AU744618B2 - Coin acceptor - Google Patents

Description

WO 99/23616 lPrTI('2i~l O If¢1' A WO 99/23616 VirT/(dO_1QQ~ftVA'

J

Coin Acceptor Background This invention relates to a coin acceptor and has particular but not exclusive application to a multi-coin acceptor for validating coins of different denominations.

Field of the invention Coin acceptors which discriminate between coins of different denominations are well known and one example is described in our GB-A-2 169 429. The acceptor includes a coin rundown path along which coins pass through a sensing station at which coils perform a series of inductive tests on the coins in order to develop coin parameter signals which are indicative of the material and metallic content of the coin under test. The coin parameter signals are digitised so as to provide digital coin parameter data, which are then compared with stored coin data by means of a microcontroller to determine the acceptability or otherwise of the tested coin. If the coin is found to be acceptable, the microcontroller operates an accept gate so that the coin is directed to an accept path otherwise, the accept gate remains inoperative and the coin is directed to a reject path.

The coin sensing station includes a number of different coils which may be energised at different frequencies and may be of different sizes so as to form individual inductive couplings of with the coin under test as it passes through the coin sensing station, on one side only or on both sides of the coin. The coils used hitherto have had a cross-sectional area which is sufficiently large relative to the coin that the inductive coupling produces eddy currents over at least a major part of the surface of the coin, with the result that the sensed parameters constitute an average of a number of different parameters of the coin, for example its metallic content, its thickness and its surface pattern.

The averaging effect produced by the relatively large sensor coils gives rise to disadvantages in certain circumstances. For example, there is an increasing popularity for coins to be minted to include regions of different materials, for T, -7771 777 WO 99/23616 PCT/GB98/03242 -2example, from more than one metal or metal alloy, and certain denominations of coins are formed of a central region of a first alloy, which is surrounded by an annular region of a second different alloy (referred to hereinafter as a "bimet" coin).

The different regions present different inductive characteristics to the sensor coils of the acceptor but the relatively large area coils tend to average the effect of the two metallic regions, with the result that certain bimet coins cannot satisfactorily be distinguished from coins of other denominations and frauds, for example washers with a central hole.

Another example is described in US-A-4 995 497 (Tamura Electric Works Ltd) in which the coils are aligned along the path are energised to detect different characteristics of a coin under test. Two different characteristics of the material of the coin and the coin thickness are detected by detecting peaks in the values of signals induced in the coils, and the coin diameter is detected by detecting the crossover in the outputs of two of the coils spaced along the coin path. The peaks and the crossover are compared with data for true coins in a memory. The aligned nature of the coils needed for the diameter measurement restricts the sensing of the coin and gives rise to difficulties with bimetallic coins.

W0-A-93/22747 (Mars Inc) discloses a coin acceptor for use with multi-metallic coins that uses two magnetic sensors each substantially smaller in width than the diameter of the multi-metallic coin, arranged essentially parallel to the coin path.

The sensors are connected in an electrical bridge circuit and the difference between the outputs of the sensors is used to discriminate between multi-metallic coins and frauds.

Summary of the invention In accordance with the present invention, it has been appreciated that by using small coils arrayed transversely of the coin path, and monitoring their outputs, it is possible to extract much more fine detail about the characteristics of the coin by seeking at least one predetermined criterion in the characteristics of sample values of at least one of the sensor outputs, and then comparing data from the sensor WO 99/23616 PCT/GB98/03242 -3outputs with corresponding stored data in order to determine acceptability of the coin More particularly, the invention provides a coin acceptor comprising: a path for coins, a plurality of sensor coils to sense a coin as it moves along the path, the sensors being arranged in an array extending transversely of the coin path so as to sense respective different regions of the face of a coin as it moves past them, and produce time varying sensor outputs as a function of the different regions respectively, sampling means to sample repetitively values of the sensor outputs produced during the passage of the coin past the sensors, and control means to successively monitor the sample values to determine when the value of at least one of the sensor outputs accords with a predetermined criterion, and in response thereto, to compare data from the sampled sensor outputs with corresponding stored data to determine acceptability of the coin.

The array of sensor coils may include coil assemblies that are arranged in one or more lines extending transversely of the coin path. They may be on opposite sides or on one side only of the coin path.

The coil assemblies preferably have an area to face the coin of less than 72mm 2 The sensor coil units may each be coupled in an oscillator circuit and the sampling means may be operative to sample an oscillatory characteristic of the circuit as the coin passes the unit such as frequency or amplitude or both.

The control means may be configured to select an ensemble of sample values of the sensor outputs that occur when one of the outputs accords with the predetermined criterion, and to compare the selected sample values with the stored data.

Alternatively, the control means may be configured to determine when the sensor outputs individually accord with a respective predetermined criterion and to compare the values thereof with the stored data.

WO 99/23616 PCT/GB98/03242 -4- The predetermined criterion may comprise a discontinuity in the value of sensor output such as a main or a localised maximum or minimum in the value of the sensor output that occurs during the passage of the coin. The predetermined criterion may also occur when at least one sample value from one of the sensors forms a predetermined value relationship with at least one sample value from another of the sensors. The value relationship may comprise a crossover of the values of the successive samples from one of the sensors with corresponding sample values from another of the sensors, or may be a function of the relative rate of change of the sample values from the sensors.

The invention also includes a coin discrimination method comprising: sensing the passage of a coin along a path with a plurality of sensor coils that are spaced apart so as to sense respective different regions of the face of a coin as it moves past them, in an array extending transversely of the coin path, so as to produce time varying sensor outputs as a function of the different regions respectively, sampling values of the sensor outputs repetitively during the passage of the coin past the sensors, successively monitoring the sample values to determine when the value of at least one of the sensor outputs accords with a predetermined criterion, and in response thereto, comparing data from the sampled sensor outputs with corresponding stored data to determine acceptability of the coin.

Brief description of the drawings In order that the invention may be more fully understood, an embodiment thereof will now be described by way of example with reference to the accompanying drawings in which: Figure 1 is a schematic elevational view of a first embodiment of a coin acceptor in accordance with the invention; Figure 2 illustrates schematically the electrical circuits of the acceptor shown in Figure 1; Figure 3 is a schematic partial cross sectional view of the acceptor taken along the line A A' in Figure 1; Figure 4a is an enlarged cross sectional view of one of the coils shown in Figure 3; Figure 4b illustrates the front face of the coil shown in Figure 4a; WO 99/23616 PCT/GB98/03242 Figure 5 is a schematic illustration of the coils and the coil drive and interface circuits shown in Figure 1; Figure 6 is a graph illustrating how the coin parameter signals vary with time as a coin moves past the sensor coils; s Figure 7 is a schematic block diagram that illustrates processing steps carried out by the microcontroller on successive ensembles of samples of the coin parameter signals taken as the coin passes the coils; Figure 8 illustrates a routine carried out by the microcontroller for determining the occurrence of a peak in the coin parameter signal xl(a); Figure 9 illustrates a routine carried out by the microcontroller for determining the occurrence of a crossover in the values of the coin parameter signals xl(a) and x2(a); Figure 10 is a schematic elevational view of a second embodiment of coin acceptor in accordance with the invention; Figure 11 illustrates schematically the electrical circuits of the acceptor shown in Figure Figure 12 is a schematic illustration of the coils and the coil drive and interface circuits shown in Figure Figure 13 is a graph illustrating how the coin parameter signals vary with time as a coin moves past the sensor coils shown in Figure Figure 14 is an enlarged portion of the graphs shown in Figure 13, and Figure 15 is a schematic block diagram illustrating processing steps carried out by the microcontroller as the coin moves through the sensing station in Figure Detailed description First emliinient A first embodiment of coin acceptor according to the invention will be described which comprises a multi-coin acceptor capable of validating a number of coins of different denominations, including bimet coins, for example the new euro coin set and the new UK coin set including the new bimet £2.00 coin.

The physical layout of the coin acceptor is shown schematically in Figure 1. The acceptor includes a body 1 with a coin run-down path 2 along which coins under WO 99/23616 PCT/GB98/03242 -6test pass edgewise from an inlet 3 through a coin sensing station 4 and then fall towards a gate 5. A test is performed on each coin as it passes through the sensing station 4. If the outcome of the test indicates the presence of a true coin, the gate is opened so that the coin can pass to an accept path 6, but otherwise the gate remains closed and the coin is deflected to a reject path 7. The coin path through the acceptor for a coin 8 is shown schematically by dotted line 9.

The coin sensing station 4 includes four coin sensing coil units Cla,b, C2, C3a,b and C4 shown in dotted outline, which are energised in order to produce an inductive coupling with the coin. Also, a coil unit CC is provided in the accept path 6, downstream of the gate 5, to act as a credit sensor in order to detect whether a coin that was determined to be acceptable, has in fact passed into the accept path 6.

The coils are energised at different frequencies by a drive and interface circuit shown in Figure 2. Eddy currents are induced in the coin under test by the coil units. The different inductive couplings between the three coils and the coin characterise the coin substantially uniquely. The drive and interface circuit produces four corresponding coin parameter data signals x 1 x 2 x 3 x 4 as a function of the different inductive couplings between the coin and the coil units C1, C2, C3 and C4. A corresponding signal xc is produced for the coil unit CC.

In order to determine coin authenticity, the four parameter signals x 1 x 2 x 3 and x 4 produced by a coin under test are fed to a microcontroller 11 which is coupled to a memory in the form of an EEPROM 12. The microcontroller 11 processes the coin parameter signals derived from the coin under test in a manner that will be described in more detail hereinafter and compares the outcome with corresponding stored values held in the EEPROM 12. The stored values are held in terms of windows having upper and lower value limits. Thus, if the processed data falls within the corresponding windows associated with a true coin of a particular denomination, the coin is indicated to be acceptable, but otherwise is rejected. If acceptable, a signal is provided on line 13 to a drive circuit 14 which operates the gate 5 shown in Figure 1 so as to allow the coin to pass to the accept path 6.

Otherwise, the gate 5 is not opened and the coin passes to reject path 7.

~I i~r WO 99/23616 PCT/GB98/03242 -7- The microcontroller 11 compares the processed data with a number of different sets of operating window data appropriate for coins of different denominations so that the coin acceptor can accept or reject more than one coin of a particular currency set. If the coin is accepted, its passage along the accept path 6 is detected by a post acceptance credit sensor coil unit CC, and the unit 10 passes corresponding data x c to the microcontroller 11, which in turn provides an output on line 15 that indicates the amount of monetary credit attributed to the accepted coin.

The configuration of the sensor coils will now be described in more detail.

Referring again to Figure 1, the acceptor has a coin door 16 that is hinged on a shaft 17 on the acceptor body 1, in a conventional manner. The coin run-down path 2 is provided between an interior wall 18 of the door 16 and a wall 19 of the acceptor body 1, as shown in more detail in Figure 3. The run-down path 2 comprises an inclined lip 20 on the door 16, down which the coin runs edgewise past the sensor coil units C1, C2, C3 and C4. The coin 8 is shown on the lip 20 of the run-down path 2 in Figure 3 The coin is shown schematically in an upright position although in practice, it will lean against one of the walls 18, 19. As known in the art, the door 16 is spring biased to the closed position shown in Figures 1 and 3 but can be hinged outwardly from the body 1 in the event of a coin jam so as to release the jammed coin and allow it to fall to the reject path 7.

In order to describe the improved operating characteristics of the acceptor, the coin 8 shown in Figure 1, is illustrated as a bimet coin and in this example comprises the new £2.00 coin. This comprises a first, central cupro-nickel core region 21 surrounded by a second, circular region or ring 22 of an alloy referred to herein as bronze, comprising 76% Cu, 4% Ni and 20% Zn. However, the invention is not restricted to the detection of bimet coins, as will be evident hereinafter.

Referring to Figure 3, the coil unit Cla,b comprises a pair of coil assemblies Cla, Clb mounted on the inside of the wall 19 of the acceptor body 1 and on the wall 18 of the door 16. The coil assemblies Cia, Clb are configured to form an inductive ell WO 99/23616 PCT/GB98/03242 -8coupling selectively with the bronze ring 22 of the bimet coin 8 under test i.e. with no significant inductive coupling to the central cupro-nickel region 21 of the coin.

As shown in more detail in Figure 4, each of the coil assemblies Cia, Clb comprises a generally cylindrical bobbin 23 of plastics material, on which windings of a coil 24 are formed. The bobbin 23 is push-fitted into a so-called half pot core 25 made of sintered ferrite material. The core 25 includes a central, cylindrical yoke 26 formed with a through hole to reduce the amount of ferrite material used, and a surrounding, concentric, cylindrical support flange 27.

As an alternative to using a bobbin, the windings of the coil 24 may be wound around a former, not shown, and the windings heated to melt their insulation, so that on cooling, a self supporting coil is formed, which is then removed from the former and push-fitted into the half pot core The support flange 27 of the half pot core 25 is push-fitted in a corresponding recess in the wall; thus the flange 27 of assembly Cia is push fitted into a cylindrical recess 28 in wall 19 and the flange 27 of assembly Clb is push fitted into a corresponding recess 29 in the wall 18. In this example, the outer diameter d, of the windings of the coil 24 is 7.3 mm. The inner diameter d 2 of the coil 24 with its bobbin 23 is 2.78 mm and the diameter of the hole through the yoke 26 is 2mm.

The faces 30 of the coil assemblies Cla,b in this example, are spaced apart by 6.24 mm. The coils 24 have an axial length of 2.78 mm. The outer diameter d 3 of the half pot cores 25 is 9 mm and thus the area A of the end face 30 of each coil unit i.e. the end which faces the coin under test, is in this example 63.62 mm 2 The windings 24 of the assemblies Cla,b are electrically connected in series. As can be seen in Figure 3, the coil assemblies Cla,b are arranged with the coils 24 arranged on a common axis, on opposite sides of the coin 8 under test.

As well known in the art of solenoid coil design, the magnetic field of a generally cylindrical coil is concentrated along the coil axis; thus, for each of the coil assemblies Cla,b, the field is concentrated mainly in the ferrite yoke 26 of the half pot core 25 and the flux around the coil is mainly channelled in a loop around the WO 99/23616 PCT/GB98/03242 -9coil by the surrounding ferrite flange 27, except in the region of face 30 where the flux passes through the surrounding material back to the yoke 26. Accordingly, the sensitivity of the assemblies Cla,b to passing coins is for the most part, restricted to the region of the coin which passes between the yokes 26. The assemblies Cla,b are positioned closely adjacent the coin rundown path 2 and the dimension d of the coils is such that the inductive coupling between the coin and the coils is restricted substantially only to the second, outer region 22 of the coin 8, with no significant coupling occurring with the first inner region 21. As can be seen from Figure 3, the half pot cores 25 extend below the coin run-down path 20 in order that the cores 26 be configured in alignment with the outer ring 22 of the coin 8.

Considering the now the coil units C2, C3 and C4, these are made up of coil assemblies identical to the assembly Cia shown in Figure 4. The coil units C1-C4 are mounted in an array extending transversely to the coin path 2, along line 31 as shown in Figure 1. In this example, the line 31 extends orthogonally of the path 2, although other configurations transverse to the coin path can be used and more generally, it is not essential for the coil units to be arranged in a line. It has been found in accordance with the invention that improved discrimination can be achieved by making the area A of the coil assembly which faces the coin, such as the coil Cla,b, smaller than 72 mm 2 which permits coin regions with individual inductive characteristics to be sensed. The or each coil C need not be circular. In fact, advantages can be obtained from square or rectangular wound coils. The transverse array of sensors preferably includes at least three of the coil units.

Referring to Figure 3, the coil unit C2 is mounted above the unit Cia, b so as to intersect a chord of the coin as it passes through the coin sensing station 4. It will be understood that as the coin rolls past the coil unit C2, an inductive coupling will first be formed with the outer ring 22 of the coin, then the inner region 21 and then the outer region 22 again. The coil unit C2 comprises a single coil assembly and thus the coil unit 22 detects the characteristics of the coin from one side only.

The coil unit C3a, b comprises a pair of the coil assemblies C3a, C3b, mounted on opposite sides of the coin path in a similar way to the coil assemblies Cia, b. The I: WO 99/23616 PCT/GB98/03242 coil unit C3a, b is mounted above the unit C2 and thus is sensitive to the coin at a different chordal position thereof.

The coil unit C4 comprises a single one of the coil assemblies as shown in Figure 4, mounted at a location above the coil unit C3a, b.

It will thus be understood that the outputs from the coil units C1-C4 will depend upon the characteristics of the coin, including its diameter, its material characteristics, its thickness, whether it is a bimet coin and a number of other io factors, with the coil units being responsive to the characteristics of respective individual regions of the coin under test as it passes the coils.

Figure 5 illustrates how the coil units are connected to the coil drive and interface circuits 10 shown in Figure 1. Considering the coil unit C1, the coil assemblies Cia, Clib, are connected in series, in the feedback loop of an inverting amplifier Al, together with a capacitor C. The circuit thus acts as an oscillator, the output of which has an amplitude and frequency dependent on the inductance presented by the coil assembly Cia, Clb. As the coin passes between the coil assemblies Cia, Cib, an inductive coupling occurs between the coin and the coil assemblies with the result that the inductance in the feedback path of the amplifier changes, which in turn changes both the amplitude and frequency of the oscillator, on a transitory basis.

The amplitude is detected by an envelope detector El for the coil unit C1, and the amplitude of the envelope is successively sampled as the coin passes between the coil assembly Cia, Clb by means of an analog to digital converter D1, so as to provide a series of successive digital sample values xl(a) as the coin passes the coil unit C1.

The coil unit C2 is connected in the feedback loop of an inverting amplifier A2 and a corresponding digital coin parameter signal x2(a) is produced by means of envelope detector E2 and analog to digital converter D2.

i WO 99/23616 PCT/GB98/03242 -11- Additionally, the frequency deviation of the oscillator A2 is detected as the coin passes the coil unit C2. A frequency detector F detects the instantaneous frequency of the oscillator A2 and the output is successively sampled and digitised by analog to digital converter D2', so as to provide coin parameter output signal x2(f).

The coil unit C3a, b is connected in the feedback loop of inverting amplifier A3, with the coil assemblies C3a, C3b being connected in series. An envelope detector E3 and an analog to digital converter D3 produce an output digital parameter signal x3(a) which comprises a series of digital samples of the amplitude deviation produced as the coin passes between the coil assemblies C3a, C3b.

The coil unit C4 is connected in the feedback of amplifier A4 and envelope detector E4 together with analog to digital converter D4 produces an amplitude deviation signal x4(a).

Whilst for the purposes of illustration, individual A/D converters D1-D4 are shown, it will be appreciated that the outputs of the oscillator circuits could be multiplexed into a single A/D converter in order to reduce the expense of the circuitry.

Figure 6 illustrates the way in which the coin parameter signals x vary with time as the coin passes through the coin sensing station 4. It will be appreciated that the shape of the individual curves shown in Figure 6 are dependent upon the characteristics of the coin under test and the curves represent individual "signatures" for the coin denomination. The passage of the coin past the sensor coils C1-4 causes a general reduction in amplitude, but the A/D converters D1-D4 produce a signal inversion so as to product the graphs shown in Figure 6.

The analog to digital converters D1-D4 shown in Figure 5 produce an ensemble A, of sample values with successive ensembles being produced for successive sample periods, spaced apart by a time Dt. Thus, an ensemble At is taken at time t where WO 99/23616 PCT/GB98/03242 -12- At xtl(a) xt2(a) xt2(f) xt3(a) xt4(a) (1) Although Figure 6 illustrates the individual ensemble values all occurring simultaneously at time t, in practice, the individual sample values x which make up the ensemble may be taken over a finite period, which is significantly shorter than Dt.

As shown in Figure 2, the microcontroller 11 receives the successive values of the coin parameter signals x that make up the successive ensembles At. As shown in Figure 7 the microcontroller 11 assembles the successive ensembles Ak ->Ak+n iro a running stack of n successive ensemble values 32.

The resulting ensemble data in the stack is processed in a step S1 in order to determine when the value of at least one of the coin parameter signals x adopts a predetermined criterion, such as the occurrence of a peak value in the sample values or when the sample value from one of the sensors forms a predetermined value relationship with the corresponding sample value from another of the sensors. This value relationship may be constituted by a crossover in the graphs, or the reaching or overpassing of a threshold, as will be explained in more detail hereinafter. The ensemble A of data which includes the predetermined criterion is stored at step S2.

Then at step S3, the individual coin parameter data values xl(a), x2(a) x2(f), x3(a) and x4(a) are compared with corresponding stored values held in the EEPROM 12 (Figure The stored values are held in terms of windows W wla, w2a, w2f, w3a, w4a each with upper and lower limits, to accommodate minor variations from coin to coin. In fact, a series of different sets of coin windows W are stored in the EEPROM 12, corresponding to different coin denominations, and the results from step S2 are compared in step S3 with all of the stored sets in order to determine whether the coin is of an acceptable denomination.

m WO 99/23616 PCT/GB98/03242 -13- If the coin is determined to be acceptable, an output is provided at step 54 indicating the denomination of the coin and its acceptability, so as to provide the outputs 13, 15 shown in Figure 2 or, indicating that the coin should be rejected.

s The manner in which a peak in the coin parameter signal xl(a) can be detected during step S1, will be described in more detail with reference to Figure 8. The process starts at step S1.10. At step S1.11, a parameter p is set to equal zero. At step S1.12, successive coin parameter samples xl(a) are selected from three successive data ensembles Ak, for k p 1, p and p 1.

At step S1.13, the three successive values of xl(a) are compared with one another.

If the intermediate value is greater than the preceding and exceeding values, this indicates that a peak has occurred. Thus, the following inequality is checked xl(a)p.i xl(a)p. xl(a)p+i (2) If the test is true, the particular value of xl(a)p indicates a peak value. If the inequalities of test are not true, the parameter p is incremented at step S1.14 and the process is repeated so as to sweep through successive ones of the data ensembles in an attempt to find a peak in xl(a).

When a peak is found, the entire data ensemble Ak is fetched from the stack, for the particular value of k p. This is the data ensemble of coin parameter signals, occurring at the peak in xl(a).

At step S2.10, the fetched data ensemble is temporarily stored and, at step S3.10, the individual coin parameter signals from the stored data ensemble, namely xl(a), x2(a), x2(f), x3(a) and x4(a) are individually compared with corresponding windows stored in the EEPROM 12 to determine if the coin is of a particular denomination to be accepted by the acceptor. As previously explained, the process may be repeated for a number of different coin denominations which have associated stored windows in the EEPROM.

I- ii~i~~i~ WO 99/23616 PCT/GB98/03242 -14- Figure 9 illustrates the routine for determining when a crossover occurs in the graphs of xl(a) and x2(a) shown in Figure 6. When the crossover occurs, the data ensemble associated with the occurrence of the crossover is used for comparison purposes with stored window data in the EEPROM 12.

At step S1.20, the routine starts and a parameter p is set to zero at step S1.21. Then the values of xl(a) and x2(a) are fetched from the stack 31 for two successive data ensembles Ak, the ensembles being selected by means of the parameter p, namely fork=p&p 1.

Then, at step S1.23, the fetched data values are compared according to the following inequality in order to determine whether a crossover has occurred for the fetched data.

xl(a)p x2(a)p AND xl(a)p+1 x2(a)p+l?

OR

xl(a)p x2(a)p AND xl(a)p+l x2(a)p+l? (3) It will be seen that these tests determine if the graphs for xl(a) and x2(a) shown in Figure 6 crossover one another.

In the event that no crossover is found, the parameter p is incremented at step S1.24 and the process is repeated for the next successive set of values in the stack 31 of data ensembles Ak shown in Figure 7.

If, however, a crossover is detected, the ensemble Ak with a particular value of k p at which the crossover occurred, is fetched from the stack 31 and is stored at step S2.20.

WO 99/23616 PCT/GB98/03242 Then, at step S3.20, the individual values stored at step S2.20 are compared with corresponding windows held in the EEPROM 12, as previously described, in order to determine coin authenticity and denomination.

The described example of the invention has the advantage that much more fine detail about the characteristics of the coin can be determined using the transverse array of coil units C1-C4. The small size of the coil assemblies relative to the coin allow the characteristics of individual chordal regions of the coin to be determined individually as shown by the individual graphs in Figure 6. As previously mentioned, the graphs of Figure 6 show the outputs derived from a bimet coin.

Hitherto, when larger diameter sensing coils were used, which produced an averaging effect over at least a major portion of the surface area of the coin, it was difficult to distinguish between a bimet coin and a corresponding washer with a central hole. In contrast, the described embodiment of acceptor according to the invention, can readily distinguish between such a bimet coin and a corresponding washer. Referring to Figure 6, the output for x2(a) adopts the generally dome shaped configuration in response to a genuine bimet coin of a particular denomination. However, if a fraudulent washer with a central hole is passed through the acceptor, a trace 33 with a central "dip" is produced. With a prior art acceptor, which averaged the effect of the entire face of the coin, it was difficult to distinguish between the genuine coin and the washer, due to the averaging effect.

However, in accordance with the invention, if the peak amplitude xl(a) is determined, so as to determine the ensemble Ax the corresponding value of the parameter x2(a) in the ensemble Ax adopts substantially different values for the true bimet coin and a corresponding washer with a central hole, namely values 34 and respectively. Thus, the stored window data in the EEPROM 12 for the true bimet coil differs substantially from the data produced for a fraudulent washer, permitting such frauds readily to be detected. Thus, in accordance with the invention, by selecting the ensemble Ax, much more fine detail can be resolved than hitherto.

Different criteria in the data can be determined on initial testing to find ensembles which uniquely characterise a particular coin denomination. For certain coins, the WO 99/23616 PCT/GB98/03242 -16criterion used to select the ensemble Ax may be when one of the coin parameter signals x equals or crosses a predetermined threshold value stored in the EEPROM 12.

For some denominations, the crossover between certain graphs in the output shown in Figure 6 is a suitable criterion. For other coin denominations, there may be local minima in the graphs of Figure 6 which may be utilised. Certain bimet coins may produce a trough in one of the graphs, which can be used as the criterion.

Also, a more complicated relationship between points on the graphs of Figure 6 may be used and the relative shapes of the curves may be considered. For example, values of two of the coin parameter signals x may be taken from an ensemble A 1 produced at a first time and then again from an ensemble A 2 produced at a later time, and the values may be processed to obtain an indication of the gradient of is each of the graphs. When the gradients adopt a predetermined relationship, the corresponding data ensemble A is selected and compared with stored data in the EEPROM 12.

Second anbodiment A second embodiment of the invention will now be described with reference to Figures 10 to 15. The second embodiment is similar to the first embodiment and corresponding parts are marked with the same reference numbers. The second embodiment differs in the manner in which the coil units are disposed at the sensing station 4, and also the way in which the coin data are processed.

Referring to Figure 10, five of the previously described coil units C1-C5 are disposed in an array extending transversely of the coin path 9. Instead of being arranged in a straight line 31 as shown in Figure 1, the coil units are arrayed in a staggered arrangement, with the axes of the coil units being disposed to be generally orthogonal of the major faces of the coin as its passes through the axes, along the coin rundown path. As previously explained, the interaction between the coil units and the coin occurs primarily in the region of the core of each coil unit and the l T i:x jn WO 99/23616 PCTIGB98/03242 -17staggered coil arrangement shown in Figure 10 permits an additional coil unit C5 to be included within the circumference of the coin 8 as it passes along the path 9.

The configuration of coil units can be used to sense coins with a diameter range of 15-33 mm. Thus, the arrangement of Figure 10 allows an additional region of the Scoin face to be analysed by the fifth coil unit. The connection of the coil units to the microcontroller 11 is shown schematically in Figure 11 and generally corresponds to the arrangement previously described with reference to Figures 1 and 5, with the provision of additional circuits for the coil unit C5, which produces a coin parameter signal xs(a).

In this example, each of the coil units C1-C5 includes a pair of coil assemblies mounted on opposite sides of the coin path, on the coin door 16 and the wall 19 of the validator, in the same way as coil assemblies Cia, b shown in Figure 1.

Referring to Figure 11, the coin units C1-C5 together with the post acceptance coil unit CC are connected through coil drive and interface circuits 10 to microcontroller 11. The operation of the circuit is similar to that described with reference to Figure 2. A random access memory RAM 31 is shown connected to the microcontroller 11.

The connection of the coil units C1-C5 is shown in more detail in Figure 12. The coil assemblies of each coil unit are connected in series. The coil assemblies of coil unit C2 are connected in an anti-phase such that the polarities of the assemblies mutually repel. The other coin units have their assemblies connected in phase such that the coil polarities attract. Each coil pair is connected in an oscillator circuit in the manner previously described with reference to Figure 5. The natural resonant frequency of each oscillator circuit is different in order to reduce crosstalk. The frequencies are, but not restricted to, between 60 and 100 KHz, and are high enough so as not to completely penetrate coins under test. Instead of using individual A/D converters for each coil unit, a multiplexer 32 is provided, which successively scans the outputs of the coil units and feeds them to a common envelope detector El and to A/D converter D1 such that samples of xl(a), x2(a), x3(a), x4(a) and x5(a) are sequentially produced on output line 33.

1- i- WO 99/23616 PCT/GB98/03242 -18- Also, frequency detector F is provided to detect frequency changes in the output of coil units C2. The output of detector F is fed to an A/D converter D2' in order to provide successive samples of coin signal x2(f). It will be understood that the multiplexer 32 can also provide samples for the post acceptance sensor CC, although this is not shown in Figure 12 in order to simplify the description.

Figure 13 illustrates the outputs from the various coil units C1-C5 during passage of the coin through the coin sensing station 4 of Figure Figure 14 illustrates an enlarged view of three successive groups of samples taken during passage of the coin through the coin sensing station. It will be seen that the multiplexer 32 shown in Figure 12 strobes between successive ones of the outputs of the coil units C1-C5 to produce successive groups of samples xl(a) x5(a), that are fed to the microcontroller 11 shown in Figure 11. Also, the frequency modulation detected by frequency detector F of Figure 12 gives rise to corresponding sample values of x2(f).

The microcontroller 11 is configured to operate as shown in Figure 15. The successive groups of coin data samples are fed at step S5 to the microcontroller 11 where a number of different criteria in the sensor outputs are detected. In this example, twelve different criteria are monitored as follows: WO 99/23616 PCT/GB98/03242 -19r e r t a k-riterio Details 1 Maximum increase in xl(a) as coin passes 2 Maximum increase in x2(a) as coin passes 3 Maximum increase in x3(a) as coin passes 4 Maximum increase in x4(a) as coin passes Maximum increase in x5(a) as coin passes 6 Maximum increase in frequency x2(f) as coin passes 7 Maximum reduction in frequency of x2(f) as coin passes 8 Maximum decrease in amplitude between two local maxima of amplitude x2(a) as coin passes 9 Maximum decrease in amplitude between two local maxima of amplitude x3(a) as coin passes The value of amplitude x5(a) when the amplitude of xl(a) reaches a preset offset from its normal no-coin amplitude 11 The value of amplitude x4(a) when the amplitude of x2(a) reaches a preset offset from its normal no-coin amplitude 12 The value derived from one or more of the results of criteria 1 to 7 listed above. The derivation method may be preselected dependent upon the particular application or currency to be detected.

As used herein, "increase" and "decrease" relate to the inverted graphs shown in Figure 13. For criteria 1-6, the maximum increase in x(a) is relative to the value which occurs when no coin is present.

Referring again to Figure 15, these 12 criteria are detected at step S6. It will be understood that as the coin passes through the coin sensing station, successive groups of the coin data samples produced at step S5 are be fed to the o1 microcontroller 11. Individual maxima and minima according to the criteria listed above, will be detected and temporarily stored in the RAM 31 connected to the microcontroller 11 shown in Figure 11. This temporary storage is shown at step S7 in Figure 15. As the coin passes, the individual maxima and minima will be updated 1 WO 99/23616 PCT/GB98/03242 as successive local maxima and minima are detected. As this occurs, the previously stored values thereof are compared with the newly produced values and an appropriate one of them is stored, depending on whether it should be the maximum or minimum. When the coin has passed, the resulting stored values of the 12 criteria are then compared with window data held in EEPROM 12 shown in Figure 11, in the manner previously described with reference to step S3 for the first embodiment. The window data stored in the EEPROM thus corresponds to values of the twelve criteria for true coins of different denominations, against which the criteria data from the coin under test can be checked for authenticity. This is carried out at step S8 in Figure 15. The coin is then accepted or rejected according to the outcome of the comparison, at step S9, in the manner previously described.

It will be appreciated that a feature of this algorithm is that local peaks and troughs in the various graphs of Figure 12 may be located and then rejected if a larger peak or trough occurs subsequently during the detection process. It will also be understood that the invention is not restricted to the particular criteria sought by the algorithm. Instead other criteria may be detected such as cross-overs, gradients and the like, as previously described.

As used herein, the term "coin" includes tokens and other coin-like items of value.

i: ~r EDITORIAL NOTE FOR 96378/98 THE CLAIMS FOLLOW ON PAGE NUMBER

Claims (21)

1. A coin acceptor comprising: a path for coins, a plurality of sensor coils to sense a coin as it moves along the path, drive circuitry coupled to the sensor coils, the sensor coils being arranged in an array extending transversely of the coin path so as to be energised by the drive circuitry to inductively couple with respective different regions of the face of a coin as it moves past them, and to sense the regions individually to produce time varying sensor outputs as a function of the different regions respectively, sampling means to sample repetitively individual values of the sensor outputs produced during the passage of the coin past the sensors to provide corresponding sample values, and control means to monitor the sample values and determine when the value of at least one of the sensor outputs accords with a predetermined criterion, and in response thereto, to compare data from the sampled sensor outputs with corresponding stored data to determine acceptability of the coin.

2. A coin acceptor according to claim 1 wherein the sensor coils include an array of sensor coil units configured to form an inductive coupling selectively with respective regions of the coin.

3. A coin acceptor according to claim 2 wherein the array of sensor coil units includes coil assemblies that are arranged in a line extending transversely of the coin path.

4. A coin acceptor according to claim 2 or 3 wherein one or more of the sensor coil units include coil assemblies on opposite sides of the coin path.

A coin acceptor according to claim 4 wherein one of the coil assemblies is coupled to the drive circuitry and the other is coupled to the sampling means. :AMENDED SHEET (ARTICLE 19) -r i~yl~ WO 99/23616 PCT/GB98/03242 26

6. A coin acceptor according to claim 2 or 3 wherein one or more of the sensor coil units include coil assemblies on one side only of the coin path.

7. A coin acceptor according to claim 6 wherein each of the coil assemblies of said sensor coil units is coupled to both the drive circuitry and the sampling means.

8. A coin acceptor according to any one of claims 2 to 7 wherein the coil units include coil assemblies with an area to face the coin of less than 72mm 2

9. A coin acceptor according to any one of claims 2 to 8 wherein the coil units include coils configured around a magnetically permeable core facing the coin path.

A coin acceptor according to any one of claims 2 to 9 wherein the sensor coil units are each coupled in an oscillator circuit and the sampling means is operative to sample a parameter of the oscillatory characteristic of the circuit as the coin passes the unit.

11. A coin acceptor according to claim 10 wherein said characteristic comprises frequency or amplitude or both.

12. A coin acceptor according to any preceding claim wherein the control means is configured to select an ensemble of sample values of the sensor outputs that occur when one of the outputs accords with said predetermined criterion, and to compare the selected sample values with said stored data.

13. A coin acceptor according to any one of claims 1 to 11 wherein the control means is configured to determine when the sensor outputs individually accord with a respective predetermined criterion and to compare the values thereof with the stored data.

14. A coin acceptor according to any preceding claim wherein the predetermined criterion comprises a discontinuity in the value of sensor output. A coin acceptor according to claim 14 wherein the discontinuity comprises a main or a localised maximum or minimum in the value of the sensor output that occurs during the passage of the coin.

AMENDED SHEET (ARTICLE 19) WO 99/23616 PCT/GB98/03242 27

16. A coin acceptor according to any preceding claim wherein the predetermined criterion comprises the occurrence of at least one sample value from one of the sensors forming a predetermined value relationship with at least one sample value from another of the sensors.

17. A coin acceptor according to claim 16 wherein the value relationship comprises a crossover of the values of the successive samples from one of the sensors with corresponding sample values from another of the sensors, or may be a function of the relative rate of change of the sample values from the sensors.

18. A coin acceptor according to claim 16 wherein the value relationship comprises a 1o predetermined function of the relative rate of change of the sample values from the sensors.

19. A coin discrimination method comprising: sensing the passage of a coin along a path with a plurality of sensor coils that are spaced apart in an array extending transversely of the coin path and energised to inductively couple with respective individual regions of the face of a coin as it moves past them, so as to the regions individually and produce time varying sensor outputs as a function of the different regions respectively, sampling values of the sensor outputs repetitively during the passage of the coin past the sensors, successively monitoring the sample values to determine when the value of at least one of the sensor outputs accords with a predetermined criterion, and in response thereto, comparing data from the sampled sensor outputs with corresponding stored data to determine acceptability of the coin. A method according to claim 19 including continually seeking to determine when the value of at least one of the sensor outputs accords with the criterion whilst the coin passes the sensor coils, and selecting the sample value which best meets the criterion if the criterion is not met more than once.

AMENDED SHEET (ARTICLE 19) WO 99/23616 PCT/GB98/03242 28

21. A coin acceptor comprising: a path for coins, a plurality of sensor coils to sense a coin as it moves along the path, drive circuitry coupled to the sensor coils, s the sensor coils being arranged in an array extending transversely of the coin path so as to be energised by the drive circuitry to inductively couple with respective different regions of the face of a coin as it moves past them, and to sense the regions individually to produce time varying sensor outputs as a function of the different regions respectively, sampling circuitry to sample repetitively individual values of the sensor outputs produced during the passage of the coin past the sensors to provide corresponding sample values, and a processor to monitor the sample values and determine when the value of at least one of the sensor outputs accords with a predetermined criterion, and in response thereto, to compare data from the sampled sensor outputs with corresponding stored data to determine acceptability of the coin. AMENDED SHEET (ARTICLE 19)