Ground Balance Theory

[UrbanFox]

Ferric, don't mind UF appearing here. Trolls do that. And nothing useful. Ever.

to you too!

It is because it's expired.

https://www.yousendit.com/download/U...MHdZY1M5TE1UQw

[mikebg]

SOLUTION IN TIME DOMAIN

Math operations in TIME DOMAIN for ground balance.

The block diagram given here explains in detail what information is required and what mathematical operations to be performed for ground balance. For explanation is used terminology of TIME DOMAIN. Analysis and explanation with terminology of FREQUENCY DOMAIN will be done separately.

COIL system
Four coils are showed in the block diagram:
1. TX coil used for excitation with time varying current i(t).
2. RX coil used to receive TGT signal, but it receives also AIR and GND signal.
3. AIR reference coil. It is used to obtain pure AIR signal. For example, this is only a turn wire on the TX coil.
4. GND reference coil. It is placed in position for induction balance with TX coil to obtain pure GND signal.
This example uses 3 RX coils to illustrate that the the ground can be balanced. In practice, a suitable software can simplify the system.

TARGET
The target is represented as a block or stage 5 with input and output. If we know the input signal x(t) and the output signal y(t)., we can identify the target by calculating its transfer function h(t). The name of this transfer function in time domain is IMPULSE CHARACTERISTIC (impulse response). For calculation of h(t) is used a complicated mathematical operation called DECONVOLUTION, because y(t) is convolution of x(t) with h(t).

AIR signal
This is the excitating signal that induces eddy currents. It is produced by mutual inductance or by self-inductance (at Monocoil search head) so its shape is like di/dt (derivative of TX coil current). A turn wire above TX coil can generate enough large AIR signal for refence

TGT signal
Since the target is buried, the energy passes through the earth twice to induce TGT signal in RX coil. Therefore, TGT signal contains distorted information for impulse response of target h(t). Excitating signal x(t) differs from AIR signal because it is the output of block 4. It is obtained by convolution of excitating signal in input of the block 4 with impulse characteristic h4.
Similarly is distorted the output signal from the target y(t). It undergoes a convolution with the impulse characteristic h6 of block 6.

GND signal
This signal contains very important information for the calculation of impulse characteristic of target h(t). Therefore, in block diagram is shown symbolically a GND reference coil.
It is assumed that the excitation signal x(t) of target has the same form as the GND signal. It is assumed also that the impulse characteristics h4, h6 and h7 are similar in shape.

Computational algorithm
1. By deconvolution of GND signal with the AIR signal is calculated the impulse characteristic h6.
2. By deconvolution of TGT signal with the impulse characteristic h6 is calculated the target output y (t).
3. By deconvolution of waveform y(t) with the GND signal is calculated the impulse characteristic of target h(t).
4. Follows comparison of the form h(t) with known forms.

This block diagram can not show what should be the function i(t) in order to obtain a better SNR and better energy efficiency. Analysis with FREQUENCY DOMAIN can do this.

[deemon]

Computational algorithm
1. By deconvolution of GND signal with the AIR signal is calculated the impulse characteristic h6.
2. By deconvolution of TGT signal with the impulse characteristic h6 is calculated the target output y (t).
3. By deconvolution of waveform y(t) with the GND signal is calculated the impulse characteristic of target h(t).
4. Follows comparison of the form h(t) with known forms.

This block diagram can not show what should be the function i(t) in order to obtain a better SNR and better energy efficiency. Analysis with FREQUENCY DOMAIN can do this.

It looks very clever , of course .... the only problem is where to get a direct ground signal

[Davor]

https://www.yousendit.com/download/U...MHdZY1M5TE1UQw

Yes. That particular one expired. Historically it was ML's IP, and now, by virtue of patent law it is not. See for yourself...

[Ferric Toes]

Thanks Qiaozhi,

I use Adobe Reader X, Version 10.1.6 on Windows XP and the second reference works, but not the first. I also took the suggestion of Aziz and tried to download the file, but couldn't even do that because the error message popped up first...

Oh well, ç'est la vie, as they say in Bulgaria...(Such is life.) I presume you've read the article and if in your opinion it does not contain any information that is conducive to the design of a simple ground balancing system that does not reject gold nuggets of any size, I really don't need to read the article.

(I know that such systems do exist, but they are complicated.)

Best regards,

Allan

P.S. I am also a student of Mandarin Chinese and I've noticed that the Swedish "ja" (yes) and the Chinese "hao" (good) have the same intonation.

Hi Allan,

I have different versions of Adobe Reader and on this computer it is Version 7 - both .pdf's open fine.

The knowledge in these and other papers is valuable to me for the following reasons. Firstly, there is a lot of conflicting information, particularly in detector patents, regarding the nature and cause of magnetic viscosity in soils and rocks. If you want to defeat the enemy (Magnetic Viscosity) then it is a good idea to find out as much as possible about its character, behaviour, and weaknesses. By far the greatest amount of published research is available from geophysics and the demining literature and while there are still a few errors and misconceptions, in general it is good information and free of smoke screens and mirrors.

Secondly, metal detecting is a scaled down version of geophysical em prospecting and as the electronic techniques used are the same, solutions should be similar. Much of the published information appears to have been missed as prior art by both applicants and examiners and I feel it is good to gather this information together in one place under "History" and "Theory", as is happening on this forum. All can then benefit by seeing what has been done before, and hopefully it will help to avoid future patent and IP conflicts.

Eric.

[mikebg]

It looks very clever , of course .... the only problem is where to get a direct ground signal

4. GND reference coil. It is placed in position for induction balance with TX coil to obtain pure GND signal.

[Prospector_Al]

Hi Allan,

I have different versions of Adobe Reader and on this computer it is Version 7 - both .pdf's open fine.

The knowledge in these and other papers is valuable to me for the following reasons. Firstly, there is a lot of conflicting information, particularly in detector patents, regarding the nature and cause of magnetic viscosity in soils and rocks. If you want to defeat the enemy (Magnetic Viscosity) then it is a good idea to find out as much as possible about its character, behaviour, and weaknesses. By far the greatest amount of published research is available from geophysics and the demining literature and while there are still a few errors and misconceptions, in general it is good information and free of smoke screens and mirrors.

Secondly, metal detecting is a scaled down version of geophysical em prospecting and as the electronic techniques used are the same, solutions should be similar. Much of the published information appears to have been missed as prior art by both applicants and examiners and I feel it is good to gather this information together in one place under "History" and "Theory", as is happening on this forum. All can then benefit by seeing what has been done before, and hopefully it will help to avoid future patent and IP conflicts.

Eric.

Hi Eric,

Aziz sent me a copy of the Buselli paper via e-mail and it opened! I don't know what's going on...

You are doing a very valuable service to those who are trying to advance the state of the art, and knowledge is the most effective tool one can use.

You are right about the fact that what's published by the geophysical community somehow falls between the cracks as far as the patent office is concerned. Many techniques are not patented, probably because they are not considered to be worthy of a patent by the inventor. And then, someone comes along and patents the idea!

How do you combat that? The examiners don't read any forums. If an old technique appears in an application, should one send a reference to to the examiner?

On a different subject: I have a "hot rock" I received from Oz that's pink in colour and weighs 57 grams. Its viscous signal is larger than several pound's worth of the local hot soil. Its signal decreases by about 50% when a rare earth magnet is brought next to it. The change occurs only when the external DC field is parallel with the field from the Tx coil. A tranverse DC field has little effect.

I also have a specimen of "Australian Banded Iron". This specimen is highly magnetic but doesn't exhibit the viscosity effect.

My explanation for the change in the viscous signal amplitued is that the external field tends to hold the domains in a fixed position, thus decreasing the effect of the Tx field from the coil.

If the stabilizing field were strong enough, one could eliminate the viscosity signal altogether. Unfortunately, it would require too much energy to do that, and the method would be effective only for superficial hot rocks.

All the best,

Allan

[deemon]

4. GND reference coil. It is placed in position for induction balance with TX coil to obtain pure GND signal.

OK , but how can you be sure that our ground is homogenous ? In another words - can you guarantee that the ground near your "special ground coil" is the same kind that the ground near your target ? Just imagine the situation when the pure ground is near the coil , but near the target we have a kind of mineral with ferromagnetic properties .... your ground balance won't work in such condition . So , as I think , your theory is beautiful , but useless in the real world

[Ferric Toes]

Hi Eric,

On a different subject: I have a "hot rock" I received from Oz that's pink in colour and weighs 57 grams. Its viscous signal is larger than several pound's worth of the local hot soil. Its signal decreases by about 50% when a rare earth magnet is brought next to it. The change occurs only when the external DC field is parallel with the field from the Tx coil. A tranverse DC field has little effect.

I also have a specimen of "Australian Banded Iron". This specimen is highly magnetic but doesn't exhibit the viscosity effect.

My explanation for the change in the viscous signal amplitued is that the external field tends to hold the domains in a fixed position, thus decreasing the effect of the Tx field from the coil.

If the stabilizing field were strong enough, one could eliminate the viscosity signal altogether. Unfortunately, it would require too much energy to do that, and the method would be effective only for superficial hot rocks.

All the best,

Allan

Well, guess what. I have an OZ hot rock with very high susceptibility and viscosity and its viscosity signal increases considerably when you bring a ferrite magnet up to it. The magnet itself gives no signal; I checked that. The same magnet causes the viscous signal to disappear on a sample of Virginia Red Hill soil. These tests were done on 10gm samples of each in my Viscosity Meter, so very well controlled. On the RH soil the VM reading returns to its original one immediately the magnet is removed. The Oz ironstone takes about 15 mins. I will do some further tests as those opposing results are interesting.

Another thing of interest for those who strive for short delay times is that the viscous signal amplitude is 2.2x as great at 5uS as it is at 10uS

Eric.

[mikebg]

OK , but how can you be sure that our ground is homogenous ? In another words - can you guarantee that the ground near your "special ground coil" is the same kind that the ground near your target ? Just imagine the situation when the pure ground is near the coil , but near the target we have a kind of mineral with ferromagnetic properties .... your ground balance won't work in such condition . So , as I think , your theory is beautiful , but useless in the real world

Dmitry, you are right. The method is sensitive to change in ground properties, but the method is useful.
I designed a metal detector used as locator for "jewelry meteorites". They can not generate signal at air test because they have no both conductivity and permeability. However when they are in the bad soil, they create local "discontinuity" in ground properties and the instrument responses. The instrument also locates caves and voids in bad ground. However when operates with ground balance, the instrument can detect and identify deep tagrets in bad soil. Conventional metal detectors can not make this because the TGT signal contains distorted information for target characteristic.

In reality, I reinvented the bicycle and tricycle in PI technology :-)
http://www.geotech1.com/forums/showt...704#post164704
http://www.geotech1.com/forums/showt...826#post155826

The genuine inventor of principle is D. E. Hughes.

Other reinventors use term "TWIN LOOP" for the PI "bicycle".
Please try the TWIN LOOP project with increased distance between loops:

http://www.geotech1.com/forums/showt...336#post128336
http://www.geotech1.com/forums/showt...ight=twin+loop

You will see that the PI "bicycle" can suppress the bad GND signal and interferences.

[moodz]

... you guys are painting a funny picture about OZ detecting .... anyone would think that the joint over here is made out of hot rock ... well guess what ... it isnt and you generally have to go out of your way to find hot rocks. If you have an AGB system it is not an issue anyway.

[Prospector_Al]

... you guys are painting a funny picture about OZ detecting .... anyone would think that the joint over here is made out of hot rock ... well guess what ... it isnt and you generally have to go out of your way to find hot rocks. If you have an AGB system it is not an issue anyway.

Hi moodz,

There's no attempt here to make Oz look bad. It's an attempt to establish some important facts about ground signals. There's a lot of misinformation floating around--part ignorance and part deliberate deception. A detector's ability to distinguish between ground signals and a desirable target is still the limiting factor for a metal detector.

The information that can be gleaned from Eric's last post will probably not interest everybody, but it's of vital importance to me. He replicated a phenomenon which I also have observed and it has immence implications for the detector technology: It's now clear that the vicous signal from hot soil and hot rocks has a relaxation time that can vary from microseconds to thousands of years. That means that ground signals can mimic a gold nugget of any size and shape.

Those looking for an explanation for the "electronic holes" in a detector's response need not look any further. Algebraic combinations of multiple signal samples can not entirely eliminate the holes--perhaps make them disappear for nuggets of certain sizes and shapes, but considering the great variation of these factors, it's apparent that the current technology has gone up an evolutionary blind alley. All those patents filed in vain! (Of course, I include myself in this futile effort.)

Several prospectors in Oz have kindly sent me samples, for which I'm grateful. The big nuggets just beyond the reach of the current crop of PI detectors are now getting nervous...
By pooling knowledge on forums such as this one, the art will evolve and everybody will be happy...

Hot rocks are just rocks that have not eroded into hot soil, yet, and soil is everywhere.

All the best,

Allan
Just my personal opinion.

[sawmill]

Allan

You don't reckon that some of the falsing,coil overloading,and just plain false
signals can be attributed to highly conductive metallic mineral salts as a binder for magnetite
do you? I dug 4 large holes in an area that gave a perfect target signal. No target
was present in any hole. The dirt from the holes would not give a peep on the detector
either. Once the soil was disturbed to a certain depth the signals went away.

There was no hot rocks or charcoal either. A few feet away a detector will over load until
you stir up the soil.

[Midas]

Well, guess what. I have an OZ hot rock with very high susceptibility and viscosity and its viscosity signal increases considerably when you bring a ferrite magnet up to it. The magnet itself gives no signal; I checked that. The same magnet causes the viscous signal to disappear on a sample of Virginia Red Hill soil. These tests were done on 10gm samples of each in my Viscosity Meter, so very well controlled. On the RH soil the VM reading returns to its original one immediately the magnet is removed. The Oz ironstone takes about 15 mins. I will do some further tests as those opposing results are interesting.

Another thing of interest for those who strive for short delay times is that the viscous signal amplitude is 2.2x as great at 5uS as it is at 10uS

Eric.

Interesting. So.. perhaps a center-tap coil one half of which you drive in the usual fashion, the other can have DC current periodically switched in (perhaps automatically, perhaps under operator control) to test whether a signal is affected and hence likely to be viscocity related. Sounds highly patentable, hopefully this discussion prevents that from happening.

Midas

[dougAEGPF]

Allan

You don't reckon that some of the falsing,coil overloading,and just plain false
signals can be attributed to highly conductive metallic mineral salts as a binder for magnetite
do you? I dug 4 large holes in an area that gave a perfect target signal. No target
was present in any hole. The dirt from the holes would not give a peep on the detector
either. Once the soil was disturbed to a certain depth the signals went away.

There was no hot rocks or charcoal either. A few feet away a detector will over load until
you stir up the soil.

If the ground is conductive then a salt coil or the salt timings on a ML should reduce or eliminate the problem. Similarly increasing the sample delay on a TDI should reduce or eliminate problems due to conductive ground.If this is not the case then perhaps one might speculate that what you observe is due to earth currents? (telluric currents). There are places in Australia where this occurs due to salty soils,conductive clays and faulted bedrock particularly near some SWER power lines where earth return currents can sometimes flow very close to the surface.
dougAEGPF