Ground Balance Theory

[Prospector_Al]

Hi Greg,

I oversimplified: If the dynamic range of the ground balancing system is inadequate, any hot soil can make a detector inoperative.

But all PI detectors are not the same--look at all the ideas floating around on this Forum!

So far, your soil is the hottest U.S. sample that I have. I put some in a big tray that fits over the coil. I'll have some news in just a few days.

What prompted my comment is the fact that there are so many analyses of soil samples and so few solutions. It seems that the soil analysis business is an end in itself--like "art for art's sake"...

Allan

[sawmill]

Hi Allan

I already did tests with an actual detector,and that magnetite does not effect the
machine enough to even mention. I can detect a penny under a plastic pan of packed
magnetite 3/4 by 6 inches at 11 inches ,from where you got your samples. The TDI will
balance out that same pan at 4 on the GB, full gain,10 Us delay,and all metal.

Spread the magnetite out on a larger surface at 1/8 to a 1/4 deep and it balances out
at 2 GB . Mix it in with a little dirt and the detector can't see it even with no GB.
The sample weighed one pound and eleven ounces. I also have a chunk of magnetite
that weighs 1 1/2 ounces, The TDI does not even notice it with no GB.

You are correct about one thing though. Someone needs to actually do a spectrograph,
assay,and chemical tests when playing with hot soils. You can't just separate what you
think is obvious and take a one eyed test.

[Davor]

Here is a report with quite a bit of information on Croatian soils.

Nice catch, I see Mr. Das took part in it as well. I found several errors regarding toponyms there, but it would not change the results. Rogoznica is more in the south, but most of the other places are in a radius of ~30km from Obrovac.

What I think is overseen in these reports is that they are missing the information on homogeneity of soils. Perhaps the soil samples are susceptible this or that much, but my point is that the very soil is a worse problem if it is mixed with rocks of different susceptibility. With rocks commensurable in size with the coil, your ground balance should be trained to reject both, and not just the mixture of them. Hence uncooperative soil.

Your susceptibility meter takes small samples, so you could measure separately the matrix and the rocks in it. A parameter you can't measure is the contrast between the two, because the size distribution is at play.

I guess the OZ soils are a bit opposite than the terra rossa in sense that there you have mild to severe soil matrix with exceptionally hot rocks here and there. In terra rossa you either have thick layers of terra rossa soil, like in Istria peninsula or throughout Italy, or you have lots of rocks with streaks and pockets of terra rossa between them, like in Obrovac, Benkovac. etc.

I guess it is much easier to train your ground balance on a homogeneous soil - there is not much contrast in it.

[Ferric Toes]

Hi Greg,

I oversimplified: If the dynamic range of the ground balancing system is inadequate, any hot soil can make a detector inoperative.

But all PI detectors are not the same--look at all the ideas floating around on this Forum!

So far, your soil is the hottest U.S. sample that I have. I put some in a big tray that fits over the coil. I'll have some news in just a few days.

What prompted my comment is the fact that there are so many analyses of soil samples and so few solutions. It seems that the soil analysis business is an end in itself--like "art for art's sake"...

Allan

Hi Allan,

I fully agree that we know where the hottest viscous ground is, and if we design for that, then everything else will be a piece of cake - almost. Also your last point is very true. Part of the reason for so many analyses is that a lot of funding was available in EU and USA for magnetic soil investigation, so all sorts of organisations, university departments and defense organisations jumped on the gravy train. I reckon I have more papers on this subject than any other aspect of metal detection, and there must be as many again that I have not seen.

The demining fraternity were only concerned about countries where mines needed to be cleared, and therefore looked at those soils only: Croatia, Bosnia, Mozambique, Cambodia, Laos, etc. They categorised those soils from neutral to very severe, not knowing that ultra severe soil existed elsewhere. In 2008, when I attended the Soil Magnetism Workshop at Cranfield University UK and showed Yogadish Das the Australian material, he was very surprised that such a level of ground mineralisation existed. Other top researchers who were there and many gave the same reaction - Das, Tabbagh, Preetz, Dabas, Hannam, Dearing, Thiesson, Pasion, Druyts, Billings: all were there, so it was a top level three day international session. https://www.msu.edu/~rvd/soilmag08/p...nday-Talks.pdf
Another large publication is .http://www.gichd.org/fileadmin/pdf/L...r_handbook.pdf Both have some interesting information, but the question still remains: why isn't the GB problem solved already?

I think it is to a large degree. The decay law is well known, although I think that some previous measurements threw up errors because the hardware was not entirely suitable. What was thought to be bad ground in the USA was often areas that VLF detectors were unworkable and later measurement found them quite weak in viscosity but high in susceptibility (black sand). Provided the grain size is above the single domain limit, then a PI will not respond to black sand or magnetite. There are bound to be a small % of grains that are below; hence a weak viscous signal in some cases. Sawmill states that the TDI works fine in these conditions, and so it should because it was tested in the "worst place on earth" i.e. Oz. I took some units over about three years ago for final testing and some minor adjustments were necessary to fully cope with the worst viscous mineralisation. There is just no way of simulating it, short of shipping a couple of tonnes back home. Having done those Oz mods. Whites can be confident that the TDI Pro will work anywhere in the world on magnetically mineralised ground.

Improvements can still be made of course. The spread in the decay law needs to be determined i.e. the +/- spread of the exponent 1/t^x. I hope to be able to do this shortly with a higher accuracy viscosity meter. Then, a better method of analysing the RX signal making use of all the information. I'll be posting on this aspect shortly as I may need some help in the DSP area.

The best soil I have so far measured in the USA for high viscosity is from the aptly named Red Hill in Virginia. LF-HF(669-552) = 117 on the Bartington MS2B

For comparison:- Sawmill's sample LF-HF (2243-2227) = 16. High susceptibility Magnetite and low viscosity of 16. All freshly measured this morning.

Just to remind you that you need to get to Oz. Victorian ironstone LF-HF (30178-28994) = 1184. Ouch!

Eric.

[mikebg]

What means "GROUND BALANCE"?

"If you can speak of technical things only in technical terms, you do not understand them."

Albert Einstein

Believe me, I searched the WEB because I can not understand the technical term "ground balance" .
In the next my posts, I will show the search result - how producers and sellers of metal detectors describe the term.

Since I know the nontechnical terms for this problem, I tried to find the math terms "convolution" and "deconvolution" of target signal in all mentioned here scientific papers and all posts in this forum.
Unfortunely the search results are frustrating.
Google showed that I use such terms and I published the solution in Time domain and in Frequency domain. Below is attached my block diagram.

Aziz uses similar solution, but he keeps this as a secret because he did not want GREEDLAB Pty Ltd to patent the method of deconvolution and CHINELAB Unlimited to use the method. However the solution was published, which makes patent pendings useless and CHINELAB Ultd will use the method.

[mikebg]

CONVOLUTION AND DECONVOLUTION

I mentioned for convolution in 04-07-2008:
http://www.geotech1.com/forums/showt...0563#post70563
"The complex plane displays things in periodic (frequency) domain. This domain is more obvious than time domain, therefore it is very handy for analysis and calculation because convolution is substituted by division."

The description of above block diagram was posted in 07-21-2011

Description in time domain:

A time varying signal TX(t) excites with magnetic field the ground with target. We can make identification of target (block 2) if we know its impulse response h2(t). To calculate h2(t), we should know the input signal x(t) and the output signal y(t) of block 2 because we know that
y(t)=x(t)*h2(t)
where the sign * (star) means convolution (свёртка) - an integral math operation (search the web for details). For calculation of h2(t) we should make deconvolution (обратная свёртка) - a complicated math operation.
At air test we know x(t) and y(t). The air test is specific case of measurement when h1(t)=1 and h3(t)=1.
The block diagram shows the common case (when the target is buried). Then the output of block 1 (input ground) excites the target (block 2) and output of target (signal y(t)) excites the block 3 (output ground). As result we receive a very convolved TGT signal from block 3. If we know only the TGT signal and TX signal, we can not calculate the deconvolution of h2(t). We need the convoluting functions.

Analysis using frequency domain

A TX current containing suitable frequency spectrum excites with magnetic field the ground with target. We can make identification of target (block 2) if we know its frequency response H2(f). To calculate H2(f), we should know the input signal X(f) and the output signal Y(f) of block 2. We know that output signal of target is
Y(f)=X(f).H2(f)
where the sign . (point) means multiplication. For calculation of H2(f) we should make division of two functions - a simple math operation:
H2(f)=Y(f)/X(f)
We know X(f) and Y(f) when we make air test. The air test is specific case of measurement when H1(f)=1 and H3(f)=1.
The block diagram shows the common case (when the target is buried). Then the output of block 1 excites the target (block 2) and output of target excites the block 3. As result we receive from block 3 a TGT(f) signal which spectrum is multiplication of several frequency responses. If we know the shape of frequency responses H1(f) and H3(f), we can calculate the shape of H2(f).
To know the shape of H1(f) and H3(f), we should receive a separate GND signal. A method for making this is illustraded in the lower left corner of the figure. A reference coil REF is positioned in induction balance to TX coil. Balance means AIR signal is zero, ie it receives only GND signal.
However if the ground has conductivity, H1(f) and H3(f) have different shape. We should make measurements to see if H1(f) and H3(f) differs in shape.

[mikebg]

DEFINITIONS for "GROUND BALANCE"

I made two definitions for "ground balance theory":

1."When search head moves, the GND signal changes. Metal detector senses the change and responses to it. Ground balance is technology to avoid the response. We need a theory that shows how to make this."

2. "When a target is buried, TGT signal represents twice convolution of targets impulse characteristic with grounds impulse characteristic. Ground balance is deconvoluting technology to obtain pure information for impulse characteristic of target."

NOTE: In East Europe is used the term "characteristic" instead the English term "response" because this is an own trasfer function of the system. For example, the gain of an amplifier is its own characteristic; it is not a response. Term "frequency response" is misleading because it relates to spectral characteristic of a system, but not how the system responses to a given frequency.

Let we see other definitions for "ground balance":

According WHITE'S
"Ground Balance - A feature that can be adjusted to ignore the masking effect ground minerals have over metal targets." I like this definition because it is something like my definition #2.

According FISHER lab:
" Every kind of soil contains minerals. In soils with high mineral concentrations, mineral signals can be tens to hundreds of times stronger than signals from buried objects. Iron minerals and mineral salts are two of the most common causes of interference.
Ground balancing allows your metal detector to ignore signals from minerals in the soil and concentrate on signals from buried metal objects. Your metal detector can be ground balanced to a particular type of soil, giving you quieter operation, more accurate target identification, and deeper target detection."

According MINELAB:
"Ground Balance is a variable setting that increases detection depth in mineralised ground. This ground may contain salts, such as in wet beach sand or fine iron particles, such as in red earth. These minerals respond to a detector’s transmit field in a similar way that a target does. Due to the much larger mass of the ground compared to a buried target, the effect of mineralisation can easily mask small targets. To correct this the Ground Balance setting removes the responding ground signals, so you clearly hear target signals and are not distracted by ground noise."

East Texas Metal Detectors:
"The Ground Balance circuity electronically samples the reciever at a location where mineralized ground doesn't produce a response. Metallic objects, however, will produce a strong response at this same setting. Since the masking ground static is cancelled and the depth of penetration is increased (due to the low operating frequency) the detector can be used for deep objects even though the soil may be heavily mineralized."

According
http://www.ehow.com/how_6068361_grou...-detector.html
"Depending on the mineral composition of the ground where you're treasure hunting, you might have to ground balance your metal detector or risk missing targets because of mineral interference. Some metal detectors come with a preset ground balance setting that cannot be changed. The setting on these detectors is intended to work across a wide variety of soil types. Other metal detectors have a feature that allows them to automatically ground balance with no user input. If your detector falls into the third category, you have to manually ground balance it."

Kellycodetectors
http://www.kellycodetectors.com/library/faq.php
"Ground balance is a form of discrimination that cancels out the effect of mineralization. Ground balancing is the physical act of finding the balance point where the effects of the ground are neither too positive nor too negative. When a detector is set with a positive ground balance, it will react to the mineralization matrix just like a target. When this happens, you will get an audio signal and targets in the ground will be masked by the mineralization. If a detector has been set up with a negative ground balance, the detector is discriminating out the ground and will go silent. A severe loss of depth and sensitivity are the results."

Garrett
http://www.garrett.com/hobbysite/hbby_terms.aspx
"Ground Balance - An adjustment made to "cancel" or ignore ground mineralization; may be done manually (See GroundTracking) or automatically.
Ground Tracking - The ability of a metal detector to continuously measure the ground’s mineralization and automatically adjust the detector’s ground balance setting for optimum performance."

Tinkerer
http://www.geotech1.com/forums/showt...GROUND-BALANCE
"A very important factor for deeper detection depth is the ground balance.
An ideal ground balance would mean that the detector only sees the targets.
It would not see the minerals in the ground. This means that one could lift the coil off the ground 6 inches or put the coil right on the ground and the output would stay 0. And this should work every where, even in the worst mineralized soils."
However the bad ground makes twice phase shifts of targets phase characteristic. How the synchronous demodulator will know that this is the genuine phase of target?

Tesoro
http://www.tesoro.com/info/faq/glossary/
"Ground Balance: A state of operation using specialized circuitry to ignore the masking effect that iron ground minerals have over metal targets.
Ground Balance - Factory Preset: A feature which eliminates the manual ground balance control and its adjustment from the operator's setup procedure. This adjustment is performed internally by the factory to optimize operation over an average range of nonconductive soils.
Ground Balance - Manual Adjusted: A feature requiring a manual control adjustment procedure to neutralize the effects of negative minerals in the search matrix.
Ground Balance - Self Adjusting: A feature which senses change in ground mineral content and continuously readjusts the ground balance while in operation. Sometimes called Ground Tracking or Automatic Ground Balance."

[Ferric Toes]

"If you can speak of technical things only in technical terms, you do not understand them."

Albert Einstein

Believe me, I searched the WEB because I can not understand the technical term "ground balance" .
In the next my posts, I will show the search result - how producers and sellers of metal detectors describe the term.

Since I know the nontechnical terms for this problem, I tried to find the math terms "convolution" and "deconvolution" of target signal in all mentioned here scientific papers and all posts in this forum.
Unfortunely the search results are frustrating.
Google showed that I use such terms and I published the solution in Time domain and in Frequency domain. Below is attached my block diagram.

Aziz uses similar solution, but he keeps this as a secret because he did not want GREEDLAB Pty Ltd to patent the method of deconvolution and CHINELAB Unlimited to use the method. However the solution was published, which makes patent pendings useless and CHINELAB Ultd will use the method.

I expect Carl will remember that he and I got berated on a couple of forums two or three years back, by a certain JH, for using "amateur, non-technical, and meaningless terms" such as GROUND BALANCE, EDDY CURRENT and MAGNETIC VISCOSITY. Are you suggesting the opposite; that they are too technical?

Eric.

[Ferric Toes]

I mentioned in an earlier post, there is almost too much material on the electromagnetic properties of soil, so I am now going to be a bit selective. The 1982 Buscelli paper is important as the investigation is based on results achieved in Australia using the SIROTEM equipment, and under "time domain measurements" concludes that "this t^-1 time dependence indicates the presence of magnetic viscosity".

The second paper by Dabas and Skinner describes further investigation of viscous remanent magnetisation for which I designed and built the PIM (Pulse Induction Meter) equipment. The non ideal TX pulse shape appears to have resulted in errors in the decay law (-1.4) instead of the later (-1.0 +-0.1) results with a more rectangular pulse shape. Errors of this order are noticeable in a ground balancing system and contribute to ground noise if different TX widths and shapes are used.

Eric.

Buselli_1982.pdf

Dabbas_Skinner_1993.pdf

[Prospector_Al]

I mentioned in an earlier post, there is almost too much material on the electromagnetic properties of soil, so I am now going to be a bit selective. The 1982 Buscelli paper is important as the investigation is based on results achieved in Australia using the SIROTEM equipment, and under "time domain measurements" concludes that "this t^-1 time dependence indicates the presence of magnetic viscosity".

The second paper by Dabas and Skinner describes further investigation of viscous remanent magnetisation for which I designed and built the PIM (Pulse Induction Meter) equipment. The non ideal TX pulse shape appears to have resulted in errors in the decay law (-1.4) instead of the later (-1.0 +-0.1) results with a more rectangular pulse shape. Errors of this order are noticeable in a ground balancing system and contribute to ground noise if different TX widths and shapes are used.

Eric.

[ATTACH]23778[/ATTACH]

[ATTACH]23777[/ATTACH]

Hi Eric,

When I tried to open the Buselli file I get an error message: "The file is damaged..." and it won't open. Does anyone else have this problem or is my Adobe reader at fault?

Allan

[Aziz]

Hi Eric,

When I tried to open the Buselli file I get an error message: "The file is damaged..." and it won't open. Does anyone else have this problem or is my Adobe reader at fault?

Allan

Hi Allan,

both PDF files being ok. "Save as" file into your harddisk and open them via file manager. Should work then.
Cheers,
Aziz

[Qiaozhi]

Hi Eric,

When I tried to open the Buselli file I get an error message: "The file is damaged..." and it won't open. Does anyone else have this problem or is my Adobe reader at fault?

Allan

I could view both files without any problems.

[ZED]

A definition of ground balance in a P.I. can be viewed as a discrimination function based on time constants because all your doing is discriminating out the
ground response and you do this by discriminating out signals with a certain time constant.
Ground balance systems or GB systems for short can encompass a broad range of time constants or a narrow range of time constants and "unfortunately" any
targets of interest that have the same time constant as the GB system will also be discriminated out as well.
Ground types with high mineralization are relatively easy to discriminate out as long as its time constant remains constant...but ! this is generally
not the case,with high mineralization you usually get high variability resulting in bigger and faster shifts in the grounds time constant and this is where
most GB systems come to grief.


Zed

[Prospector_Al]

I could view both files without any problems.

What version of the Adobe Reader do you have? The problem must be at my end...

Allan

[dougAEGPF]

Ground types with high mineralization are relatively easy to discriminate out as long as its time constant remains constant...but ! this is generally
not the case,with high mineralization you usually get high variability resulting in bigger and faster shifts in the grounds time constant and this is where
most GB systems come to grief.

Zed

Zed how can the above effect you describe be reduced? Taking extra GB samples for example? Using an integrating ADC over the full pulse length to average out all noise(including ground noise)?
dougAEGPF