Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Electroseismic "thumping" for low permeability in subsoils

Status
Not open for further replies.

fastline

Member
We are getting ready to drill a water well in an unfounded area. The question comes in "where do we drill?" Obviously the driller will have no idea. The ways of witchcraft don't appeal to me as a a man of science.

I stumbled into the current industry use of electroseismic testing methods which, though probably not absolute, may give an idea of the best prospective areas based on subsoil responses.

I probably need some help with my understanding as well as possibly coming up with a stupid simple way to try this for myself. I will explain the concept as I understand it.

A simple dipole antenna is created at the ground surface by driving two conductive rods in the ground at a certain distance apart. A seismic generator is used by way of a hammer/plate or large weight to generate a P wave proximal to one of the electrodes in the ground. Low permeability zones may contain sand or gravel and allow water to move within it. As I understand it, when the P wave strikes an area containing water, a dipole moment is created in the water which would return to the surface at the speed of light as a charge that would be detected by the dipole antenna.

Possibly the amplitude and timing of this return could indicate both the abundance and depth to water.

Now, I realize this is really simplified and may require some filtering or intense interpretation of the wave forms but this process is being used commercially right now. I don't need a fancy 3D image or software to tell me the depth if I can simply compare my results in time with a scope. What I intend is to mount a piezo generator on a large hammer handle which would act as the trigger for the event. This would function as the point in time the P wave was generated, then compare this with any returned waves to the dipole antenna. One of the variables that must be estimated is the speed at which the P wave travels in the soil. Each formation would be a bit different but somewhere between 2,000-10,000ft/sec.

I have areas I can test this device with known variables such as abundance of water, and depth to water as well as the geology between.

I am curious if you guys think we can employ this technique with a simple scope with 1GZ/sec sample rate and set to trap about .5sec of data?

I do know this. I WILL hit water regardless of where I drill. It comes down to finding a highly permeable zone that will have free flowing water. I know the water will be found between 20-50ft if we get a good vein but other sources may be down to 100ft. Traditionally no well around here is continued past 120ft. We are not shooting for an aquifer, only a water table that bears the right formation for production on a domestic well.
 
Hy fastline,

Care to tell use where you are and show your location in your user window on the left of your posts. This helps us to answer question as it indicates what your mains voltage is and what components you can access.

We get some unusual and intriguing questions on ETO, but I would guess that yours takes top prize. :cool:

Firstly, let me say that I have no experience of seismology but I do know the principals and have worked on sonar.

In my opinion your objectives are achievable but here are some initial thoughts:

(1) Stimulation

I do not think that a piezoelectric transducer would be able to launch enough power at an appropriate frequency into the ground to get a reasonable amplitude return.

(2) Detection

I suspect that a standard microphone and amplifier could be adapted to realise a low frequency response to detect the returned waves.

(3) Display

The frequency response of the oscilloscope that you describe would be adequate.

(4) Interpretation

With just a single trace on the scope, the returned signal would be difficult to interpret in comparison to a 3D map.

I know you haven't asked this, but my advice would be to hire a well proven water diviner. Like you I do not believe in any mambo-jumbo but they do seem to find water. :wideyed:

spec
 
Thanks! To clarify, the piezo transducer is not to generate the P wave, only to provide a trigger signal to the scope to trap and record the data. I need to know precisely when the wave occurs to compare in time to the antenna responses. In talking with my colleague, I think we can simply set this near the impact zone and set a weight on it.

For the P wave generation, we intend to test with a large hammer/steel plate but not opposed to stepping to a buffalo pyro gun if needed. I am hoping at the depths being tested, we can get ample data the easy way.

My general theory of interpretation of data after considerable studying is the higher the return pulse activity, amplitude, and sharpness of rise time will dictate more presence of water than returns of minimal wave activity. I obviously cannot predict what each wave might mean in terms of GPM, size of perm zone, etc, but a general understanding might need to be applied that we "drill where we get the best responses and feedback"

I have high confidence that a competent hydrologist can simply look at the responses and know where to go. The pretty 3D pictures just help in looking at it. This is similar the DAQ, pretty dials and displays are nice but the raw data can do just the same.
 
Thanks! To clarify, the piezo transducer is not to generate the P wave, only to provide a trigger signal to the scope to trap and record the data. I need to know precisely when the wave occurs to compare in time to the antenna responses. In talking with my colleague, I think we can simply set this near the impact zone and set a weight on it.

For the P wave generation, we intend to test with a large hammer/steel plate but not opposed to stepping to a buffalo pyro gun if needed. I am hoping at the depths being tested, we can get ample data the easy way.

My general theory of interpretation of data after considerable studying is the higher the return pulse activity, amplitude, and sharpness of rise time will dictate more presence of water than returns of minimal wave activity. I obviously cannot predict what each wave might mean in terms of GPM, size of perm zone, etc, but a general understanding might need to be applied that we "drill where we get the best responses and feedback"

I have high confidence that a competent hydrologist can simply look at the responses and know where to go. The pretty 3D pictures just help in looking at it. This is similar the DAQ, pretty dials and displays are nice but the raw data can do just the same.

Hy fastline,

Apologies, it was late and I completely misread your post, especially the bit about the hammer. :banghead:

spec
 
Last edited:
Just some general info from Wiki about seismology.

spec

Body waves
There are two types of body waves, Pressure waves or Primary waves (P-waves) and Shear or Secondary waves (S-waves). P-waves, are longitudinal waves that involve compression and expansion in the direction that the wave is moving. P-waves are the fastest waves in solids and are therefore the first waves to appear on a seismogram. S-waves are transverse waves that move perpendicular to the line of propagation. S-waves are slower than P-waves. Therefore, they appear later than P-waves on a seismogram. Fluids cannot support perpendicular motion, so S-waves only travel in solids. [8]

Seismogram.gif

Seismogram records showing the three components of ground motion. The red line marks the first arrival of P-waves; the green line, the later arrival of S-waves.
Surface waves
The two main surface wave types are Rayleigh waves, which have some compressional motion, and Love waves, which do not. Rayleigh waves result from the interaction of vertically polarized P- and S-waves that satisfy the boundary conditions on the surface. Love waves can exist in the presence of a subsurface layer, and are only formed by horizontally polarized S-waves. Surface waves travel more slowly than P-waves and S-waves; however, because they are guided by the Earth's surface and their energy is thus trapped near the surface, they can be much stronger than body waves, and can be the largest signals on earthquake seismograms. Surface waves are strongly excited when their source is close to the surface, as in a shallow earthquake or a near surface explosion.[8]

https://en.wikipedia.org/wiki/Seismology#Body_waves
 
Last edited:
Hy again fastline,

Can you confirm that you are proposing to drive two electrically conductive rods into the ground, then vibrate the ground and at the point that the vibration starts the scope will be triggered. You will not be monitoring vibrations in the ground but instead the voltage between the two rods. Can you also confirm that you will not be using microphones to measure the mechanical vibrations in the ground.

Assuming this is correct, can you indicate the likely amplitude of the voltage between the two conductive rods. I suspect that the voltage will be very small and that a low noise amplifier will be required to amplify the returned voltage to a level suitable for the scope.

spec
 
I suspect that the voltage will be very small
I suspect that the voltage will have a DC bias from galvanic effects and the wanted signal will be buried under a lot of electrical noise, so good filtering will be needed in addition to the low-noise amplification.
 
I suspect that the voltage will have a DC bias from galvanic effects and the wanted signal will be buried under a lot of electrical noise, so good filtering will be needed in addition to the low-noise amplification.
Yes, good point Alec.

Careful choice of the pole surface material would be important: gold or carbon perhaps.

I wonder what the output impedance of the voltage between the two poles would be.

spec
 
I wonder what the output impedance of the voltage between the two poles would be.
Highly dependent on near-surface soil composition and condition.
 
Highly dependent on near-surface soil composition and condition.
I suppose a very high input impedance amplifier would cater for the varying input impedance. Also, a low frequency cut off at around 1Hz? would cater for the DC galvanic effects.

I wonder if the the amp would need to be gated to prevent saturation during the initial 'thump'. Or perhaps limiting will be sufficient to ensure an acceptable recovery.

spec
 
Not knowing how the responses might look, I planned to just connect the scope directly to the electrodes and see what happens first. I might need a little help in filtering. I know the voltage response will be low but hoping due to the distances to water, the response will be strong enough.

Yes, I am indeed looking for a voltage response through the principles of dipole moment in water. I planned to use copper grounding rod for the electrode for now. Some are just using steel.
 
Not knowing how the responses might look, I planned to just connect the scope directly to the electrodes and see what happens first. I might need a little help in filtering. I know the voltage response will be low but hoping due to the distances to water, the response will be strong enough.

Yes, I am indeed looking for a voltage response through the principles of dipole moment in water. I planned to use copper grounding rod for the electrode for now. Some are just using steel.

If you set the scope to AC input (rather than DC) and if the scope is sensitive enough you might get a readout, but it is difficult to be specific without knowing the likely voltage between the rods.

Can you give the planned dimensions of the rods, how deep they will be driven into the ground, and how far apart the rods will be.

This has practical implications because the connecting leads will need to be screened from noise and interference.

spec
 
Well, here is where I am. I got a piezo transducer working as a trigger but it appears, as was suspected, noise and response will be tricky. I need a little help understanding a few things here. The noise you see on the yellow CH2 is 60hz but not sinusoidal. I need to look at the scope more but I noticed my setting of 5mv/div showing about 19mv pk-pk but the Vp measurement below is showing 199mv. Not sure on that one just yet.

Any thoughts or ideas how to proceed?

I used 1/2" Copper coated grounding rod for this test. About 2ft long and 4ft apart. I am unable to get a resistance value because I have effectively created a capacitor so my meter will start at about 30ohms and climb on up to 3000 or so. I did not leave it to see if it would stabilize.
 

Attachments

  • 20160805_113527.jpg
    20160805_113527.jpg
    519.2 KB · Views: 142
Last edited:
What are you using as a seismic source? One industrial type consists of a motor-driven rotating eccentric weight (think cell-phone vibrator on a giant scale :) ). This has the advantage that you have a well-defined input frequency (or frequency sweep), making it easier to filter out interference from the receiver signal by using a very narrow-band filter or correlation techniques.
Perhaps one of these is a tad OTT for your project :D.
 
Last edited:
Well, here is where I am. I got a piezo transducer working as a trigger but it appears, as was suspected, noise and response will be tricky. I need a little help understanding a few things here. The noise you see on the yellow CH2 is 60hz but not sinusoidal. I need to look at the scope more but I noticed my setting of 5mv/div showing about 19mv pk-pk but the Vp measurement below is showing 199mv. Not sure on that one just yet.

Any thoughts or ideas how to proceed?

I used 1/2" Copper coated grounding rod for this test. About 2ft long and 4ft apart. I am unable to get a resistance value because I have effectively created a capacitor so my meter will start at about 30ohms and climb on up to 3000 or so. I did not leave it to see if it would stabilize.

Hy fastline,

The source impedance of any signal voltage between the two rods cannot be measured by a meter as you tried but thanks for trying anyway.

As we suspected it looks like noise will be a problem.

How are you connecting the rods to the scope?

I think you will need a battery powered differential amplifier (instrumentation amplifier) placed physically between the rods so that the connecting cables are as short as possible and equal lengths.

The cables from the rods to the differential amplifier need to be screened cables (coaxial cable) with the screens connected to the scope earthy point. But the cable screens should be left unconnected at the rods.

spec
 
Last edited:
I do not currently have an amplifier but looking for some recommendations on that as I need one for other things. We have used an audio amp in a pinch in the past.

By connecting things now, my only goal here was to review the noise levels knowing that it will be 10-100x worse with an amp. I need to get real on signal conditioning apparently.

Right now it is only connected with a pair of automotive jumper cables and back to the scope leads. Total length is about 15ft. Obviously my leads are not grounded if "earthy point" and "screens" are how you describe grounds.

I am sure I could reduce some of the noise but it is very obvious at 60hz. I made a quick Low pass RC filter to see if I could improve things. I used a 500ohm resistor and 1uf cap so I think that is about a 300hz filter. It cleaned things up a bit and reduced the stray voltage but still way beyond the signals I will need to see.


Alec - I was thinking the same thing on the seismic source that I continuous low freq device could sure help home in on things and filter everything else out. Right now I am using a hammer/plate method but research has proved it can work if you work the scope signals right.
 
You certainly need a good differential amplifier, but I don't think a simple low-pass filter is going to help you much. What I'd suggest is a notch filter, such as a bootstrapped twin-Tee, tuned to reject 60 Hz. From the looks of your scope waveform, it might be necessary to have notch filters to get rid of the 3rd and 5th harmonics, as well.
 
I was thinking the same thing on the seismic source that I continuous low freq device could sure help home in on things and filter everything else out.
Low frequencies are better at penetrating ground than high frequencies.
If you used a vibration source of, say, 40Hz and a sharply tuned receiver, then the 60Hz noise and its harmonics might be less troublesome.
You will definitely need screened cables for your receiving electrodes.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top