

(2)

Basically yes - go search the web there is a fair amount of information to be found.

Forget the longwinded approach and simply use two different metal probes and a simple volt meter.

The cheap shop moisture probe I have uses this circuit and has two different metals separated by an insulator like coax cable and a half inch long pointed copper tip. The shaft is silver coloured but I'm not sure what metal it is.

There are different voltages produced between different metals in a salt solution and the moisture level in the soil will cause the voltage produced to change.

Drawback is the probe will corrode, but if you only want to test it once per day it is easy. Stick the 1/4 inch diameter 12 inch long probe into the soil and look at the attached meter.

It cost me less than $5.00 Australian and is now about five years old. It works happily for my girlfriend's pot plants but I don't know how accurate it is. It does show a difference when it's wet to when it's dry but I think different soils would have different salt levels which would cause calibration problems.

Hope this helps, Peter

That certainly is easier ... the problem is, when I said I only want to test it once a day, what I want to do is to hook up my PC so _it_ tests it once a day - so it's not that I want to run outside once a day and stick the probes into the soil, the probes gotta stay there...

and

different

metal

produced

per

soil

think

test

a

probes

What are you complaining about? It's the perfect excuse to justify building a robot to drive out there every day, poke the probe into the ground, and read the dial with its camera. Now you've turned your wife's rose garden into Mars!

Chris this is what I did. Five years ago I started where you are now but I used 3" lengths of 1/8" brazing rod spaced at 1/2" and buried 4" below ground. The probes are connected via twisted pair to the PC. The hard part is overcoming noise generated by ground currents. 50hz and its harmonics in my neck of the woods. I tried low voltage AC and a whetstone bridge but couldn't get reliable readings with multiple sensors and large ground currents. Maybe I did something wrong. After some very good ideas from s.e.d I applied near AC from an osc (using the +&- 12V rails of the PC) to the probes and used a full wave rectifier and averaging filter (national an20 page 10) to feed DC levels to an ADC in the PC. This has been a stop start project since I installed the probes 5 years ago and I've only recently got it up and running but the probes are still in good shape. I think the gypsum probes I've read about would be better as they should overcome the conductivity differences of soil types. Gordon my address needs gj in front for mail

test

a

probes

It

accurate

Or use thicker probes to allow for the corrosion that will occur. If you use

12" thick probes you will only have to change them every couple of years.

Still lots cheaper than the fancy ones if you get them from the local scrap dealer for $X.XX per pound.

Although I like the robot idea, (WIFE = Washing, Ironing, F_cking, Etc!) it doesn't say much for your gardening enjoyment if you don't go out to the garden! Might as well replace them with plastic flowers! (Same applies to wife!) You could then tell the moisture content of the soil by seeing if they have fallen over 'cause the ground is mud after not turning off the sprinkler.

If you want to see the moisture profile of the soil you really need a spade!

Peter

I dunno if my first post got thru (#@%^& Google) So I'll try agen.. I reckon zinc (galv) would be the worst choice, since zinc is in batteries. Stainless Steel would be better, I'd say try using some stainless steel cutlery, but a S/S pot scourer or SS wool would be better because of its larger surface area.

T.

Any 'battery effect' by one electrode is simply compensated for by the 'battery effect' of the other electrode - if electrodes are same material.

As stated earlier, currents are flowing through earth. Those earth borne currents make conductivity appear different in reverse directions. Testing requires AC current sourced by the tester, typically at frequencies that are not harmonics of

60 Hz (or 50 Hz > I dunno if my first post got thru (#@%^& Google) So I'll try agen.. I