An improved metal detector

Abstract

Abstract A method to detect a target in soil using a metal detector, comprising: transmitting a transmit magnetic field using a transmitter; receiving a receive magnetic field using a receiver to produce a receive signal; processing the receive signal to produce an output indicative of the target; wherein the step of processing the receive signal is configured such that, in response to a specific transmitted magnetic field, the receive signal is processed with corresponding models such that the output signal is simultaneously nulled to components of the receive magnetic field due to log-uniform distributed VRM particles, log-linear distributed VRM particles and log-quadratic distributed VRM particles of the soil. Abstract A method to detect a target in soil using a metal detector, comprising: transmitting a transmit magnetic field using a transmitter; receiving a receive magnetic field using a receiver to produce a receive signal; processing the receive signal to produce an output indicative of the target; wherein the step of processing the receive signal is configured such that, in response to a specific transmitted magnetic field, the receive signal is processed with corresponding models such that the output signal is simultaneously nulled to components of the receive magnetic field due to log-uniform distributed VRM particles, log-linear distributed VRM particles and log-quadratic distributed VRM particles of the soil. 20 25 20 50 47 02 J ul 2 02 5 2 0 2 5 2 0 5 0 4 7 0 2 J u l 2 0 2 5 2 0 2 5 2 0 5 0 4 7 0 2 J u l 2 0 2 5

Description

2025205047   02 Jul 2025 TECHNICAL FIELD

[0001] The present disclosure relates to a metal detector. BACKGROUND

[0002] Hand-held metal detectors are used to locate gold, explosive land mines or ordnance, coins and treasure. These metal detectors usually consist of transmit electronics generating a repeating transmit signal cycle of a fundamental period, which is applied to a transmit inductive winding, which transmits a resulting varying magnetic field, sometimes referred to as a transmit magnetic field or transmitted magnetic field.

[0003] These metal detectors also contain a receiver that receives a received magnetic field, and contain receive electronics that processes a receive signal from the received magnetic field, during one or more receive periods during the repeating transmit signal cycle, and the receive signal is processed to produce an indicator output signal, the indicator output signal at least indicating the presence of at least metal targets within the influence of the transmit magnetic field. Invariably, in hand-held metal detectors, the receiver is a receive inductive winding. In all well-known metal detectors purpose-built to detect gold, the indicator output signal is via modulation of an audible output.

[0004] Metal detectors may be classified into two groups depending on their transmit signal and synchronous demodulation (or sampling) signal processing functions, namely frequency-domain or timedomain. Frequency-domain metal detectors may be thought of as having signal processing that includes synchronous demodulators responsive to specific frequencies of the transmitted signal (plus possibly harmonics to varying degrees), whilst time-domain metal detectors may be thought of as having synchronous demodulators sensitive to specific periods of time usually following transitions in a transmitted magnetic field. Examples of time-domain are well-known pulse-induction (PI) metal detectors, or PI-like metal detectors with periods of alternating constant transmitted magnetic fields with rapid magnetic field transitions connecting the said alternate periods of constant transmitted magnetic fields, for example see US8,614,576, US9,250,348, US9,348,053. Examples of frequency-domain metal detectors include the most common form of metal detector, the single-frequency sine-wave transmitting type, and multi-frequency metal detectors that usually transmit and receive several different frequencies, for example see US4,942,360, US7,432,715.

[0005] The present disclosure provides an alternative to improve the operation of a metal detector to detect a target. 2025205047   02 Jul 2025 SUMMARY

[0006] According to a first aspect of the present disclosure, there is provided a method to detect a target in soil using a metal detector, comprising: transmitting a transmit magnetic field using a transmitter; receiving a receive magnetic field using a receiver to produce a receive signal; processing the receive signal to produce an output indicative of the target; wherein the step of processing the receive signal is configured such that, in response to a specific transmitted magnetic field, the receive signal is processed with corresponding models such that the output signal is simultaneously nulled to components of the receive magnetic field due to log-uniform distributed VRM particles, log-linear distributed VRM particles and log-quadratic distributed VRM particles of the soil.

[0007] In one form, the step of processing the receive signal is configured such that, in response to a transmitted single magnetic ramp of duration followed by a constant magnetic field of a same value as that at a termination of the magnetic ramp, the output comprises an output signal that is simultaneously nulled to rates of change of components of the receive magnetic field in following forms across a processing bandwidth of the handheld metal detector: Ln[(t + ) / 1]; / n2 (t + t) — Zn2 (t); / n3 (t + t) — In3 (t); and a constant value; where t = 0 at the termination of the magnetic ramp.

[0008] In one form, the step of processing the receive signal is configured such that, in response to a transmitted magnetic isolated half cosine wave connecting a constant transmitted magnetic field to a different constant transmitted field, the output comprises an output signal that is nulled to rates of change of components of the receive magnetic field in following forms across a processing bandwidth of the handheld metal detector: fj cos =^ In3 (x + ))dx, where the transmitted half-cosine pulse is of the form H = a when t<=-T, H = b when t>=0, — — a + — — a) cos (?) when -T<t<0, with a receive period commencing at t=0 at the termination of the said half cosine-wave.

[0009] In one form, the transmit magnetic field comprises at least one current ramp period, and (i) at least one period of zero transmitted magnetic field or (ii) at least one constant current period.

[0010] According to another aspect of the present disclosure, there is provided a metal detector configured to perform the method of the first aspect.

[0011] According to another aspect of the present disclosure, there is provided a non-transitory computer readable medium, comprises instructions to perform the method of the first aspect. 2025205047   02 Jul 2025 BRIEF DESCRIPTION OF DRAWINGS

[0012] Embodiments of the present disclosure will be discussed with reference to the accompanying drawings wherein:

[0013] Figure 1 depicts a general form of one embodiment of the present disclosure;

[0014] Figure 2 provides an example of a synchronous demodulation function that simultaneously nulls log-uniform, log-linear and log-quadratic distributed VRM;

[0015] Figure 3 depicts an exemplary basic block diagram of a metal detector; and

[0016] Figure 4 shows a normalised first-order target frequency response on X axis versus response for a multi-frequency example. DESCRIPTION OF EMBODIMENTS

[0017] Many magnetic soils have relatively high magnetic permeability, indeed for some such soils, most of the soil sticks well to a magnet. Most of the soil magnetic material consists of multi- or singledomain ferrimagnetic particles, with the most common being magnetite and maghemite. If these single domain crystals are sufficiently small nanoparticles, the crystals’ magnetisation may randomly flip direction under the influence of temperature (with the field most often aligned to the easy axis; for more details see for example, M. Knobel, W. C. Nunes, L. M. Socolovsky, E. De Biasi, J. M. Vargas, and J. C. Denardin, Superparamagnetism and Other Magnetic Features in Granular Materials: A Review on Ideal and Real Systems, 2008, Journal of Nanoscience and Nanotechnology Vol. 8, 2836-2857). The mean time between flips is called the “Neel relaxation time.” It is Arrhenius in nature: t = Toe-KV / (kr), where T is the temperature and k is the Boltzmann constant, V the particle’s volume, and K its anisotropy energy density, and thus KV is the energy barrier to flip the nanoparticle’s magnetic field from an easy axis direction, through its hard plane axis to the opposite direction of the easy axis. ro is material dependent and of the order on 0.1 to 1 nanosecond. As can be seen from the equation, for a given shape and magnetic material, the Neel relaxation time is exceptionally dependent on size, that is a small relative change in size has a large effect on the relaxation time. Applying a magnetic field to these randomly flipping magnetic nanoparticles biases the flips to align more on average to the applied field. If the applied field is then removed, the magnetic nanoparticles then re-randomise (entropic), and thus generate a decaying magnetic field during the re-randomisation. If the measurement time for these randomly flipping magnetic field nanoparticles is much longer than the typical Neel relaxation time of a sample, 2025205047   02 Jul 2025 then the sample is said to be “superparamagnetic.” This occurs when all these magnetic moments of the individual magnetic atoms add to contribute to effectively a “single giant” magnetic moment. On the other hand, if the measurement time was much shorter than the typical Neel relaxation time, the sample would be magnetically blocked.

[0018] For a log-uniform distributed sample of “all sized” nanoparticles, the magnetic susceptibility equation is known to those skilled in the art, namely (^ = ^+1^ J      1 + i^r where the applied frequency is ^ / 2 , / sp is the superparamagnetic susceptibility, and jb;that of the blocked state. This type of magnetism is usually referred to as Viscous Remnant Magnetism, or Viscous Remnant Magnetization, or VRM, in the metal detector industry, and will from here on be referred to as such. Most magnetic soils do contain a significant quantity of VRM particles with mostly log-uniform frequency distribution over typical multi-frequency or time-domain metal detector effective detection bandwidths, but usually with a small log-linear distribution component in addition. US8,106,770 discloses that a log-uniform distribution with a randomly related log-linear VRM distribution component may be simultaneously universally cancelled by subtracting relatively medium frequency measured resistive VRM components from relatively high-frequency resistive VRM components added to relatively low frequency resistive VRM components, in a specific ratio relationship to achieve the said log-linear plus log-uniform VRM component null. Various patents, e.g. US9,348,053 and US9,250,348 disclose a rate-of-change of a log-uniform VRM received signal in response to a transmitted magnetic field.

[0019] Induced EMFs in the receive inductive winding directly proportional to the rate of change of the transmitted magnetic field are called the (received) reactive component, usually called the “X” component, and in terms of targets, this arises due to the energy lossless components of ferro- or ferrimagnetism, and the “inductive” component of eddy currents, whereas the “R” component is manifest due to energy loss associated received signals, such as the associated energy loss component of soil VRM and eddy currents. Unless otherwise stated, received and processed soil VRM components below refers to the resistive components, not the reactive (X) components. In the time domain, it is assumed herein that X plays no role in receive signals, unless otherwise stated.

[0020] Before synchronous demodulation signal processing, the received rate of change of magnetic signal induced in a receive inductive winding is convolved with the transfer function of the receive inductor circuit plus that of the metal detector receive electronics preamplifier. For example, an induced EMF in a second order critically damped “parallel LCR” magnetic receiver circuit, from a response for a log-uniform VRM sample to which a single magnetic linear ramp that transitions between an initial 2025205047   02 Jul 2025 (“infinitely long period”) constant valued magnetic field to a final constant magnetic field is applied, wherein the linear ramp is of duration , in the time domain would be proportional to: Convolved induced EMF: x [oj(t + t) + 1^-^+^57(^) - 5i[((i)(t + t)]} + Ln ( + (o>t + 1)6-^57(0)0 - y - Ln(o)0] + (a + / j«7)e“"t where Ei is the exponential integral, a>2 =   , and a and P define the transient conditions in the LCR network when t=0 at the termination of the linear ramp and commencement of the altered constant magnetic field. y is the usual Euler’s constant symbol.

[0021] The received signal from the magnetic soils may be thousands of times higher than the signal from a relatively deeply buried target, and thus the receive signal from soils needs to be cancelled very precisely in order to detect the faintest metal target signals.

[0022] As implied above, the magnetic grain size range contributing to the received soil VRM signal components that a metal detector effectively nulls out via ground balancing, is relatively very narrow. Thus, it is likely that soils that are well-homogenised due to being extensively exposed to weathering, have a wide range of randomized particle distribution size spanning across and well beyond this said narrow VRM magnetic nano-particle size range. Thus, the VRM distribution is very likely to respond as being log-uniform with a very small log-linear component. Indeed, a range of a few dozen so called “hot rocks” collected over several years by gold prospectors using metal detectors several decades ago, were found to be nulled out accurately when using synchronous demodulation functions designed to cancel both log-linear VRM and log-uniform VRM simultaneously. This evidence supported the soil log-linear VRM model well. Furthermore, the most popular state of the art high-end metal detectors purposely designed to seek metal targets in magnetic soils, are the most popular because they include receive synchronous demodulation functions designed to cancel both log-linear VRM and log-uniform VRM simultaneously, which also suggests that this soil model is at least reasonably accurate.

[0023] However, the at the time of writing, these said high-end state of the art commercially available metal detectors purposely designed to work in highly magnetic soils, do exhibit spurious soil signals. These spurious soil signals have been attributed to many factors, namely: A. Soil partial magnetic saturation: details of which are disclosed in US18 / 266962. In some soils the indicative response from this non-linear source can be relatively severe depending on how the operator swings the coil above the soil surface. 2025205047   02 Jul 2025 B. Imperfect electronics such as ringing in coils (disclosed in US11474274) C. Soil conductivity. D. Possibly some other hitherto unknown mechanisms.

[0024] However, since then, new coil developments described in US18 / 266962 and US11474274 have eliminated the problems of points A., B., and C. above, leaving only point D: Thus, only now at the time of writing has it become unambiguously certain from tests with the use of these next generation of metal detectors and coils, that the log-linear VRM soil model is not accurate in some soils, but is accurate others. Further, these spurious soil signals are now more obvious because without the above problems of A, B and C, it is possible to increase the signal amplification of the metal detector thereby rendering the un-nulled soil sources overtly clearer without the confounding interfering signals from saturation, electronic ringing and soil conductivity.

[0025] It was noticed that these unexplained spurious signals appear to be associated with clays. Furthermore, it was noted that car tracks, tree roots, and soil around previous underground fires (such as burnt tree stumps) exhibited relatively high levels of these unexplained spurious signals in the said next generation of metal detectors using log-linear VRM nulling synchronous demodulation functions. In addition, some target-like sounding signals are generated in soils by small local concentrations of clay. (“Small” meaning e.g. 5-25cm scale.)

[0026] Follow-up research showed that clays contain nanoparticles. It was then hypothesised that VRM magnetic nanoparticles may be transported via water into and out of clays, for example via diffusion or under differential pressure. If so, then if a wet clay is placed under differential pressures, it should force out VRM magnetic nano-particles, because clay nano-particles bond together considerably more tightly than they do to the VRM magnetic nano-particles, and thus the latter could be transported out of (or into) the clay via water movement. This water magnetic nano-particle transportation mechanism into and out of clays could substantially bias the VRM particle size distribution as it is forced through the clay nanoparticles, to a sufficient extent so as to distort the grain size distribution characteristics away from the log-linear VRM soil model. Such a mechanism would be consistent with the hypothesis that soil differential pressures caused by vehicles traversing dirt road tracks and growing tree roots forcing a magnetic nanoparticles redistribution in wet clays, have water borne movement rates being a function of magnetic grain size.

[0027] To check the above hypothesis, an experiment was performed: It was guessed that the rocks of the hot-rock collection were formed via geological processes resulting in log-linear VRM, and that the said non-log-linear VRM soil source is mainly associated with clays. It was also guessed that an approximation to satisfy the distorted VRM distribution component away from log-linear could be 2025205047   02 Jul 2025 approximated as log-quadratic. A mathematical model was developed for log-quadratic VRM, and then a synchronous demodulation function designed to cancel such a component (plus log-uniform VRM and log-linear VRM simultaneously and “D.C.”). A next generation state of the art metal detector with stereo audio then was employed to test outcomes of this synchronous demodulation function, simultaneously with a log-linear VRM nulling synchronous demodulation function, wherein the left stereo audio was modulated (only) by an output from the said log-quadratic nulling synchronous demodulation function, and the right stereo audio was modulated (only) by an output from a log-linear VRM distribution nulling synchronous demodulation function. This metal detector was then taken to a gold field containing very wet magnetic soils containing clay, and a next generation coil that cancelled out soil conductivity relatively well was used (for example the coil disclosed in US18 / 266962). The coil was swept over the soil and the metal detector “ground balanced.” Next the operator stood on the soil for some time which caused soil compression. This left shallow footprint “hollows” in the wet soil. It was then noted that these shallow imprints filled with water, but not just water, rather water transporting what looked like black magnetite fine particles. The use of a magnet showed that these particles were highly magnetic which strongly suggested they were indeed mostly magnetite. It was assumed that these magnetite particles were nano-particles forced out of the clay. The metal detector coil was then passed over the dammed-up footprint containing the alleged magnetite nano-particles. The log-linear nulled right stereo produced a relatively loud signal, indicating the footprint’s VRM distribution was not accurately defined by a log-linear soil model, whereas the log-quadratic nulled left stereo produced hardly any signal, supporting all of the above hypothesises. This metal detector was further tested on car tracks, tree roots, and clay soils generally, all of which supported the same hypothesis. In regard to the same findings for magnetic clay soils subjected to underground fires, it is suspected that the resulting (rapidly) expanding fire heated water and steam passing through the clays involved the same size dependent magnetic nano-particle transportation mechanism, thus likewise distorting the VRM particle size distribution away from the log-linear VRM soil model. Further it seemed odd in hindsight that no “hot-rocks” previously tested included log-quadratic VRM distributed magnetic particles. Thus a far bigger collection of hot rocks was tested, and it was then found that some hot-rocks do also require log-quadratic receive synchronous demodulation functions to null out their VRM signal sources, but most hot-rocks in the collection appear to be defined satisfactorily as having a log-linear VRM distribution. Further it was noted that some hot-rocks that exhibited relatively high X components compared to VRM, which suggests a relatively higher percentage of larger grain size bias compared to the average VRM hot rock, satisfied the log-quadratic model well, but not the log-linear distribution. However, some hot rocks, including ones of relatively low mass density but having relatively high VRM to X ratios and relatively high absolute VRM responses, suggesting a bias away from larger magnetic particles in favour of smaller, also clearly satisfied the logquadratic VRM model, but in the opposite spectral slope sense of the log-linear component. Examples are some hot-rocks associated with volcanic magnetite (for example from the Golan Heights or Natal), and various high X rocks found in various Australian locations. Lastly, clays relatively expand and shrink significantly when wet and dry respectively, and do so with such high pressures as to notoriously causes 2025205047   02 Jul 2025 cracks in houses, sometimes excessively so. This also suggests a differential pressure mechanism for changing VRM away from a log-linear VRM distribution in clays. For the sake of understanding, it should be noted that Stokes’ law v = -^-^(^--) provides a settling velocity of the particle of density p in the water of density Y (=1) and viscosity n (~0.001). This is effectively zero for VRM sized magnetic particles. Hence Brownian motion dominates the movement of particles rather than gravity. The crossover size is roughly 1gm. In other words, the change in wet soils away from log-linear VRM is not because the larger of VRM magnetic particles are drifting down under gravity faster than the smaller VRM particles, thereby causing the said increasing log-quad VRM components, rather, diffusion dominates, not gravity. Instead rather, diffusion tends to re-homogenise the VRM particles to reduce the log-quadratic VRM distribution towards a log-linear VRM distribution, and similarly, to re-homogenise the VRM particles to reduce a log-linear VRM distribution towards a log-uniform VRM distribution. This may explain why log-quadratic VRM appears worse during the wet seasons, but usually returns back to the same “dry soil base line” during the dry seasons.

[0028] In retrospect, it is unlikely that had a log-quadratic receive synchronous demodulation been installed in the current (or previous) generation of high-end metal detectors with their attendant said electronic, soil saturation, and soil conductivity issues, plus their monaural audio compared to the next generation stereo, may not have been of benefit to the average detector operator. Rather, now that these said issues are resolved, plus the advantages of stereo implemented as above, allows for the advantages of log-quadratic VRM nulling synchronous demodulation in certain soils to excel if included in the next generation of high-end metal detectors, especially when used to prospect for gold in clay magnetic soils often to be found in many gold fields, or, to recover land mines.

[0029] Generally, the main problem is an obscuring of the faintest subtle target signals that get buried below the spurious log-quadratic spurious ground “noise” signal if only log-linear VRM nulling synchronous demodulation functions are used: hence in essence the problem to be solved is a “target signal-to-soil-noise” issue.

[0030] The receive rate of change of magnetic VRM responses to a change in a transmitted constant magnetic field from one value to another via a linear magnetic ramp of duration t is for:

[0031] log-uniform VRM with no higher orders: aLnr / ) ,

[0032] log-linear VRM with no higher orders: « Mim[Ln2(t + t) - Ln2(t)] + 1^(-^) ,

[0033] log-quadratic VRM with no higher orders: 2025205047   02 Jul 2025 « ^quad [Ln3 (t + T) - Ln3 (t)] + Mim [Ln2 (t + t) - Ln2 (t)] + Ln(^) , where t=0 is at the termination of the linear ramp and the commencement of the immediately following constant magnetic field, and ^nn and gquad are relative VRM log-linear and log-quadratic soil content coefficients respectively.

[0034] Note that the above three expressions are rate-of-change expressions (derivatives) because this is the form that is induced as an EMF in the receive inductive winding, and it is this EMF that is then further processed. Hence, the said magnetic linear ramp is in effect an induced rectangular response EMF signal in the receive inductive winding, but as above, convolved with the receive inductive winding transfer function. The coefficients Mquad and ^nn vary considerably for different soils. For example one soil may be very magnetically homogeneous, but the magnetic concentration varying, and hence from a practical point of view, only the log-uniform component needs to be cancelled with the aid of “ground balancing.” Another soil may have a significant varying log-uniform to log-linear ratio of VRM, and this will require the cancellation of both log-linear and log-uniform components, and varying ratios of all three will require all three said components to be cancelled simultaneously. The “linear ramp” may not be perfectly linear due to inductive coupling of VRM soil components, or soil conductivity, or other eddy currents such as those arising from the PCB, connectors or metal targets. Yet further, the time-constant of the transmit inductive winding ensures an exponential waveform for a given constant voltage applied to the transmit inductive winding during the linear ramp period, or, such a said applied voltage may not be constant, and so forth. Thus the expressions above may not be exact due to small deviations away from a linear current ramp. For example, consider just the finite resistance of the transmit inductive winding. Suppose the transmit inductive winding has an effective series inductance of say 300^H, and an effective total series resistance of 1Q that includes the transmit electronics output impedance. Further suppose that the applied voltage across the transmit inductive winding is 180V during the said linear current ramp, and the current flowing through the transmit inductive winding changes by a total of 3A, and any external inductive mutual coupling to the transmit inductive winding is ignored. Then the duration ( ) of the assumed linear ramp is ~ ^10003 = 5, and if the “best fit waveform” is for example is assumed if the applied voltage is instead approximated to 180.9V, then the actual transmit inductive winding current is 180.9(1 lxt 10.0003 - 1.5 Amps which has a standard deviation of about just 9mA away from a linear ramp; that is, a good approximation to a linear ramp. 2025205047   02 Jul 2025

[0035] The net received VRM resistive responses depends on the total history of the applied field, for example, suppose the transmit signal is a repeating quasi-square wave current of normalised fundamental period of 2 units, consisting of periods of alternating sign constant magnetic field, each of the same absolute magnitude, with each constant magnetic field of period being 1-t, separated by periods of linear current ramps of duration t, then the net rate of change in received magnetic response to log-uniform VRM during constant current periods, without higher-order VRM distribution contribution, is proportional to: }, where t=0 at the commencement of a constant current (transmitted magnetic) period, at the termination of a current ramp, and r is as usual the symbol for a gamma function.

[0036] This signal, as above, is convolved with the receive LCR critically damped receive inductor, and this then convolved with the forward transfer function of the receive preamplifier before demodulation occurs. Suppose these said convolutions may be approximated as a group delay , which is not sufficiently accurate in practice, but reasonably usefully indicative of a close VRM soil null outcome. Using this approximation, the accumulated contribution to the averaged signal (low-pass filtered) of the synchronous demodulation processing (for log-uniform VRM only), of n synchronous demodulation gains ¢j at between times t = aj and t = jj during only the current (“present”) transmitted constant current period for a generalised multi-period waveform, where the duration of the i-th previous constant current period plus its associated preceding period of linear current ramp period, was of duration 07, is proportional to ^jj=±¢j Xj=o — 1j{ [Pj — d + t + o’;]Ln(Pj — 8 + t + oj) — [dj — 8 + t + o]]Ln(ctj — 6 + t + 0 t) — [Pj — 8 + crj]Ln( Pj — 8 + at) + [ dj — 8 + (y^Ln^aj — 8 + aj)}, wherein as above, the transmitted current of each successive constant current period alternates in sign but all have the same absolute current magnitude, and linear transmitted current ramps each of duration transitions the transmitted current between each successive transmitted constant current period.

[0037] Similarly, the same processes apply to the log-linear and the log-quadratic contributions. Assuming that VRM distributions of higher order than log-quadratic are insignificant, then to create a universal VRM null, the said average for each of the log-uniform, log-linear and log-quadratic VRM contributions must be zero, and also, the net “DC” integrated synchronous forward gain functions must be zero too: to null asynchronous received rates-of-change of magnetic fields, that includes coil movement in the earth’s magnetic field, etc. In regards to the latter, suppose the fundamental transmit waveform is a 2025205047   02 Jul 2025 repeating sequence of m constant current (receive processing) periods long, where m is an even number, with the sign of each successive constant current period changing, then X&ilj^kj (P kj-akj) = 0, where Qkj is the gain of j-th synchronous demodulation period of the kth constant current period of the fundamental repeating transmitted sequence, and pk— — akj is the duration of the said jth synchronous demodulation period of the k-th constant current period of a fundamental repeating sequence with a gain of ^kj, and nk is the number of discrete different demodulation gain periods of the kth constant current period.

[0038] Note: in practice, metal detectors that ideally are supposed to transmit constant magnetic fields during the receive demodulation periods, do not in fact transmit ideal constant currents because of nonideal electronics behaviour, and because of soil and target energy loss components that alter the “ideal constant transmitted field.” Thus, herein the term constant magnetic field or constant transmit current should be understood to have such limitations.

[0039] Figure 1 depicts a general form of one embodiment of the present disclosure. In this form, a method to detect a target using a metal detector is provided. The first step indicated as 1 in Figure 1 comprises transmitting a transmit magnetic field using a transmitter. The transmit magnetic field may take many forms. For example, it may comprise at least one current ramp period. As another example, it may comprise one current ramp period and zero or more constant current periods. Other forms of transmit magnetic field of known time-domain and frequency-domain metal detectors may be used too.

[0040] The next step is as indicated as 3 in Figure 1, which is to receive a receive magnetic field using a receiver to produce a receive signal.

[0041] The next step is as indicated as 5 in Figure 1, wherein the step of processing the receive signal is configured such that, in response to a specific transmitted magnetic field, the receive signal is processed with corresponding models such that the output signal is simultaneously nulled to components of the receive magnetic field due to log-uniform distributed VRM particles, log-linear distributed VRM particles and log-quadratic distributed VRM particles of the soil.

[0042] In one embodiment, the step of processing the receive signal is configured such that, in response to a transmitted single magnetic ramp of duration followed by a constant magnetic field of a same value as that at a termination of the magnetic ramp, the output comprises an output signal that is simultaneously nulled to rates of change of components of the receive magnetic field in following forms across a processing bandwidth of the handheld metal detector: Ln[+ + t) / t]; fn2(r + t) — fn2(t); 2025205047   02 Jul 2025 in3 (t + t) — Zn3 (t); and a constant value; where t = 0 at the termination of the magnetic ramp. The output signal is nulled to rates of change of components of the receive magnetic field due to log-uniform distributed VRM particles, log-linear distributed VRM particles, log-quadratic distributed VRM particles.

[0043] Note, that this excitation waveform is chosen for practical reasons; it is an approximate subcomponent of many transmit waveforms. Alternatively, the corresponding frequency domain mathematics could equally suffice to define the relationships. The mathematics provided for the said magnetic ramp is sufficient for a person skilled in the art to derive log-quadratic distributed VRM decay waveforms from other transmit waveforms. For example, the rate of change of the received magnetic field unique to logquadratic VRM (ignoring components from both log-linear VRM and log-uniform VRM) excited by a transmitted magnetic isolated half cosine wave connecting a (“infinitely long period”) constant transmitted magnetic field to a different constant transmitted field, is of the form: fj cos ™ In3 (x + t)dx, where the transmitted half-cosine pulse is of the form: H = a when t<=-T, H = b when t>=0, H — a + — — ao cos (y) when -T<t<0, with a receive period commencing at t=0 at the termination of the said half cosine-wave.

[0044] Accordingly, in another embodiment, the step of processing the receive signal is configured such that, in response to a transmitted magnetic isolated half cosine wave connecting a constant transmitted magnetic field to a different constant transmitted field, the output comprises an output signal that is nulled to rates of change of components of the receive magnetic field in following forms across a processing bandwidth of the handheld metal detector: fj cos ™ In3 (x + t)dx, where the transmitted half cosine pulse is of the form H — a when t<=-T, H — b when t>=0, = — a + — — ao cos (^) when -<t<0, with a receive period commencing at t=0 at the termination of the said half cosine-wave. The output signal is nulled to rates of change of components of the receive magnetic field due to log-uniform distributed VRM particles, log-linear distributed VRM particles, log-quadratic distributed VRM particles.

[0045] There are other possible embodiments for the transmitted magnetic field and their corresponding models to process receive signal, as long as the output signal is simultaneously nulled to components of the receive magnetic field due to log-uniform distributed VRM particles, log-linear distributed VRM particles and log-quadratic distributed VRM particles of the soil.

[0046] Figure 2 provides an example of a synchronous demodulation function that simultaneously nulls log-uniform, log-linear and log-quadratic distributed VRM, for a repeating quasi-square wave transmit current with a fundamental period of 256 time-units, and rapid change-of-transmit current linear ramp periods of 6 time-units between successive constant current periods of opposite polarity of 122 2025205047   02 Jul 2025 time-units each. The “X-axis” of Figure 2 commences at zero at the transition between the termination of a period a rapid change of transmit current and a constant current polarity of a first polarity.

[0047] A response from soils with only log-uniform VRM and no higher-orders is shown as graph 19, and a response from soils with log-uniform VRM plus log-linear VRM but no higher orders is shown as graph 17, and, a response from soils with all three of log-uniform VRM plus log-linear VRM plus logquadratic VRM, but no higher-orders is shown as graph 15. Graphs 15, 17 and 19 are arbitrarily normalised to a magnitude of four at a time indicated by 11. The contributions of log-linear VRM and especially log-quadratic VRM are very highly exaggerated in Figure 2 solely for the purposes of aiding understanding and are far from reality where it would be hard, if not impossible, to spot the differences from real soils visually on the scales shown in Figure 2. An example of a synchronous demodulation forward gain function is shown as graph 13 that commences at time 11 post the commencement of the shown constant current period. For the opposite polarity, a second polarity constant current period (not shown for simplicity), all four graphs 13, 15, 17 and 19 are inverted; that is the same as Figure 2, but, mirror-imaged about the horizontal axis. This then guarantees a null to demodulated asynchronous magnetic fields (outside the post synchronous demodulator low-pass filter bandwidths) and also a null to DC received signals. To repeat to ensure understanding, the synchronous demodulation multiplying function 13 cancels out simultaneously all three components of log-uniform VRM, log-linear VRM and log-quadratic VRM.

[0048] The same principles apply to the frequency domain. For example, consider four processed frequencies, a, p, x, 8 with associated resistive component gains of “1” for a, y for P, n for X, and 9 for 8, and, a gain of k for a reactive difference component of 8 and a, that is gains of +k for frequency 8 and -k for a. These gains take account of the transmitted magnitudes at each frequency and forward gain transfer function of the receive preamplifier (and receive winding plus coil cable transfer function as well). The said five synchronous demodulation outputs are added. Note: to ensure understanding of signs, the log-uniform VRM reactive X component decreases with increasing frequency. Then to create a simultaneous null to log-uniform, and log-linear, and log-quadratic VRM distributions, and also soil conductivity, but no higher order VRM distribution terms:

[0049] y = {[ / Ln(5) - 5Ln( / )] ^4Ln3(a) + Ln (-) [12Ln2(a) — n2]} + [%Ln(a) — aLn( / )] (4Ln3(5) — ^(-) [12Ln2(8) — n2]} + [5Ln(a) — aLn(5)] (in (-) [12Ln2( / ) — n2]} + 6{[Ln2(a) — Ln2 (5)](5 — a)Ln2( / ) + (a + 5)Ln2(a)Ln2(5)} + 2[( / — 3)Ln4(a) + ( / — a)Ln4(5)] + 4 |aLn(^) Ln3(a) + 8Ln Q) Ln3(5)] — 12 / Ln2(a)Ln2 (5)} / {[ / Ln(5) — 8Ln( / )] ^4Ln3(a) + Ln (-) [12Ln2( / 3) — n2]} + [8Ln( / 3) — / 3Ln(8}] ^4Ln3(a) + Ln (-) [12Ln2( / ) — n2]} + [(^Ln(x) — xLn(P)] (4Ln3(a) + Ln (-) [12Ln2(8) — 7(2|] + 6[Ln2(8)—Ln2(a)]{[P — 2025205047   02 Jul 2025 6]Ln2(x) + [5 - / ] / ,772( / ?)} + 2Ln2(8){[x - / 7][ / ,T72(5) - 3in2(a)] + 2Ln(5)[(5 - / 7) / ,77( / ) + ( / - 5) / ,77( / 7)]}}

[0050] r] = -{[(3Ln(a) - aLn(P)} {4 / ,773 (5) - Ln (-) [12Ln2(8) - re2]} + [(3Ln(8) -8Ln( / 7)] {4 / ,773 (a) + Ln (-) [12Ln2(a) - re2]} + [5Ln(a) - aLn(8)][12Ln2( / 3) - re2]Ln (-) + 6[Ln2(a) - Ln2(5)](5 - a)Ln2( / 7) + 6(a + 8 - 2 (3) Ln2 (a) Ln2 (8) + 2[( / 7 - 5)Ln4(a) + ( / 7 - a)Ln4(5)] + 4[8Ln (-) Ln3(8) + aLn (-) Ln3(a)]} / {[ / / ,77(5) - 5 / ,77( / )] }4Ln3(a) + Ln (-) [12Ln2 ( / 7) - re2]} + [8Ln((3) - (3Ln(8)] [4Ln3(a) + Ln (-) [12Ln2(x) - re2]} + [ / 3Ln(x) - / / ,77( / 7)] }4Ln3(a) + Ln (-) [12Ln2(8) - re2]} + 6[Ln2(8)-Ln2(«)]{[ / ? - 5] / ,772( / ) + [5 - / ]Ln2( / 7)} + 2Ln2(8){[x - £][ / ,772(5) - 3Ln2(a)] + 2Ln(5)[(5 - ^Ln()) + ( / - 8)Ln^])]}}

[0051] <p = {[xLn(3) - / ? / ,77( / )]{4 / ,t73(5) - Ln(-)[12Ln2 (a) - re2]} + [ / 3Ln(a) -aLn( / ?)]{4 / ,773(5) - Ln(-)[12 / ,772( / ) - re2]} + [aLn(x) - / / ,77(a)]{4 / ,773(5) - Ln(-)[12 / ,772( / 7) -re2]} + 6[Ln2(a) - Ln2(5)][( / - a)Ln2( / 7) + (a - / 7) / ,772 ( / )] + 2Ln2(a){[x - / 3][3Ln2 (5) -Ln2(a)] + 2Ln(a)[((3 - a)Ln(x) + (a -x)Ln())]}} / {[ / / ,77(5) - 5 / ,77( / )] }4Ln3(a) + Ln (-) [12Ln2 ( / 7) - re2]} + [8Ln((3) - (3Ln(8)] [4Ln3(a) + Ln (-) [12 / ,772( / ) - re2]} + [ / 3Ln(x) - / / ,77( / 7)] }4Ln3(a) + Ln (-) [12Ln2(8) - re2]} + 6[Ln2(8')-Ln2(«)]{[ / ? - 5] / ,772( / ) + [5 - / ] / ,772( / ?)} + 2Ln2(8){[x - (3][Ln2(5) - 3Ln2(a)] + 2Ln(5)[(5 - ))Ln(x) + ( / - 5) / ,77(6)]}}

[0052] k = -6re{[ / 7Ln(a) - aLn(6)][ / ,772(8)-Ln2( / )]+[ / ^71( / 7) - (3Ln( / )][L772(5)-L772(a)]+[aL77( / ) - xLn(a')][Ln2(8)-Ln2(l3)]+[j3Ln(8) -8Ln(6)][ / ,772 ( / )- / ,772 (a)]+[5 / ,77(a) - aLn(8)][Ln2 (x)-Ln2 (3)] + [ / / ,77(5) -8Ln( / )][ / ,772(a)- / ,772(6)]} / {[ / / ,77(5) - 5 / ,77( / )] {4 / ,773(a) + / ,77(-) [12 / ,T72( / 7) - re2]} + [5 / ,77(6) - / 7 / ,77(5)] {4 / ,773 (a) + Ln (-) [12 / ,772 ( / ) - re2]} + [ / 7 / ,77( / ) - / / ,77( / 7)] ^4 / ,773 (a) + Ln (-) [12 / ,772 (5) - re2]} + 6[ / ,772 (5)- / ,772 («)]{[ / ? - 5] / ,772( / ) + [5 - / ] / ,772( / ?)} + 2Ln2(8){[x - P][ / ,772(5) - 3Ln2(a)] + 2 / ,77(5)[(5 - ))Ln(x) + ( / - 5)177(6)]}}

[0053] Thus, for example, if a = 2, / 7 = 4, / = 16,5 = 64, then y = -1.651,7] = 1.246, <p = -0.2396, K = -0.1612. Note, as the frequency domain resistive components are particularly sensitive to soil conductivity, there is typically no point in nulling out the log-quadrature VRM component if the soil conductivity is also not nulled out (simultaneously); hence the inclusion of this constraint above. 2025205047   02 Jul 2025

[0054] A block diagram of a metal detector for implementing one embodiment of the present disclosure is shown in Figure 3. A transmitter inductor 20 transmits a magnetic field 23 into soils 22. A received magnetic field 24 from the soils 22 is received by a receiver inductor 21, that is connected to an input of a receive preamplifier 25. The transmit inductor 20 is energised by current flowing from transmit electronics 26, that is controlled by a master clock and timer 28 via control line 29. An output 27 from receive preamplifier 25 is connected to inputs of synchronous demodulators 31. The forward gains of these synchronous demodulators 31 are controlled by the master clock and timer 28 via connections 30 that are synchronised to the transmit magnetic field 23. The outputs 32 of the synchronous demodulators 31 are fed to the inputs of filters 33, that include at least low-pass filtering to remove transmit related frequencies and limit the noise and audio modulation control input bandwidths. Outputs 34 of the filters 33 are yet further processed in the further processing modules 35, which may include, for example, such processes as the well-known “automatic ground balancing.” Outputs of the further processing modules 35 also control an indicator output provided to the metal detector operator, such as for example, modulated audio that is provided to headphones 37 via headphone cables 36. Any other modification known to a person skilled in the art may be applied to the block diagram of Figure 3.

[0055] Figure 4 shows as graph 41 the normalised first-order target frequency response on X axis versus response for the above multi-frequency example, where the said frequencies are in units of kHz. Here target frequency in Hz = 1 / (2n * target time-constant).

[0056] Field tests have shown the simultaneous cancellation of log-uniform VRM, plus log-linear VRM, plus log-quadratic soil VRM distributions clearly affords a net target-signal-to-residual-soil-noise advantage for the location of a range of metal targets in soils with the more significant log-quadratic VRM present.

[0057] In this specification the terms “ground” and “soil” are used interchangeably. As understood by a person skilled in the art, the terms “ground” and “soil” mean surfaces of earth where targets may be contained within. The surfaces are often solid, may be homogenous or may be a combination of various soil types, and may contain moisture or water.

[0058] Those of skill in the art would understand that information and signals may be represented using any of a variety of technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips referenced throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof.

[0059] Those of skill in the art would further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly 2025205047   02 Jul 2025 illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure.

[0060] The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. For a hardware implementation, processing may be implemented within one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, micro-controllers, microprocessors, other electronic units designed to perform the functions described herein, or a combination thereof. Software modules, also known as computer programs, computer codes, or instructions, may contain a number of source code or object code segments or instructions, and may reside in any computer readable medium such as a RAM memory, flash memory, ROM memory, EPROM memory, registers, hard disk, a removable disk, a CD-ROM, a DVD-ROM or any other form of computer readable medium. In the alternative, the computer readable medium may be integral to the processor. The processor and the computer readable medium may reside in an ASIC or related device. The software codes may be stored in a memory unit and executed by a processor. The memory unit may be implemented within the processor or external to the processor, in which case it can be communicatively coupled to the processor via various means as is known in the art.

[0061] Throughout the specification and the claims that follow, unless the context requires otherwise, the words “comprise” and “include” and variations such as “comprising” and “including” will be understood to imply the inclusion of a stated integer or group of integers, but not the exclusion of any other integer or group of integers.

[0062] The reference to any prior art in this specification is not, and should not be taken as, an acknowledgement of any form of suggestion that such prior art forms part of the common general knowledge.

[0063] It will be appreciated by those skilled in the art that the disclosure is not restricted in its use to the particular application described. Neither is the present disclosure restricted in its preferred embodiment with regard to the particular elements and / or features described or depicted herein. It will be appreciated that the disclosure is not limited to the embodiment or embodiments disclosed, but is capable of numerous rearrangements, modifications and substitutions without departing from the scope of the disclosure as set forth and defined by the following claims.

Claims

1. A method to detect a target in soil using a metal detector, comprising:transmitting a transmit magnetic field using a transmitter;receiving a receive magnetic field using a receiver to produce a receive signal;processing the receive signal to produce a target indicative output; andproviding, based on the target-indicative output, an indicator output indicating the presence of the target;the processing comprises:processing respective resistive components of the receive signal at first, second, third and fourth different frequencies;processing a reactive difference component of the receive signal associated with two of the first, second, third and fourth different frequencies; andcombining the respective resistive components and the reactive difference component such that the target-indicative output is simultaneously nulled to components of the receive magnetic field due to log-uniform distributed VRM particles, log-linear distributed VRM particles, log-quadratic distributed VRM particles and soil conductivity.

2. A method to detect a target in soil using a metal detector, comprising:transmitting a transmit magnetic field using a transmitter;receiving a receive magnetic field using a receiver to produce a receive signal;processing the receive signal to produce a target-indicative output; andproviding, based on the target-indicative output, an indicator output indicating the presence of the target;wherein the processing comprises synchronously demodulating the receive signal using a timedomain synchronous demodulation function comprising four demodulation windows having respective gains, the time-domain synchronous demodulation function being configured such that the targetindicative output is simultaneously nulled to components of the receive magnetic field due to log-uniform distributed VRM particles, log-linear distributed VRM particles and log-quadratic distributed VRM particles of the soil.

3. The method of claim 1 or 2, wherein the step of processing the receive signal is configured such that, in response to a transmitted single magnetic ramp of duration followed by a constant magnetic field of a same value as that at a termination of the magnetic ramp, the output comprises an output signal that is simultaneously nulled to rates of change of components of the receive magnetic field in following forms across a processing bandwidth of the handheld metal detector: Lh[(t + t) / t]; Zti2(t + t) — Zn2(t);In3(t + t) — / n3(t); and a constant value; where t = 0 at the termination of the magnetic ramp.2025205047   31 Jul 20264. The method of claim 1 or 2, wherein the step of processing the receive signal is configured such that, inresponse to a transmitted magnetic isolated half cosine wave connecting a constant transmitted magneticfield to a different constant transmitted field, the output comprises an output signal that is nulled to ratesof change of components of the receive magnetic field in following forms across a processing bandwidthof the handheld metal detector: fj cos   In3 (x + )')dXw where the transmitted half-cosine pulse is of theform H = a when t<=-T, H = b when t>=0, H = a + (b — a) coswhen -T<t<0, with a receiveperiod commencing at t=0 at the termination of the said half cosine-wave.

5. The method of claim 3, wherein the transmit magnetic field comprises at least one current ramp period, and (i) at least one period of zero transmitted magnetic field or (ii) at least one constant current period.

6. The method of claim 4, wherein the transmit magnetic field comprises at least one half-cosine pulse, and (i) at least one period of zero transmitted magnetic field or (ii) at least one constant current period.

7. A metal detector configured to perform the method of any one of claims 1 to 6.

8. A non-transitory computer readable medium, comprising instructions to perform the method of any one of claims 1 to 6.

Images