X vs R

[Jackalope]

Can someone help explain and/or correct my understanding ...

1. Resistive (R) signal is generated from energy loss to primary field (absorption)
1a. Energy loss is from eddy formation and/or magnetic domain switching (iron)
1b. Highly inductive metals borrow more energy from primary field, take longer to form and end (> phase shift)
1c. Rx coil becomes imbalanced due to secondary EMF from 1a above and generates voltage/current

2. Reactive (X) signal is generated from nearly instantaneous reaction to target permeability (redistribution)
2a. No energy loss involved. Iron (magnetic) concentrates flux lines but metal targets distort field away
2b. Eddies in skin of metal inductive target where current lags voltage. Phase shifts < 90 degrees in target current
2c. Secondary field from eddies shifts phase again < 90 degrees. Net phase shift < 180; tends to negate/exclude primary field
2d. Metal/iron distortion of primary field imbalances Rx coil and generates voltage difference which drives a current

3. The current in the Tx coil is compared against the current in the Rx coil, at least at two time points
3a. When Tx current is passing zero the Rx coil voltage is measured. This measures the X component phase (R is zero)
3b. When Tx current is near peaked the Rx coil voltage is measured again. This measures the R component phase (X is zero)
3c. The arctan of the voltages of R/X give the phase shift, caused by the target, between the Tx and Rx coils

Questions:
Why does the secondary target field generate a nearly 180 degree phase shift as compared to phase of Tx coil?
One would assume only R signal can produce a phase shift due to current delay but X signal is said to produce phase shift?
How is 3a & 3b related to the X and R component - for instance, why is the X component found by this procedure?

Thanks for any clarification

[Davor]

It is a point of reference dilema. If you use pure resistive response, e.g. wet salts as point of reference, you'll find it logical. It is ~90° against the ferrite response, and also ~90° against the short circuit perfect conductor on the other side. From ferrites with no conductivity and all ferro-magnetism to all conduction and diamagnetism you have 180° arc. Salts that are resistive are in a middle. +X and -X stand for ferro-magnetism and diamagnetism.

[Aziz]

Hi John,

first of all, your assumptions 1. - 3c. aren't fully correct. Secondly, everyone in the art has a different understanding and definition of X and R.
Better to correct the assumption step by step first.
Aziz

[Tepco]

Can someone help explain and/or correct my understanding ...

1. Resistive (R) signal is generated from energy loss to primary field (absorption)
1a. Energy loss is from eddy formation and/or magnetic domain switching (iron)
1b. Highly inductive metals borrow more energy from primary field, take longer to form and end (> phase shift)
1c. Rx coil becomes imbalanced due to secondary EMF from 1a above and generates voltage/current

2. Reactive (X) signal is generated from nearly instantaneous reaction to target permeability (redistribution)
2a. No energy loss involved. Iron (magnetic) concentrates flux lines but metal targets distort field away
2b. Eddies in skin of metal inductive target where current lags voltage. Phase shifts < 90 degrees in target current
2c. Secondary field from eddies shifts phase again < 90 degrees. Net phase shift < 180; tends to negate/exclude primary field
2d. Metal/iron distortion of primary field imbalances Rx coil and generates voltage difference which drives a current

3. The current in the Tx coil is compared against the current in the Rx coil, at least at two time points
3a. When Tx current is passing zero the Rx coil voltage is measured. This measures the X component phase (R is zero)
3b. When Tx current is near peaked the Rx coil voltage is measured again. This measures the R component phase (X is zero)
3c. The arctan of the voltages of R/X give the phase shift, caused by the target, between the Tx and Rx coils

Questions:
Why does the secondary target field generate a nearly 180 degree phase shift as compared to phase of Tx coil?
One would assume only R signal can produce a phase shift due to current delay but X signal is said to produce phase shift?
How is 3a & 3b related to the X and R component - for instance, why is the X component found by this procedure?

Thanks for any clarification

Not exactly like that. Magnetic materials (Fe) are capable to store some energy in magnetic field, nonmagnetic are not. In a sense, Fe will act like transformer core, non-Fe like shorted transformer winding. R signal is generated by absorption (non-Fe) or addition of stored energy (Fe) producing different polarity signal. This is reactive energy so it is 90deg out of phase.
Any conductive target will however produce coil imbalance, as in 2. In phase, (actually 180 deg out but not important) generating X. In mineralized but nonconductive soil R component will be large, X very small (10:1 ratio roughly but may vary slightly), even X can occasionally go to negative, when magnetization loss exceed conductivity component. You already answered first question in 2. So resulting waveform will vary in amplitude and is shifted in phase referenced to TX, sampling at two points 90deg apart produce X/R ratio, like this:
Ideally conductive nonferomagnetic material will produce signal exactly matching TX waveform, no phase shift, maximum X component and no R component. R component is extracted 90deg out, at that point RX voltage will zero cross, so no R. Less than ideally conductive target will produce less than ideal phase match and some R component signal in one direction, depending on conductivity losses, by extracting some energy from field.

Ideal conductive loseless Fe material is nonsense in physics, so I will avoid that part in simplified description. Normal, conducting Fe target, good old rusty nail for example, will produce both X and R signals. X part due to conductivity, and some phase shift will be added due to energy stored in magnetic field due to its magnetic properties producing R but this time of opposite polarity compared to non-Fe material.
Ideal nonconductive loseless Fe (ideal ferrite) will produce only R signal, it is capable for energy storage too, and this reactive energy will be exactly 90deg out of phase, energy returned, no X signal generated, now waveform will zero cross at 0 deg. Signal polarity will be same as for conductive Fe. Normally soil or ferrites have some small conductivity and some X signal will be produced. Under some conditions, opposite polarity from conducting target. Now, all mess with ground balancing starts here. By shifting initial phase, or by introducing another channel to process ground X-R ratio and subtract from X and what not…
Hope this simple explanation is useful (also hope im right on this) best regard.

[Qiaozhi]

As you correctly stated, there are two factors at work here, absorption and redistribution.

Metal targets are conductive, and eddy currents are generated in the surface of the object. These eddy currents generate their own magnetic field that tends to exclude the applied field from the volume of the target (Lenz's Law). This is an example of redistribution.
If the target is non-conductive (and has a permeability greater than air) the applied magnetic field can occupy the volume of the target. This is an example of absorption.
It is these two effects that cause the applied magnetic field to become distorted, and are responsible for the reactive (X) and resistive (R) components of the received signal.

If you take a highly conductive target (copper, for example) and apply an alternating magnetic field, eddy currents are generated in the surface. The magnetic field created by these eddy currents are phase-shifted when compared to the applied field, and the amount of phase-shift depends on the target conductivity.

However, in the real world, metal targets exhibit a combination of absorption and redistribution, which can make the task of discrimination less than ideal.

[golfnut]

Can someone help explain and/or correct my understanding ...

1. Resistive (R) signal is generated from energy loss to primary field (absorption)
1a. Energy loss is from eddy formation and/or magnetic domain switching (iron)
1b. Highly inductive metals borrow more energy from primary field, take longer to form and end (> phase shift)
1c. Rx coil becomes imbalanced due to secondary EMF from 1a above and generates voltage/current

2. Reactive (X) signal is generated from nearly instantaneous reaction to target permeability (redistribution)
2a. No energy loss involved. Iron (magnetic) concentrates flux lines but metal targets distort field away
2b. Eddies in skin of metal inductive target where current lags voltage. Phase shifts < 90 degrees in target current
2c. Secondary field from eddies shifts phase again < 90 degrees. Net phase shift < 180; tends to negate/exclude primary field
2d. Metal/iron distortion of primary field imbalances Rx coil and generates voltage difference which drives a current

3. The current in the Tx coil is compared against the current in the Rx coil, at least at two time points
3a. When Tx current is passing zero the Rx coil voltage is measured. This measures the X component phase (R is zero)
3b. When Tx current is near peaked the Rx coil voltage is measured again. This measures the R component phase (X is zero)
3c. The arctan of the voltages of R/X give the phase shift, caused by the target, between the Tx and Rx coils

Questions:
Why does the secondary target field generate a nearly 180 degree phase shift as compared to phase of Tx coil?
One would assume only R signal can produce a phase shift due to current delay but X signal is said to produce phase shift?
How is 3a & 3b related to the X and R component - for instance, why is the X component found by this procedure?

Thanks for any clarification

I dont think the terminology is important. X , Y, Z, jam, potato....



To extract amplitude information you only need one Rx. (no fine phase info available).

To extract phase info you need a pair of Rx'es phase offset from each other. (phase split by the maximum you are likely to encounter across all target alloys)



The phase offset of a target will more closely align with one of the Rx channels, this channel gives out more signal during the pass. - the ratio of the two Rx output amplitudes gives the phase result as you mention.

big
short = gold/ally

short
big = silver

medium
medium = bronze/lead



At a basic level, I dont think there is much more too it.

S

[mikebg]

Hi John,
I apologize that during the nearly three years I have no answered your question, which you put at the beginning of 2010:
http://www.geotech1.com/forums/showt...eferrerid=2910

At the same time, another participant in the forum (nick_f) was also put similar question:
http://www.geotech1.com/forums/showt...eferrerid=2910

Then I just pointed out what to read:
http://www.geotech1.com/forums/showt...light=response

Tinkerer in 2010 also sought similar information:
http://www.geotech1.com/forums/showt...766#post113766

I've decided in 2010 to write a lecture titled "Coils and targets beyond basics. Mutual inductance and self-inductance". Then I was spokesman of (R)EMI group - great educators for amateur radio and experts on electromagnetic induction. When I asked experts to publish in forum my material, they banned me, because the lecture is for very beginner in the field of electricity, which would lower the forum level where participants are worldknown designers.

I was not of the same opinion, first because for understanding of my lecture requires a solid education in mathematics and physics, the second because in the Wikipedia this issue was not made precisely to an understandable level in time domain and in frequency domain. Then I lost a lot time for continuous smoothing controversy with (R)EMI group, so I did not think I could write a lesson for dummies and beginners, and can send it via E-mail. This does not violate their desire not to lower the level of the forum.

The lecture is still stuck in my computer but not in English. It remains only to translate it and do some drawings. Now because of Christmas and New Year I will probably have enough time to do this even with a delay of three years.

[Jackalope]

Interesting.

There must be a logical cause and effect explanation to the relationship between a buried target's reaction to the primary field and the current flow in the Rx coil as related to the Tx coil. In other words, I know that the route is there and that some people probably know it but I have yet to find anyone who is capable to lay out that route in its entirety, which to me is very strange. I've read the books, studied detector patents, searched the web, and yet there is not a single explanation that goes to the details of what is really happening.

As a meteorologist, I can logically present the cause and effect between the occurrence of an event and its origins, as far as is known without requiring you to know everything about that field of study. What I don't see is the same plainness and clarity with the most basic origins of a target's response to a magnetic field carried forward to the detector's side and the final result. Certainly there are pieces of the puzzle that are used repeatedly to explain a ferrous and non-ferrous reaction, such as eddies and permeability changes, but the next few links in that cause and effect chain that should move logically toward the final goal of explaining how we receive that information at the detector and process it is lacking; without the missing links I cannot explain the relationships adequately.

I can understand that non-conductive metal with some permeability value will absorb the primary field. And then because of this, what happens? Is it because the field is distorted that the inductively balanced coils drive a voltage in the Rx coil, and if so, how is that markedly different than when it is a conductive metal that absorbs the primary field to induce eddies. Is the secondary field received at the Rx coil producing a different reaction that can be measured by demodulators, which must be the case, but in what way is the current and voltage between the Rx and Tx coil affected by that difference? How is that difference transmitted from target to coil? Why is a certain measurement taken at 90 degrees apart successful in separating the R and X components? How did the composite signal generated such that it is made up of these components in the first instance?

You see, I think answer can be given without going into deep electronic theory - a little bit is needed to prove the case should be sufficient. For instance, giving the reason why the secondary EMF from a metal target excludes the primary field should explain how the waveform is 180 degree apart from the primary sine wave, enough detail to see the logical connection is all that is needed. It is not enough to say it is so without explaining the mechanism behind it. It should be a process of "this does that" because "that did this." The end result of the individual explanations from link to link is the conclusion of causality: a measurement of phase, polarity, and amplitude that says iron or metal in an IB detector.

What I find curious is that up to this point, as far as I have been able to determine, such a systematic explanation has never been presented by anyone anywhere. There have been bits and pieces presented but not with the thoroughness to pull it all together, which I find frustrating. When I think I understand a piece of the puzzle it is a dead end - there is no mechanism to move to the next step in the process of putting the parts together due to a void of information, or often enough other sources of explanation do not agree, or there is oversimplification of the process given because it is inferred that the reader is too dull a chap to follow the logic and would be like a pig looking at a watch. Sometimes it seems to me that the theory of modern detector operation is made up of one part science mixed with two parts fairy dust - there appears to be similarities between the explanation of IB detector operation and the explanations used to justify how a dowsing stick bends over a buried target.

There is still the probable explanation that the lack of clarity is simply because no one really knows in entirety how the darn thing works. I'd like to think that this is not the case, but it does seem to fit the evidence.

[Carl-NC]

John, there's a lot going on, not easy to explain in just a few sentences. One could even write a book about it.

[IBGold]

Hi John,
I apologize that during the nearly three years I have no answered your question, which you put at the beginning of 2010:
http://www.geotech1.com/forums/showt...eferrerid=2910

At the same time, another participant in the forum (nick_f) was also put similar question:
http://www.geotech1.com/forums/showt...eferrerid=2910

Then I just pointed out what to read:
http://www.geotech1.com/forums/showt...light=response

Tinkerer in 2010 also sought similar information:
http://www.geotech1.com/forums/showt...766#post113766

I've decided in 2010 to write a lecture titled "Coils and targets beyond basics. Mutual inductance and self-inductance". Then I was spokesman of (R)EMI group - great educators for amateur radio and experts on electromagnetic induction. When I asked experts to publish in forum my material, they banned me, because the lecture is for very beginner in the field of electricity, which would lower the forum level where participants are worldknown designers.

I was not of the same opinion, first because for understanding of my lecture requires a solid education in mathematics and physics, the second because in the Wikipedia this issue was not made precisely to an understandable level in time domain and in frequency domain. Then I lost a lot time for continuous smoothing controversy with (R)EMI group, so I did not think I could write a lesson for dummies and beginners, and can send it via E-mail. This does not violate their desire not to lower the level of the forum.

The lecture is still stuck in my computer but not in English. It remains only to translate it and do some drawings. Now because of Christmas and New Year I will probably have enough time to do this even with a delay of three years.

Hi Mikebg,

I would be very interested in reading your lecture in English but am unable to understand the group banning you from publishing the lecture as , as well we all know sometimes we have to go back to the very basics to be able to properly explain things in an understandable manner as the people to who the lecture is directed at are not all experts probably far from it keep up your good work.

Regards, Ian.

[Qiaozhi]

Interesting.

There must be a logical cause and effect explanation to the relationship between a buried target's reaction to the primary field and the current flow in the Rx coil as related to the Tx coil. In other words, I know that the route is there and that some people probably know it but I have yet to find anyone who is capable to lay out that route in its entirety, which to me is very strange. I've read the books, studied detector patents, searched the web, and yet there is not a single explanation that goes to the details of what is really happening.

Let's start with a simple inductor and apply a potential difference across the coil. What happens?
At time t=0 the inductor looks like an open-circuit, and no current flows through the windings. As time progresses, the current flow increases exponentially until it reaches a maximum value that is limited by the DC resistance of the windings. As you can readily see, the current in an inductor lags the applied voltage. If a sine wave is applied to an ideal inductor, the current would lag the voltage by 90 degrees.

Now let's take a transformer. The magnetic field from the primary interacts with the secondary and generates a current in the secondary windings. Again, the current in this winding lags the voltage across the coil, resulting in a total phase shift of 180 degrees.

Then apply the same thought process to a metal detector. The TX coil creates a magnetic field that creates eddy currents in a nearby metal target. These eddy currents generate their own magnetic field that interacts with the RX coil. So we have a 90 degree lag at the TX coil, a 90 degree lag at the target, and a third 90 degree lag at the RX coil (270 degrees in total). The result is a 90 degree phase-offset at the receiver (ignoring polarity).

For an ideal resistive target, (one that exhibits absorption, rather than redistribution) there is no associated phase-shift, but there will be a reduction in amplitude. In the real world, metal targets have both absorption and redistribution properties, and a metal detector can measure these two properties separately to give an idea of the target type.

VLF detectors generally use quadrature sampling to provide discrimination. The first sample is taken at the zero-crossing of the RX waveform, and the second sample is taken 90 degrees later at the peak. By plotting these two values on the X and Y axes of a graph, it is possible to measure the angle formed by the resulting vector and the X-axis. This angle can be determined in a digital metal detector by directly calculating the arctangent of Y/X (and displaying the result on a VDI meter). Alternatively, an analog detector can provide discrimination by logically ANDing the outputs of the X and Y channels, and providing a means of adjusting the second (peak) sample to allow for selective rejection of undesired targets.

Have you read "Inside the METAL DETECTOR"?
At lot of this is explained in Chapters 7 and 9.

[Skippy]

Qiaozhi: I'm thinking this is wrong:"...and a third 90 degree lag at the RX coil (270 degrees in total)". The voltage generated by the receive coil will LEAD the induced current by 90 degrees, so this final phase shift is -90 degrees, giving 90+90-90 = 90 degrees in total. This matches what is seen in practice

[Qiaozhi]

Qiaozhi: I'm thinking this is wrong:"...and a third 90 degree lag at the RX coil (270 degrees in total)". The voltage generated by the receive coil will LEAD the induced current by 90 degrees, so this final phase shift is -90 degrees, giving 90+90-90 = 90 degrees in total. This matches what is seen in practice

Skippy: I agree. My explanation of how the phase-shift at the RX coil occurs is grossly over simplified.

In fact, the more I think about it, the less I like it, (this part wasn't in ITMD, by the way) ... so let's try again:

I think the explanation is a lot simpler than that.
If we consider the search coil (with its TX and RX windings) as a loosely coupled transformer, then according to theory the secondary voltage will be in phase with the primary current. In other words, the RX voltage will be 90 degrees out of phase with the TX voltage. You can easily confirm this for yourself in a SPICE simulation. Drive a 1mH inductor (3 ohm resistance) from a sine source (10kHz), and link this to a second 1mH inductor (3 ohm reistance) with a parallel 253nF capacitor using a coupling coefficient of 1m. This will emulate an induction balanced coil arrangement. The results clearly show the RX voltage in phase with the TX current, which lags the TX voltage by 90 degrees.

In you want to get adventurous, you can add a third inductor to emulate a target. e.g. 1mH with a parallel 100 ohm resistor. Then link this inductor to both the TX and RX coils with two separate K statements using a coefficient for both of 40m; and you will notice the RX voltage (and current) phase-shift to the right. This phase-shift increases as the target signal coupling is increased.

Now ... what was the question?
I hope I'm not digging a huge hole here.

[Sergey_P]

http://www.md4u.ru/download/file.php?id=14187


(RU)

[mikebg]

Sergey, the main error in your theory is a wrong formula
Iv=E*р, where
Iv is eddy current,
E is electromotive voltage and
р is conductivity.
This is not the Ohm's law because conductivity has geometric dimension (Siemens pro meter).
The Ohm's law for an eddy current loop is
Iv=E*Y or Iv=E/R, where
Y and R are conductance and resistance of the eddy current loop. They depend on conductivity and eddy diameter.
However the eddy loop has not only resistance; it has inductance L which together with Y or R forms in time domain a timeconstant
Tc=L*Y or Tc=L/R
The Tc forms in frequency domain a cutoff frequency
wc=1/(2*pi*Tc)=0.159/Tc,
at which the eddy current has phase delay 45 deg and amplitude drop 3 dB. At other frequencies we have different phase lags.
You can SPICE simulate this in complex plane for whole frequency spectrum:
http://www.geotech1.com/forums/showt...669#post101669
Please search WEB to download
"NI_Multisim_Analog_Devices_Edition_10_0_1.exe "