Ground resistivity meter - DIY

[Baresel]

Hello Everybody, I'm new to this forum.

Recently, I have been experimenting with home made ground resistivity meters. What I made is simply a wood rod (1m long) with 4 electrodes on it, placed in even distances.

A 9V battery is connected to the outsidfe rods, while the inside ones are on a digital V meter.

When I stict it into the ground, I get readings in the range 0.05V-0.60V, depending on the type of soil.
I tested it in various places, where I exactly know what is underground.

In the place where a large block of concrete is burried, the voltage drops.

In places where some grass, branches and leaves from my garden were burried, the voltage goes up.

So, definetly if detects something.

My questions:

1. Now I'm using 9V DC battery. Should I change to AC? Some websites suugest that, but in no place it is expalined clearily enough.

2. What is a vertical range of such a device? Some sources suggest around 1/4 of the outside rods?

Thanks,
G

Look like this:

[homefire]

Your set up looks like you would be measuring the voltage created by the Rods you have connected to the DVM.

Electrolysis created by the rods.

Read This!

http://www.cflhd.gov/agm/geoapplicat...larization.htm

[Baresel]

Well, I'm a little confused.

This setup is made according to basic description of a simplest soil resistivity meter.
I guess it's called the Wenner method.

Should I change the V meter?

[homefire]

I guess it could measure the voltage drop set up like that?

Expert I are not. I was just making a observation.

[Max]

Well, I'm a little confused.

This setup is made according to basic description of a simplest soil resistivity meter.
I guess it's called the Wenner method.

Should I change the V meter?

Hi,
you must keep the currents small... otherwise already explained... you'll not just measure resistivity but electrolitic process take place and false readings.

So, the method is not wrong itself... just need some current limiting... and calibration. Better you'll look for a magazine project of this kind I think...

Kind regards,
Max

[homefire]

In fact that's what it looks like they are doing. Measuring the Voltage Drop of the sample area.

http://www.scribd.com/doc/13337594/E...tivity-of-Soil

I have a freind that does this for a living. But I do not know if they use this system.

They use A Pulsed 500 volt source.

They can determine the Ground Resistance.

Buy measuring the Decay Time of the Pulse , they can determine the chargeability of the material.


http://www.scribd.com/doc/13337594/E...tivity-of-Soil

[Carl-NC]

Instead of a DC voltage, you should use an AC current. Electrolysis won't be a problem.

[Baresel]

Thanks for your responses. Now it's much clearer!

I will try to switch to AC.

Is there any commercially available portable AC generator? I mean something independent of wall socket, for field work.

[Aziz]

This job is destined for a pure laptop solution. You don't even need any extra hardware for this. Only electrodes and a sophisticated software (you need to develop this).
You need:
- electrodes
- laptop with high fidelity sound card (stereo version)
- sophisticated software (using lock-in amplifier)

You can make a high accurate solution with this without getting problems with EMI. The sound output will be driven in differential mode (to increase the output voltage). The sound input will be driven in differential mode (to get rid of EMI). You need to implement a lock-in amplifier. Then you can resolve sub µV. As this is not time critical, you can even resolve nV signals.

No, this isn't a joke! It is really possible.

Aziz

[Baresel]

Wow, that is a very interesting solution.
Perhaps I will try it one day. For now, I am experimenting with the simplest setup possible.

[Aziz]

For (very) high ground resistances (i.e. dry ground), you very likely need some impedance matching electronics. A simple stereo j-fet or c-mos amplifier or buffer would increase the signal source impedance.
That would give more flexibility and precision. The sound card input impedance is quite low. The microphone input is a bit higher.

In the electrode graphics above, additionally ground electrode in the center position is necessary for differential measurements. The outer one would be driven with stereo output signal (of same frequency but one channel is 180 degree phase shifted forming an inverted polarity stimulation) and the inner electrodes would be fed into amplifier, buffer or directly to the stereo input line.

I did not try this yet. But should give superior measurement results.

Aziz

[Baresel]

Hello,
I have been experimenting with using my laptop that way.

Actually there is a software perfectly suitable for such a task

http://www.daqarta.com/

I did some field testing. The current from typical soundcard output seem to be just enough for such device with 6 meter array.
The software generates signal, and records it through microphone socket, measuring the voltage.

However, I cant get sensible results. I think it must be something with the signal generated.
I used the left channel with a sine wave at 1Hz. It seems to produce za current, but it seems to be in only one polarity. Im a bit confused, how the signals should look. Should they be off-phase by 180 degrees? I dont get it. I thought one sine wave was enough.

Could you please have a look at the software, and generate a proper signal for me? Then, you could save a config file nad send it to me, for testing.

[Aziz]

Hello Baresel,

the sound output offers only few volt output. It depends on the type of brand.
Stimulating with differential lines, you can double the output voltage.
Line1: sin(w*t)
Line2: sin(w*t+pi) = -sin(w*t), pi=180 degree phase shift

Differential output voltage: Line1-Line2 = sin(w*t)-(-sin(w*t)) = 2*sin(w*t),
where w=2*pi*f, f=frequency, t=time, pi=3.141592654

Most sound cards did not have a stereo microphone input (most mono only). As microphone input line has often additionally 20 dB gain (x10) option for increased gain, the input impedance might be still quite low for this application.

For a better and independent configuration, a simple stereo fet amplifier/buffer or even an instrumentation amplifier could give the possibility to measure high impedance and tiny signals. Probably some of the low noise microphone pre-amps could also be used.

As the signals/currents through the electrodes are ultra tiny on dry ground, you need to measure in differential mode only. Any EMI can be cancelled and will not disturb your measurement much.

I don't know the software, you have found. I only use self coded software. And the lock-in amplifier technique is really destined to this application. The time constant for the lock-in amplifier could be increased heavily (tens of seconds or even minutes), to resolve even nV signals. You can outperform most of the measurement devices then.

Aziz

[Aziz]

Hello friends,

if somebody wants to experiment with ground resistivity measurements using laptop and lock-in amplifier, the following basic schematics can be used. I have tested it and it works perfect for simple measurements. As I am using a high impedance op-amp (max. 10 pA input bias current, 10^12 Ohm input impedance), high ground resistances (like dry ground) can be measured.

The "difference amplifier" doesn't exist here and it is done in the software by simply doing "input signal = input L - input R" or vice versa. The lock-in amplifier can take the transmitted signal as a reference. For much higher ground resistances, an JFET instrumentation amplifier with high gain would be convenient. In this case, a single ended signal can be fed into the sound card input (one channel only).

Aziz

[Aziz]

Hi again,

I am not an expert on ground resistivity measurements. Some conventional measurement devices "sense" the current through the stimulation probes, which is necessary to calculate the ground resistance. There should be other way to solve this without sensing the ground current:

Just imagine only single side of the probes for clarification (right side here):
GND --[_R_]-- RXR --[_R_]-- TXR

We do not know the ground resistance R and the current I through the ground. This is a typical voltage divider circuit through the ground. The stimulated voltage is known. The measured voltage is also known. If we take a known resistor reference (RRef) and parallel this to GND and RXR connection, we can measure the voltage drop change. And this could be used to calculate the ground resistance R. We have:
GND --[RRef]-- RXR
GND --[_R_]-- RXR --[_R_]-- TXR

U1 measured without RRef,
U2 measured with paralleled RRef (RRef||R)
U2 should be lower than U1 then

It should be possible (provided that I am not wrong) to calculate the ground resistance R as this is a loaded voltage divider with known reference resistor RRef.

On differential configuration, the reference resistor RRef should be connected through RXL and RXR of course and make the calculation for differential mode.

There are many other measurement modes. Just google. It is very interesting.

Aziz