JP2966933B2 - Coin inspection method and device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a coin inspection method and apparatus.

In this specification, the term "coin" is used to encompass genuine coins, tokens, counterfeit coins, and any other things that can be used in attempting to operate a coin-operated device.

The coin is tested by measuring the degree of interaction between the coin and the magnetic field through the coin in a path into the oscillating magnetic field created by the inductor, and the resulting measurement depends on one or more characteristics of the coin. Coin inspection devices that compare each of a reference value or a set of reference values corresponding to measured quantities obtained from one or more denominations of acceptable coins are well known. Apply one or more such tests, where each test responds to different coin characteristics, and accept a tested coin only if all test results are appropriate for one acceptable denomination of the coin It is most common to judge. An example of such a device is
GB-A-2 093 620.

At least one of the tests is usually sensitive mainly to the material from which the coin is made,
Such testing may depend on the electrical conductivity of the coin material and, in the case of magnetic materials, the magnetic conductivity of the coin material. Said inspection arranges for the coin to cross the front of the inductor and thus in its oscillating magnetic field, and the effect of the coin on the frequency or amplitude of the oscillator, of which the inductor forms part, by approaching the inductor. Was performed by measuring the Mostly, the peak value of the effect obtained when the coin was centered on the inductor was measured.

However, this type of measurand is sensitive to the distance between the coin and the inductor in a direction perpendicular to the front of the inductor at the time the measurand is made. This undesirable effect can be countered to some extent by designing the mechanical design of the device so that the coin is always encouraged to cross the inductor at a fixed distance from the inductor, but this is completely achieved. Requires design features that cannot be done and can be otherwise objectionable. The measurand scatter caused by changing the lateral position of the coin can be accounted for by setting a wider coin acceptance range so that acceptable coins can be placed at different distances from the inductor. Even if it crosses it, it will always pass the inspection, but this in turn affects the reliability of the device in rejecting unacceptable coins. Also, by utilizing the combination effect of the two inductors arranged on each side of the coin passage, at least to some extent
It is known that the effect of the displacement of the coin position between the two inductors can cancel each other out, but this involves the provision of a second inductor.

One object of the present invention is to provide a coin inspection method that is responsive to the coin material and relatively insensitive to the distance of the coin from the inspection inductor.

The present invention provides, from one aspect, a coin inspection method in a coin inspection device, wherein a step of exposing a coin inserted into the device to an oscillating magnetic field generated by an inductor, and a reactance of the inductor when the coin exists in the magnetic field. And the step of measuring the loss, and the direction in the impedance plane of the displacement line representing the displacement of the coin present point limited by the measured amount related to the coin absence point representing the reactance and loss of the inductor when the coin is absent, Checking whether the directions match each other.

The "impedance plane" referred to above is the reactance (reactive impedance) of the circuit or inductor
And the loss (resistance impedance) is in the plane 2
A plane represented as a measurand or vector along two mutually perpendicular axes. The term "displacement line" will be explained later in connection with FIG.

An example will be described in which a free running oscillator is used to make inductance and loss measurements. However, another preferred embodiment uses a phase discrimination method, which avoids the need to use large capacitors and allows all timing phases of the measurement network to be determined by the clock of the microprocessor, and This simplifies operation.

Since the direction of the displacement line is substantially independent of the lateral position of the coin, the present invention can be implemented using only one inductor. This simplifies the necessary electrical wiring,
A typical coin device where the coin path is between the body and the releasable lid avoids the need to provide flexible wiring through an inductor attached to the lid.

In some of the described embodiments, it will be apparent that the reference direction in the impedance plane is set as an angle with respect to one of the reactance and loss axes.

The reactance of the coil itself and the loss of the coil itself are
The location of the coin-absent point in the impedance plane may not be constant, as it may change with temperature, and as a result may change with time, and may cause slight deformation of the coin device structure. .

For this reason, the reactance and loss of the inductor are measured both with and without the coin in the magnetic field. The direction of the displacement line is determined by the two points at which the measurand is taken. In particular, two reactance measures are subtracted, two loss measures are subtracted, and the ratio of the two differences is taken, which represents the tangent of the angle that the displacement line makes with one of the axes. .

This tangent can then, of course, be set and stored as the corresponding angle tangent for the acceptable coin, expressed as a number in digital form when used for performing digital processing and storage. .

The movement of the coin absent point in the impedance plane may be inconsequential, or if possible, steps may be taken by compensation techniques to prevent such a shift. In such a situation, instead of the angle-only reference information, it may, for example, draw a reference displacement line whose direction is the reference direction and whose position extends through a substantially absent coin absent point. 1 in the impedance plane, limiting together
A set of stored coordinates may be constructed. Next, checking whether the direction of the displacement line matches the reference direction need not include actually measuring the coin absent point. It can be assumed that the point has not changed, and as a result, by checking whether the coin present point is on the reference displacement line, 2
You can easily see if the two directions match or vice versa. If so, the coin would have caused displacement of the coin absent point in the direction of the reference displacement line.

In another aspect of the invention, the reference direction is set as an angle with respect to the total impedance vector in the absence of the coin of the inductor, instead of with respect to the loss or reactance axis. This consists of specific values when the reactance and loss measurements are made in a phase discriminating manner, as explained below. The use of the phase identification method has the advantages described above, but can also introduce errors due to the reference signal used not being correctly phased. Measuring the direction of displacement of the impedance plane created by the coin relative to the total impedance vector of the inductor and setting the reference direction as an angle relative to the total impedance vector also reduces or eliminates such errors.

From another aspect, it is checked whether the direction of the displacement line matches the reference direction in the impedance plane suitable for the particular coin type, whether or not the phase identification method is used to determine the direction of the displacement line. Was made, and
It is determined whether the difference between the coin absence value and the coin presence value of the reactance of the inductor matches a reference value suitable for the same specific coin type.

This additional inspection allows the identification of different coin types by diameter. As for the diameter of the coin, the direction of the displacement line on the impedance plane has a property that the sensitivity is very poor.

In the preferred embodiment to be described, the direction of the displacement line is calculated from the signal ratio. Since the ratio is taken, this result is independent of the gain of the path handling the relevant signal. However, if it is also desired to use the difference between the reactance in the presence of the coin and the reactance in the absence of the coin as an acceptance criterion, the gain of the path becomes important.

Another feature of the present invention that can be used independently of whether the measurand is taken using a phase identification method is that it sometimes simulates the expected change in reactance of the inductor when the coin is not in its magnetic field. Thereby compensating for the effect of a change in system gain due to the difference between the reactances, and detecting a resulting change in the signal affected by the reactance and affected by the system gain. Comparing the change with a reference value, applying the compensation coefficient obtained from the result of the comparison to the reactance-dependent signal, and adjusting the signal to substantially match the reference value; Maintaining the application of said compensation factor until the next time the change is simulated.

From yet another aspect, the present invention is a coin inspection method in a coin inspection device, comprising exposing a coin inserted into the device to an oscillating magnetic field generated by an inductor; and reacting the inductor when the coin is in the magnetic field. And the step of measuring the loss, and the direction of the displacement line in the impedance plane representing the displacement of the coin present point limited by the measured amount with respect to the coin absence point representing the reactance and loss of the inductor when the coin is absent is the reference in the impedance plane. The process of checking whether the directions match, the frequency of the oscillating magnetic field generated by the inductor is
A coin inspection method is provided wherein the skin depth for the coin material is sufficiently low that it is greater than the coin thickness, whereby the direction of the displacement line is affected by the thickness of the coin being inspected.

Such a method can also be used whether or not the reactance and loss measurements are taken by a phase identification method.

Another aspect of the invention is a coin inspection device for performing the method according to the invention as mentioned above.

In order that the invention may be more clearly understood,
Next, an embodiment of the invention will be described as an example with reference to the accompanying schematic diagrams below.

FIG. 1 represents the impedance plane for the inductor of the coin inspection device shown in FIG.

FIG. 2 schematically illustrates a circuit for generating X and R signals using a phase identification method.

FIG. 3 is another impedance plane diagram useful in explaining the operation of the circuit of FIG.

FIG. 4 shows how X and R change over time as the coin traverses the inductor.

FIG. 5 shows how the angle θ changes over time as the coin traverses the inductor.

FIG. 6 is another impedance plane diagram useful in explaining a coin inspection method further developed by the present invention.

FIG. 7 illustrates a significant portion of the circuit that is the same as the circuit of FIG. 2, but includes additional features.

FIG. 8 is another impedance plane diagram useful in understanding the function of the circuit of FIG.

FIG. 9 is another impedance plane diagram that is helpful in understanding the effects of the offset applied in the circuit of FIG.

FIG. 10 shows how the angle θ measured in the impedance plane varies with thickness and frequency when the measurand is taken for multiple test disks of the same material but of different thicknesses. It is a graph.

FIG. 11 schematically illustrates another coin inspection device using the present invention, which is deployed using a free-running oscillator instead of a coil in which the X and R signals are driven.

FIG. 12 shows the relationship between frequency, phase and effective resistance in the tuning circuit of FIG.

In FIG. 1, the vertical axis represents the imaginary component or reactance X of the impedance of an inductor, such as the coil 104 of the device shown in FIG. 2, as affected by any coins that may be near the coil. The horizontal axis also represents the real component of impedance, ie its resistance or loss R, as affected by any coin that may be near the coil.

If X and R are measured when the coin is not close to the coil, the resulting value will be characteristic of the coil only and will limit point a in the impedance plane (the plane represented in FIG. 1).

Then, if the coin was carried close to the coil, the effective reactance and loss of the coil would both change, ie, if X and R were measured for the coil plus the coin, the resulting The value will define a different point b in the impedance plane.

If the coin goes back and forth vertically to the front of the coil at its center position relative to the coil, point b moves along a substantially straight line ab.

As a result, the same coin passes several times through the same device,
If in each case the X and R values are measured when the coin is centered with respect to the coil but each time at a different distance from the coil, the resulting X and R measurement amounts to three points b, Limit c and d.
Although the X values for these points are all different, and consequently the R values are different, each set of these values will limit the points on the same line ab.

Over time, due to aging of circuit components, the effects of temperature changes, or deformation of the physical shape of the device, the line a
The position of -b may move in the impedance plane, for example to a parallel position a'-b ', but this gradient or angle θ remains the same for coins of the same type. In other words, the direction of the line in which the point representing the coin / coil combination on the impedance plane has moved relative to the coil-only point (herein referred to as the "displacement line") indicates the type of coin and substantially the coin Be independent of lateral position.

Thus, a reference value for θ, which is a characteristic of a particular acceptable type of coin in a particular coin inspection device, can be set, and then the value of θ for an unknown coin is measured in the same device. In that case, a comparison between the measured value of θ and the reference value will give an indication of the acceptability of an unknown coin, as long as the coin material properties affecting θ are concerned. This indicator is independent of the distance each coin has traversed the coil, and is independent of time-varying factors that do not cause a change in the angle θ for acceptable coin types.

If the coin contains high magnetic permeability magnetic material, the loss is
Increased by the additional factor of hysteresis loss, reactance may increase instead of decrease. Because the coin will act to some extent as a core for the coil. In such a case, the angle θ will be opposite to the angle shown in FIG. This can be used to distinguish between magnetic and non-magnetic coins.

The above method over the prior art in which the measured X and R values are individually compared to a reference has yet another advantage. The criterion is usually not a specified value, but an upper limit and a lower limit that limit a certain range. If a different measurement is compared to each reference range, the coin will be accepted if each measurement is anywhere within its respective reference range. For example, if the measured quantities were X and R measured quantities as described above,
Even if both the X and R measures are at the limits of each range, i.e., the combination of these measures is likely to be the result of a coin that should not actually be accepted, the coin is accepted. Will be paid. In this way, a coin whose X metric would be at the limit of an individual reference range for X, would have its R metric displaced in one direction from the center of the reference range for R but not in the other direction. Will only be accepted if The latter indicates that this particular combination of X and R measurements suggests that the coin should be rejected even if the coin was accepted using the prior art.

In the device that will be described, the values of X and R are measured when the coil is not present, then when the coin is near the coil, both X values are subtracted and both R values are subtracted.
The values are subtracted, resulting in ΔX and ΔR shown in FIG.
Is generated. This value indicates how much the coin has changed the effective reactance and loss of the coil, and then △ X / △ R is taken. That is, this is the tangent θ for the unknown coin. Acceptability, this
Inspection is performed by comparing with a reference value of tangent θ corresponding to the ratio of the measured values of ΔX and ΔR for the acceptable coins.

Next, the apparatus of FIG. 2 will be described in detail. Means are provided for positioning a coin, shown in dashed lines, at 10 near the coil 104. This means is schematically shown as a coin path 12 along which the coin passes the coil bordering the edge. A practical arrangement for passing a moving coin near an inductive test coil is described, for example, in GA-A-209.
3 620. As the coin 10 passes through the coil 104, the total effective loss of the coil increases, peaking when the coin is centered relative to the coil, and then decreasing to a rest level. The total effective reactance decreases to a negative peak and then returns to its rest level. In this embodiment, the device uses the peak value.

The circuit of FIG. 2 has a real component (R) of the coil impedance.
And the imaginary component (X) is separated using a phase identification method. Circuit, including a signal source composed of a digital frequency generator 100, the output of which is filtered by a filter 102, the output of the filter 102 controls a constant current source 103, and the output of the constant current source 103 controls a coin detection coil 104. Drive. Thus, components 100, 102, 103 appear to the coil as constant current sources. The output of the generator 100 is close to a sine wave but is digitally generated, which includes harmonics, but the function of the filter 102 is to filter them out.

The signals at both ends of the coil 104 are applied to a phase sensitivity detector 106 to which two reference signals are similarly input from the generator 100. One reference signal is on signal line 108, ideally in phase with the voltage across coil 104 so that the phase sensitivity detector can generate a signal representing X at one of its outputs. I have. On the other signal line 110, a 90 ° relative to the initial reference signal so that the phase sensitivity detector can generate a signal indicating the R of the coil at its other output.
A reference signal that is ゜ and is in phase with the coil current is applied. The voltage signal applied to the phase sensitivity detector and the voltage signal output therefrom can only be trusted as a reference for X and R while the peak coil current remains constant over time. It should be noted.

The R and X signals are filtered by respective filters 112 and 114, and the resulting signal is applied to a microprocessor 116, which performs further processing of the signals and performs other functions required for coin inspection. Is programmed as Further, the microprocessor 116 controls the signal generator 100 to alternately generate a reference signal on signal lines 108 and 110, and also controls the phase sensitivity between the R and X output paths in synchronization with the switching of the reference signal. The output of the detector 106 is also switched.

Referring to FIG. 3, the vector 118 represents the total impedance of the coil 104 when no coin is present, so that its end corresponds to point a in FIG. When the passing coin is at the center of the coil, vector 118 moves along displacement line 120 to become vector 118 '. The end of the vector 118 'corresponds to the point b, c or d in FIG. Microprocessor 116 receives as input signals representing the X and R components of both of these vectors from phase sensitivity detector 106, and thus obtains △ X and △ R and the tangent θ as previously mentioned.
And their ratio △ X / △ R can be calculated.

Since the angle θ is calculated from the difference between the X value and the difference between the R values, any inadvertent offsets applied in the circuit to the signals representing X and R will not cause errors. This is because they make the difference value unaffected.

Although the inductor is shown as a single coil, it may have other forms, for example, a pair of coils opposed across the coil path and connected in a parallel or series forward or reverse connection.

FIG. 4 shows how, for a single coin, X and R (both measured in ohms) change over time as the coin traverses the coil. ΔX and ΔR are also shown. X reaches a relatively smooth and flat negative peak during the middle portion of the coin passage, while R indicates a relatively smooth plateau at the center of the peak and each end of the plateau. It can be seen that it has a small peak further superimposed on it. Note that these small peaks are caused by the edge effect when the edge of the coin passes through the center of the coil.

As the coin traverses the coil, the locations of the points defined by the X and R values in the impedance plane are shown by the key curves in FIG.

In the plane, prior to arriving coins i.e. at time t _1, X-R coordinate point is at the top of the key of FIG. 5, which corresponds to a point in FIG. When in the center with respect to the coin in the coin reaches the time t _3, X-
The point limited by the R measurement moved to the tip of the key. This corresponds to point b in FIG. The presence of a small added peak at the beginning of the main peak of the R-measurement causes the point to draw the bulge of the key in FIG. 5 as the coin moves toward the center position. As the coin moves away from the coil to move from the central position, t ₄ to t ₅ the points from t ₃
Turn around until you return.

It will be seen that the vector 120 from the coin absent point to the point defined by the current XR measurement of the moving coin will extend and rotate until it reaches the tip of the key, and then perform the reverse movement.

From this, △ X and △ that occur in the middle during the coin passage
The variable value of R is stored, the corresponding time-varying value of △ X / △ R (that is, tangent θ) is calculated, then the maximum value of the calculated value of tangent θ is detected, and this maximum value can be accepted. It will be appreciated that the calculation can be performed by comparing to a reference value of tangent θ for the coin.

As described, it is preferable to take a measurand for moving coins so that coins can be inspected quickly and continuously, but it is also possible to measure loss and reactance with stationary coins. is there.

The advantages of driving the coil as shown in FIG. 2 are that a large capacitor is not required when compared with the method using a free-running oscillator, and that all signals in the detection network can be synchronized with the clock frequency of the microprocessor. This is a considerable simplification. However, the phase of the reference signal via signal lines 108 and 110 (defining the phase identification axis) is incorrectly related or related to the phase of the current in coil 104 (defining the true R and X axes). If so, it may be possible to make the phase identification method of FIG. 2 a little less accurate than ideally desired.

The relative precision of these phases is higher than the digital generator 10
This is possible because it is limited by a resolution of 0 and the analog filter 102 itself introduces an unidentified phase delay into the signal applied to the coil 104 that can vary with temperature. The effect of the phase error is that the components of the total impedance vectors 118 and 118 'of FIG. 3 will be measured with respect to the identification axes X _d and R _d rotated with respect to the true reactance and loss axes. is there. Thus, the calculated value △ X _d is greater than the desired true value △ X, and the calculated value △ R _d is less than the desired true value △ R. Their ratio △ X _d / △ R
_d is the angle θ that was going to be measured, as seen
Become tangent of a larger angle θ _d. To leave it to another method, the angle θ is measured, but with an error of a magnitude determined by the angle error of the phase identification axis.

One method for eliminating this will be described with reference to the impedance plane diagram shown in FIG. This is consistent with FIG. 3 except that for ease of understanding, the angularly displaced identification axes X _d and R _d are shown as solid lines and the true X and R axes are shown as dotted lines. . It is important to note that the error in the identification axis is the total impedance vector 118 when there is no coin, the total impedance vector 118 'when there is a coin, and the vector 11 for the end of the vector 118.
That is, the shape of the triangle formed by the displacement line 120 representing the end point displacement of 8 'is not changed. Its shape and, therefore, the interior angle indicated by C, is determined only by the length and direction of the two total impedance vectors 118 and 118 ', which are independent of any phase error.

The measurands taken for the identification axes X _d and R _d can be used to derive the angle C as follows. It should be noted that angle C is equal to the sum of angles A and B as shown in FIG. FIG. 6 shows that R _d / X _d is the tangent of angle B, so that angle B can be calculated from those measurements. Also, the tangent of the angle A is △ R _d / △ X _d , and as a result, the angle A can be calculated from their difference values. Thus, instead of trying to measure its direction with respect to the true R and X axes, which may introduce errors due to unknown phase errors in the phase identification method as described, the axis against which the direction of the displacement line 120 is measured By using the vector 118 as a reference, a coin test criterion is reached that is independent of both the lateral position of the coin with respect to the test coil and any phase errors that may be present in the circuitry used in the phase identification method.

Given angles A and B, the product of both tangents is 1
It can be shown that the tangent of angle C is simply △ R _d / △ X _d plus R _d / X _d (which would be the case in practice). Thus, in these situations, the process is simplified by measuring the direction of the displacement line 120 by the sum of the tangents of the angles A and B.

In general, if the angles referred to herein are sufficiently small, they may correspond to an acceptable degree of accuracy over their tangents, and in these situations the terms "tangent" and "angle" each include the other It should be understood that

In FIG. 7, the constant current source is in the form of a transistor 103 and related components. Additional components when compared to FIG. 2 are a calibration and offset circuit, indicated generally at 130, and the X and R signals taken from the lower end of coil 104 prior to amplification for phase sensitivity detector 106. 132, and a second offset circuit 134
And outputs of the filters 112 and 114 to the microprocessor 11
A digital-analog converter 136 for converting to a digital format for handling in 6. One filter or integrator 112/114 is shown in FIG.
Are equivalent to the two circuits 112 and 114 shown separately. In practice, the analog-to-digital converter 136
It would be preferable to use a microprocessor that actually incorporates

At this stage, the X and R signals are simply components that are 90 degrees out of phase with the in-phase component of the coil voltage signal, respectively.
It will be appreciated that the output signal from coil 104 is always amplified by preamplifier 132. Thus, preamplifier 132 serves as a common path for both the X and R signals. The phase sensitivity detector 106 generates an X signal from the output thereof when an in-phase reference signal (with respect to the coil voltage) is applied to a signal line 108 and a 9
The X signal and the R signal are separated by generating an R signal when a reference signal having a phase different from 0 degree is applied. As a result, from the phase sensitivity detector 106 to the microprocessor 116
Circuit components have served as a common path for the X and R signals, but always deal with only one or the other of them.

A first important function of the network of FIG. 7 is another way of dealing with the problems that arise with the angular displacement of the phase discriminating axis with respect to the true X and R axes, namely the angle between the displacement line 120 and the total impedance vector 118. C is affected by the error angle θ _d
It is to provide an alternative to the method previously described with respect to FIGS. 3 and 6 calculated instead.

The first step is to measure the phase error angle θ _e (see FIG. 3) by the method described below. theta _e that is the difference of erroneous angle theta _d desirable angle theta can be seen from Figure 3. Once θ _e is known, one or both of the two steps can be employed. First, the microprocessor 116 determines that the phase of the reference signal on both signal lines 108 and 110 is θ
Digital generator 100 can be adjusted to shift _e in a direction that reduces _e to zero. This would not normally be possible. This is because the generator 100 is a digital, the output phase is only can be adjusted stepwise, therefore normally, there is a residual value of theta _e that can not be erased by the adjustment. However, theta _e is because also measured found to residual value can be subtracted from the measured angle theta _d erroneous it get a true value theta. It is of course possible to reduce as much as possible the value of the angle theta _e is desirable the adjustment. This makes the simplicity assumption that the angle and its tangent are equal, as described above, more accurate. Next, a method of measuring _θe will be described with reference to FIG.

The principle is that the operation of the calibration and offset circuit 130 simulates the displacement of the reactance of the coil 104 when there is no coin in the magnetic field. If the phase error angle θ _e is zero and the X component of the coil impedance vector 118 changes without changing its R component, then any of the R components as read or measured from the output of the phase sensitivity detector 106 It can be seen from the study of FIG. 3 that there will be no change. However, rather than the phase error angle theta _e is 0, if the axis R _d does not coincide with the axis R in FIG. 3 therefore will change in the R value as measured along the axis R _d is present.

This can be better understood with reference to FIG. FIG. 8 shows that if the simulated change .delta.X _d is added to the X component of the total impedance vector 118 and replaces it with the vector 118 ", there is no change in that R component measured along the true R axis. shows the state. However, when the phase discrimination axes X _d and R _d is incorrect angle theta _e cliff as before, as measured on the axis R _d, there is a change &Dgr; R _d in the measured R value it was found. Further, δR _d / δX _d is that a tangent of the angle theta _e also can be readily seen from FIG.

The calibration and offset circuit 130 of FIG. 7 simulates the change in the X component of the coil impedance to make sure that this simulation does not affect the R component of the coil, and then measures from the output of the phase sensitivity detector 106. relationship between the change in X measured amount of change in R as is used as a basis for calculating the error angle theta _e.

In the normal operation mode of the calibration and offset circuit 130, the transistor T2 is off and the transistor T1 is on.
Next, the current flowing through the coil 104 is divided into series resistances Rb and Rc on the one hand and parallel resistance Ra on the other hand. These are all precision resistors. In the circuit of FIG. 7, it must be remembered that it is a voltage component across the coil 104, which is taken as a measure of the coil loss R and is in phase with the current flowing through the coil 104. This is a true expression only while the magnitude of the coil current is constant. It is a measure of the coil reactance X, which is 90 ° out of phase with the coil current.
4 is the value of the voltage component between both ends. In fact, this latter voltage is connected by a tap between the resistors Rd and Rc to obtain a voltage in phase with the coil current, and the phase of the voltage divided by the tap connection is changed by 90 ° by the capacitor Ci. Applying the resulting phase-shifted voltage to the input of preamplifier 132 has an offset added to it for reasons explained below. This offset voltage is
Is 180 ° out of phase with the imaginary component of the voltage between both ends, that is, the reactance-related component.
, Only adds a fixed offset to the voltage component corresponding to the coil reactance X. This offset voltage is AC and is essentially phase adjusted so as not to affect the loss related components of the input voltage to the preamplifier 132.

To measure the phase error, transistor T2 is turned on and inserts a precision resistor Rd in parallel with resistor Rc, thereby reducing the tap divider voltage provided through capacitor Ci. This voltage decrease simulates a decrease in the reactance X of the coil 104, ie, δX _d in FIG. 8, at the input of the amplifier 132. However, if it were only done, the coil current would increase as the total resistance in series with coil 104 was reduced. To compensate for this and keep the coil current constant, turning off transistor T1 disconnects resistor Ra. Resistance Ra
Is chosen to keep the coil current constant, so that the simulation of the change in X is essentially arranged so that coil loss R is not simulated as well, ie, represents R. Therefore, the condition required for shifting the voltage between both ends of the coil 104 by 90 ° is maintained. Next, if there is a change in R measured by microprocessor 116 from the signal output from preamplifier 132, that change will result in a phase discriminating axis being displaced relative to the R and X axes. ΔR _d in FIG.

If the resulting angle, having the calculated θ _e or at least the tangent θ _e as △ R _d / △ X _d , is large enough for the minimum adjustment that can be applied to the digital generator 100, the microprocessor 116 , Instruct the digital generator 100 to adjust it to the extent that it reduces the phase identification error. When the measured error angle is such that less than the minimum adjustment step, microprocessor 116 obtains a desired angle theta to the coin testing by adding it with the measurement value theta _d. theta _e was positively or negatively, it will be seen that it is possible to result added to either increase or decrease the measured value theta _d.

If necessary, the above calculation of θ _e may be adjusted, for example, by 3
Automatically under the control of microprocessor 116 at intervals of every second, but only when no coin is at the coil. After each time, the transistors T1 and T2 return to the normal operating state in which T2 is off and T1 is on.

Instead, the network does not simulate any change in X, and from the measured △ R _d and any resulting △ X _d measurements, θ _e or tangent θ
It can be adapted to simulate changes in R without calculating _e .

The second function of the calibration and offset circuit 130 has already been described briefly, but will now be described. It is the amplification of an offset voltage 180 ° out of phase with respect to the X component of the voltage across coil 104 at the input of preamplifier 132 via Ci.
The reason for this is that, in fact, X is much larger than R, typically about 30 times larger. Further, the changes △ X and △ R caused by the coin can typically be within 20% of the X and R coin absent values. Both the X and R signals must be processed in a common path of the preamplifier 132 and the phase sensitivity detector 106, and one signal is a very poor signal because it is about 30 times as large as the other signal. A noise ratio is obtained, which may make any meaningful extraction of the ΔR measurand impossible. There is an offset added to the X signal via the capacitor Ci, so that it makes the X signal at the input of the preamplifier 132 quadruple in magnitude with the R signal. Therefore, a greatly improved use of operational amplifiers
Made with 132 dynamic range, the signal-to-noise ratio can be acceptable.

It should be noted that the exact value of the offset voltage is not important as long as it is constant. This is because the offset voltage is added to both the X value when the coin is present and the X value when the coin is absent, and therefore does not cause any change in the angle θ or the difference ΔX used in calculating its tangent. Offset is preamplifier 132
Is not added to the R signal at the input of.

Although the calibration and offset circuit 130 has a third function, it is necessary to refer to yet another method used for coin inspection using the circuit of FIG. 7 before describing it.

The measurement of the direction of the displacement line in the impedance plane is
It has been described above that it is a good indicator of coin material and is substantially independent of the distance of the coin from the coil. This forms a valid coin check, but by itself is usually not enough to distinguish different types of coins, as different types of coins may be made of the same material.

Therefore, it is desirable to detect at least one other coin characteristic, and the diameter of the coin is a valid one. However, the direction of the displacement line (eg, angle θ) is
Even though the coil is made about the same or larger than the largest diameter coin to be tested, it is not sensitive enough to the diameter of the coin to provide a valid diameter test. Using the circuit of FIG. 7, and if the diameter of inductor 104 is about the same as or larger than the largest coin to be tested, the value of ΔX is effectively sensitive to the diameter of the coin, 2, it can be seen that a coin will only be accepted if its △ X value matches the △ X value of an acceptable coin of the same type as taking its displacement line direction. .

However, unlike the ratio of ΔX and ΔR, only the value of the ΔX signal depends on the system gain, which can be expected to change with time and temperature.

In order to compensate for the effect of the gain change on the measured value of ΔX, the calibration and offset circuit 130 operates periodically (for example, at the time of switching on or every few seconds) as follows. As described, transistor T2 is off during normal operation of the circuit. To calibrate the gain change, transistor T1 is also off, thus disconnecting resistor Ra from the circuit. Since this is in parallel with Rb and Rc, the total impedance increases and the current through coil 104 decreases. Since the three resistors Ra, _Rb and Rc are precision resistors, disconnection of Ra can be selected to cause a very precise constant percentage change in coil current, eg, 2%, repeatedly. As long as the X component of the coil voltage is concerned, this will appear as a 2% reduction in coil reactance. Of course, the system would be designed to operate at some desired overall gain from the coil 104 to the digital-to-analog converter 136. For example, the desired overall gain is 2 of the X component of the coil voltage.
% Change in the analog-to-digital converter output is 200
Let's assume that the count change is small enough. When T1 turns off causing a 2% change, the resulting change in count at the output of the analog-to-digital converter is checked by microprocessor 116. If it is 200, no operation is performed, but if it is not 200 but, for example, n, a compensation coefficient 200 / n is calculated.
Thereafter, transistor T1 is turned on again, returning the circuit to normal operation mode, after which ΔX is
Each time calculated at 116 (based on the count output of the analog-to-digital converter 136 for the X value in the presence and absence of coins, of course), the result is multiplied by a compensation factor of 200 / n, whereby the system gain This produces a ΔX value that compensates for the change. In short, the change in gain of the analog component is measured and then compensated by multiplication in the digital stage so that the gain between the output from the coil and the final calculated ΔX value is kept constant. .

The analog-to-digital converter 136 forms yet another common path in which the X and R signals are to be processed. As the coin traverses the coil 104, the X signal decreases and the R signal increases. In order to make the best use of the dynamic range or resolution of the analog-to-digital converter and / or to enable a converter with lower resolution and thereby less costly consumption, An offset is added to both the X and R signals so that the coin absent value of each signal is near the appropriate limit of the dynamic range of the analog-to-digital converter 136. These are DC offsets, added by the second offset circuit 134 under the control of the microprocessor 116,
Each has a different value, one for the time when the X signal is processed or obtained, and the other for the value when the R signal is processed or obtained. Are switched in synchronism with the switching between the two phase identification reference signals which are accordingly phased differently.

The cumulative result of all offsets can be understood with reference to FIG. FIG. 9 shows that X is much larger than the R component.
Based on a more realistic measure using the components, the impedance vectors 118 and 118 'for the same coin and without coin of FIG. 3 are shown. Coin presence and coin absent X values are X ₁ and X _2, respectively. The R values when the coin is present and when the coin is absent are R _{1 and} R ₂ , respectively, and the two difference values are shown as △ X and △ R at the top right of FIG.
These define the displacement line 120. Substantial first offset voltage applied through the capacitor Ci as described previously are represented as X _0, X ₁₀ to the X ₁ and X _2, the size of R ₁ and R ₂ are comparable And X _{20 so} that line 120 is 1
Shifted to 20 '. The second X offset voltage applied by the second offset circuit 134 is represented as X ₀ ′ and shifts the voltages X ₁₀ and X ₂₀ to X ₁₀ ′ and X ₂₀ ′, respectively, thus shifting line 120 ′ to 120 ′. Circuit 13
The R offset voltage from 4 is _denoted as R ₀ ′ and the voltage R
Shift ₁ and R ₂ to R ₁₀ 'and R _20' each, the result line 120 to "shift to 120. Quiescent i.e. X component value X ₁₀ during coin absent" is be close to zero from Figure 9 It can be seen that it can be converted to an analog to digital converter 136.
At the bottom of the dynamic range. The coin absent value R ₁₀ ′ of the R component signal is located near the top of the dynamic range of the analog-to-digital converter 136. Difference value △ X
And ΔR and the resulting θ remain unchanged with the application of the offset, as shown near the bottom left corner of FIG. The different but substantially overlapping portions of the dynamic range of the transducer are occupied, so that the dynamic range is fully utilized.

If the coin is large enough to affect the entire magnetic field of the coil 104 at the frequency most commonly used for coin inspection, the angle θ described above and shown in the figures,
The angle C shown in FIG. 4 is constant for a given coin material. However, as the frequency decreases below the most commonly used range, for example below 20 kHz, the angle θ begins to change, which change depends on the thickness of the coin. FIG. 10 shows a set of three curves representing the value of angle θ for three test disks of the same material but of different thicknesses. The values of θ are shown over the frequency range (on a logarithmic scale) in which the coil 104 can be driven. The thinner the disk, the higher the frequency at which the thickness begins to affect the angle θ, and vice versa.
In general, the thickness dependence of the angle θ becomes significant when the frequency is reduced to a point where the skin depth of the magnetic field in the material becomes about 1/3 of the thickness of the material. It can be seen from FIG. 10 that when the frequency is so high that the skin depth is almost smaller than the thickness of all the test disks, the angle θ does not depend on the thickness. The higher the conductivity of the material, the lower the skin depth at a given frequency. As a result, lower frequencies need to be advanced to obtain effective thickness dependence for higher conductivity coin materials. US coin sets mainly consist of relatively high conductivity, and it is preferable to use a frequency of 10 kHz or less, for example, 6 kHz or less, in order to obtain thickness sensitivity for the coin set and magnetic coins. Regarding cupronickel, which is common to UK coin sets,
Since the conductivity is lower and the skin depth is greater at a given frequency, a significant thickness dependence can be obtained at frequencies below 100 kHz, and more effective at frequencies below 50 kHz, more preferably at 35 kHz. can get. In these lower frequency ranges, the angle θ depends on the coin thickness, just like the material, but for the most part is independent of the distance of the coin from the coil, so that reliable thickness-dependent measurements Can be performed using one coil provided on one side of the coin passage.

The actual coin inspection device was configured to use two inspection inductors comparable to the inductor 104, using the techniques described individually with respect to FIG. Both inductors were provided on the same side of the coin path. The first inductor, consisting of an annular coil placed in a ferrite pot core, was 14 mm in diameter and was driven at 8 kHz. The second inductor, with respect to the coin travel direction, had the same configuration, but was driven at 37.5 mm in diameter and 115 kHz. The first inductor is smaller in diameter than the smallest coin to be accepted and is placed on the coin path so that the coin is always completely blocked by the coin when it is centered relative to the coil. Since this inductor is driven at a relatively low frequency of 8 kHz, the value of the angle θ drawn with this coil depends on both the material and the thickness of the coin. The second inductor had a diameter greater than the largest coin to be accepted and was placed in its lowest plane with respect to the coin path. The higher frequency of 115 kHz ensured that the angle θ drawn with this inductor was substantially independent of coin thickness, but the large diameter of the coil meant that the angle θ was not only Sensitivity was also provided for diameter or area. This inductor was positioned downstream of the coin path to stop the coin from moving. Otherwise, the momentum of this coin will affect the diameter sensitivity measurement of the coin. Such a bounce will have little effect on the output of smaller thickness sensitive inductors.

Both coils are driven by the same digital signal generator 100,
The outputs from both coils were processed by the same preamplifier 132 and other components to the right passing to the microprocessor 116, with reference to FIG. Each inductor has its own filter 102, drive transistor 103, and configuration and offset circuit 130, and these two components alternate under control of microprocessor 116 at point P in FIG. Was put into the network of FIG. 7 and cut off. Microprocessor 116 simultaneously switched generator 100 to a lower and higher frequency suitable for the two inductors.

As described, the measurand is generated when the displacement line direction and ΔX itself are at an extreme value, but as is known,
Measured quantities taken at other times while the coin is passing through the sensor can be used, and the techniques described can be used in the method as well.

In the embodiments described above, the phase identification method is used to derive X, R, △ X, and △ R, but various new and inventive aspects of those embodiments are useful even if they are derived. Other methods (such as will be described with respect to FIGS. 11 and 12), such as using ΔX as an acceptance criterion in addition to the displacement line direction, or using a displacement line at a lower frequency as a thickness response measurement Using directions is also available when used.

The described techniques for compensating for gain changes are available to coin devices regardless of the source or significance of the signal being processed.

Next, the apparatus of FIG. 11 will be described in detail. The pi-type tuning circuit 2 includes an inductor in the form of one coil 4, two capacitors 6 and 7, and a resistor 8. The resistor 8 should not be considered as a separate component, but rather represents the effective loss of the tuning circuit, which will mainly consist of the intrinsic loss of the coil 4.

Means are provided for positioning the coin, shown in dashed lines, at 10 near the coil 4, which means is schematically shown as a coin path 12 along which the coin passes the coil bordering the edge. As the coin 10 passes through the coil 4, the total effective loss of the tuning circuit increases, peaks when the coin is centered relative to the coil, and then decreases to the rest level. In this embodiment, the device responds to this peak value of the effective loss.

The tuning circuit 2 has a feedback path so as to form a free-running oscillator. The feedback path is indicated generally at 14 and includes a signal line 16 carrying the voltage developed at some point in the tuning circuit, and a switching circuit 18
And an inverting amplifier 20 for providing a feedback path gain. twenty four
The phase delay circuit schematically indicated by
Depending on the condition of 18, the vehicle is alternately put into the return path or bypassed. When the phase delay circuit 24 is not in the feedback path, the phase shift around the feedback path will be 180 ° and then the phase shift across the pi-tuned circuit will also be 180 °.
In this state, the oscillator is operating at its resonance frequency.

Reference is now made to FIG. Figure 12 shows the total effective loss of the tuning circuit from a relatively low value R1 to a relatively high value.
The relation between the oscillation frequency and the phase shift amount (φ) of the feedback path is shown for five different values up to R5. In general terms, for a pi-tuned circuit with a variable effective loss, the magnitude of the effective loss of the tuning circuit at any particular time will change the amount of phase shift in the feedback path from one known value to another (or It can be ascertained by changing (by some known amount) and measuring the resulting frequency change. The relationship between phase shift change and frequency change is illustrated in the figure.
Effectively representing the slope of one of the curves shown in FIG.
The result indicates on which curve the circuit is operating, and thus what the current effective loss of the circuit is. For example, as shown, the phase shift is from 180 ° to the amount φ ₁ (about 30
゜) and the frequency changes by △ fNC,
The effective loss amounts to a low value R1. However, if the frequency changes by a larger amount ΔfC, the effective loss will be at a higher value R4.

This is performed with the circuitry shown schematically in FIG. The description of FIG. 11 will end with the following.

The frequency of the oscillator is detected by the frequency detection circuit via the signal line 26.
Supplied to 28. The control circuit 30 operates the switching circuit 18 via the signal line 32 so as to move the phase delay circuit 24 into and out of the oscillator feedback path. Via the same signal line 32, the control circuit 30 also operates the switching 34 in synchronization with the switching circuit 18, so that the frequency value detected by the detection circuit 28 is stored in the storage 36 (this ,
This is the frequency value when there is no phase delay in the oscillation circuit) and is stored in the memory 38 (this is the frequency value when the phase delay is introduced into the oscillation circuit). FIG. 11 and the following discussion can be better understood by referring to the following table of notations used for various frequencies and frequency differences. That is, f0 = frequency without phase shift fφ = frequency with phase shift Δf = fφ−f0 ΔfNC = Δf without coins ΔfC = peak value of Δf with coins f0C = coins The peak value of f0 at the time f0NC = the value of f0 when there is no coin The subtractor 40 subtracts f0 from fφ to develop Δf, and in the normal state of the switch 42, the value of Δf is stored in the memory 44. Sent. This normal state prevails while the coin is not close to the coil 4, in which case the effective loss of the tuning circuit is low (ie the low value R1 in FIG. 12) and is stored in the memory 44. The frequency difference value becomes ΔfNC (shown in FIG. 12), which indicates the inherent effective loss of the tuning circuit itself when the measurement is being made.

As coin 10 starts to reach near coil 4,
The output f0 of the frequency detection circuit 28 starts to change. Control circuit
Part 46 of 30 detects the onset of this change from signal line 48,
In response, the state of the switch 42 is changed via the signal line 50, and the latest pause value of ΔfNC is stored in the memory 44.

When the coin 10 approaches and reaches the center position with respect to the coil 4, the frequency f0 drops until it reaches a low peak value. Circuit portion 46 is adapted to detect this peak occurrence, causing switch 42 to direct the value of Δf that occurred when the coin was centered to memory 52 in response. This becomes, for example, the value ΔfC as shown in FIG. 12, and becomes the maximum value of the frequency shift resulting from the additional phase change φ1 generated while the coin passes through the inductor. The frequency shift is such that the total effective loss of the tuning circuit is now a relatively high value R4, which is the effective loss inherent in the circuit plus the effective loss introduced into the circuit by the particular coin currently in the center of the coil 4. Show. The effective loss R of the coil is k ₁
Δf. Here, k ₁ is a constant. Value indicating the effective loss introduced only by the coin is derived from the circuit 54 for multiplying a constant k ₁ by subtracting △ FNC from △ fC. This is equal to ΔR as previously mentioned.

The circuit of FIG. 11 also measures ΔX, that is, the amount of reactance introduced into the tuning circuit 2 by a coin as follows. f _O
(Ie, the oscillation frequency without any additional phase shift) is applied to switch 62 via signal line 64. As a result, the absence of cost without phase shift, i.e., the idle frequency, is stored in the memory 66, and the peak low frequency of the coin, which has reached without a phase delay when the coin passes through the inductor 4, is stored in the memory 68. Become so. These frequencies represent the total reactance of the tuning circuit itself, each with the additional effect of a coin. The effective reactance X of the coil becomes k ₂ / f0. Here, k ₂ is a constant. ΔX is
Taking the inverse of the two frequencies is drawn by the circuit 70 for multiplying the constant k ₂ is subtracted them.

The outputs of circuits 54 and 70 are provided to divider 72, which
Is △ X / △ R (ie, the
n.theta.) and send it to a comparator 74, where it is compared with a reference value of tan .theta. from reference circuit 74. If they match, comparator 74 provides an output to AND gate 76.

In practice, one or more tests will be performed on the coin. Then, yet another input for each test value comparable to the reference value for coins of the same type is applied to the unhung gate 76. All inputs, each for each test, but when the tested coin yields a complete set of values comparable to each reference value for a given denomination of the coin, AND gate 76 accepts at its output A signal is generated to accept coins, for example, by operating an accept / reject gate in a known manner. Of course, additional tests may be used in conjunction with those previously described with respect to FIGS.

Although the embodiment of FIG. 11 has been described and illustrated above with switches and functional blocks, in practice, all components shown within the dashed box 80 are suitably provided by a suitably programmed microprocessor. . The programming is within the skill of a skilled programmer,
Given the functions to be achieved as explained above.

Although the inductor is shown as a single coil,
Other forms, such as opposing across the coin path and connected in a parallel or series forward or reverse connection 1
Sets of coils.

As described, the measurand is generated when the oscillation frequency is at the peak value, but it is also possible to take a measurable measurand at other times while the coin passes through the coil, as is known And the techniques of FIGS. 11 and 12 can be used in the method as well. In order to take into account the fact that identical acceptable coins of a given denomination vary in character to some extent, any comparison made to check acceptability in any embodiment is limited, for example, by upper and lower limits. It will be understood that this may be taken into account by providing a reference value in the form of a given range, or by applying a tolerance to the measurement before comparison with the exact reference. Would. All the reference values can be stored, for example, in the memory of a microprocessor or in a separate digital memory, or can be calculated from the stored coin-related information whenever necessary.

Continuation of front page (72) Inventor Bailey, John William UK. RG 62 Utility, Berkshire, Reading, Airlie, Adams Way 18 (72) Inventor Ralph, Allan United Kingdom. S.O. 427. Doubled, Hampshire, Near Brockenhurst, East Boulder Road (no address) (72) Inventor Chittleborough, Michael United Kingdom. H.P. 10 Zero N.P., Bucks, High Wiyum, Wood Burn Moor, Falcons Croft, Shelly Crows 3 (72) Inventor Sagadi, Curly United States. 19335 Pennsylvania, Downington, Conlays Way 115 (58) Field surveyed (Int. Cl. ⁶ , DB name) G07D 5/08

Claims (23)

(57) [Claims] 1. A method for inspecting a coin in a coin inspection device, comprising: exposing a coin to an oscillating magnetic field generated by an inductor; measuring reactance and loss of the inductor; Of the displacement line representing the displacement of the coin present point in the impedance plane representing the reactance and loss of the inductance obtained in the measurement when the coin is in the magnetic field with respect to the reactant tank and the coin absent point in the impedance plane representing the loss. Determining whether the direction in the impedance plane matches a predetermined reference direction for the test coin in the impedance plane. 2. A circuit including a coin path and an inductor, wherein the inductor generates an oscillating magnetic field in the coin path, means for measuring reactance and loss of the inductor, and when the coin is not in the magnetic field. FIG. 6 shows displacement of a coin present point in the impedance plane representing the reactance and loss of inductance obtained in the measurement when a coin is in the magnetic field with respect to a coin absent point in an impedance plane representing reactance and loss of inductance of FIG. Means for determining whether the direction of the displacement line in the impedance plane matches a predetermined reference direction for the test coin in the impedance plane. 3. Apparatus according to claim 2, wherein said means for measuring inductor reactance and loss includes phase discrimination circuitry and includes a signal source for driving the inductor. 4. The apparatus according to claim 3, wherein said signal source is a constant current source. 5. The apparatus of claim 3 wherein said means for measuring inductor reactance and loss comprises a phase discrimination network, and the angle in the impedance plane of the phase discrimination axis with respect to the true reactance and loss axis. An apparatus comprising means for measuring displacement. 6. Apparatus according to claim 5, including means for angularly shifting a phase identification axis on which said phase identification network operates to reduce said angular displacement. 7. The apparatus according to claim 5, wherein said checking means includes means for adding a compensation factor derived from said angular displacement measurement. 8. Apparatus according to claim 2, including means for setting said reference direction as an angle with respect to one of the reactance and loss axes. 9. Apparatus according to claim 8, including phase identification circuitry adapted to measure reactance and loss of the inductor, wherein said identification means is coupled to one of the phase identification axes. An apparatus adapted to evaluate the angle of the displacement line for the same. 10. The apparatus according to claim 9, wherein:
The checking means includes means for adding a compensation coefficient derived from the angular displacement measurement amount, and the compensation coefficient applying means combines the measured angular displacement of the phase identification axis and the angle of the evaluated displacement line. A device that is adapted to: 11. The apparatus according to claim 2, wherein:
The measuring means is further adapted to measure the reactance and loss of the inductor when the coin is absent, and to set a point representing the reactance and loss of the inductor when the coin is absent, from the coin presence measured amount and the coin absence measured amount. An apparatus comprising means for confirming the direction of the displacement line. 12. The apparatus according to claim 11, wherein:
An apparatus including means for causing the measuring means to take a measured amount when the coin is absent every time the coin is inspected. 13. An apparatus according to claim 2, wherein:
A means for providing an indication of a reference displacement line whose direction in the impedance plane is the reference direction and whose position in the impedance plane extends through the point at which the coin is absent; A device adapted to ascertain whether the reactance and loss measurements of the sub-range substantially limit a point lying on the reference displacement line. 14. An apparatus according to claim 2, wherein:
Apparatus adapted to evaluate the angle of the displacement line with respect to the total impedance vector of the inductor in the absence of a coin. 15. The device according to claim 14, wherein
Apparatus including means for setting said reference direction as the total impedance vector of the inductor in the absence of a coin in the impedance plane. 16. The apparatus according to claim 2, wherein:
A common path in which the reactance and loss dependent signals of the inductor are processed, and means for adding an offset to the reactance dependent signal to substantially reduce its value to match the value of the resistance dependent signal. An apparatus adapted to use the difference between the coin presence value and the coin absence value of the reactance dependent signal. 17. The apparatus according to claim 16, wherein:
From said common path, the signal goes to yet another common path,
The confirmation means is adapted to use the difference between the coin presence value and the coin absence value of both the reactance dependent signal and the loss dependent signal in the confirmation step, and before the further common path, the coin absence value of each signal. The value approaches the end of the dynamic range of yet another common path component, so that means for adding an offset to one or both of the signals to maximize the dynamic range of said component is provided. Equipment provided. 18. The apparatus according to claim 2, wherein:
The reference direction is suitable for a specific coin type, and the checking means further checks whether a difference between a coin absence value and a coin presence value of the inductor matches a reference value suitable for the same specific coin type. Equipment that is adapted to. 19. The apparatus according to claim 2, wherein:
A device wherein the frequency of the transmitted magnetic field generated by the inductor is low enough that the direction of the displacement line is affected by the thickness of the coin being inspected. 20. The apparatus according to claim 19, wherein
The device wherein the frequency is low enough such that its skin depth for the coin material is at least 1/3 of the coin thickness. 21. The apparatus according to claim 2, wherein
A device wherein the inductor is on only one side of the coin path. 22. The apparatus according to claim 2, wherein
Apparatus comprising means for detecting the direction of the displacement line that reaches an extremum while the coin passes through the inductor, wherein the confirmation means is adapted to use the direction of the extremum. 23. The apparatus according to claim 22, wherein
The checking means repeatedly evaluates the direction of the displacement line as the coin passes through the inductor along the edge,
A device operable to detect when the direction of the displacement line is at an extreme value from the results of the evaluation.