This invention relates to a sensor apparatus for a coin validator.

Coin validators are used in a large variety of devices for detecting and identifying coins and similar items. The term "coin" is used herein as an umbrella term encompassing all items that may pass through a coin validator type device, and hence should be taken to include not only monetary coins but also any type of similar items such as tokens, counterfeit coins, components of composite coins, or washers. Coin validators are found in vending machines, change machines, slot machines and any other machines that rely on coin based payments.

Early coin validators relied on physical mechanical characteristics of the coin to identify acceptable and unacceptable coins. Mechanical sorting devices would identify coins based on thickness, diameter, weight and so on. At present, known coin validators can also include various electrical sensor devices for detecting different parameters related to the coin in order to identify the coin or to reject the coin as unacceptable. Light sensors detecting a shadow thrown by the coin are used to measure the size of the coin. These optical sensors can be analogue sensors, in which the output indicates a measure of the amount of light on the sensor, or they can be digital, wherein a linear array of pixels is provided and the sensor output indicates which pixels are illuminated. Such pixels can be in the form of photodiodes.

Electromagnetic systems can provide a measurement relating to the electromagnetic properties of the coin, which will be affected by the coin material as well as the size of the coin. The technology for the electromagnetic system is similar to the oscillating fields used in metal detectors. The coin passes between a pair of coils; one coil that transmits a magnetic field and the other that receives a magnetic field. As the coin rolls by, it alters the magnetic field and the variation is interpreted by the validator's control system. Differences in size, shape, or metal content produce different, very precise electronic signatures that determine type and

denomination. US 7520374 discloses an example of an electromagnetic sensor system.

The measured data is compared to stored data to determine if the coin should be accepted or rejected. A motor or solenoid can be used to open or close a rejection passage for a counterfeit or unrecognised coin, or to accept validated coins into the correct denomination tube or slot.

Figures 1 a through 1d show various prior art arrangements for an optical coin sensing device. The arrows show movement of the coin along a coin path. Sensor strips 2a, 2b, 2c, 2d take the form of linear arrays of photodiodes. Light is projected toward the sensor strips and as the coin 3 passes along the coin path the coin generates a shadow over the sensor strips. The size of the shadow is used to measure the diameter of the coin.

In Figure 1 a, the length of the optical sensor array 2a and the position of the sensor array relative to the coin path are chosen such that for at least the smallest diameter coins of a predetermined range of coins the coins will obscure the entire sensor array. This provides a simple screening process to reject coins that are too small and when a digital optical sensor is used the size of the coin can be derived from measurements of lengths of chords of the coin as disclosed in WO 2009/063197, or alternatively by measurements as described in WO

92/18952. This technique will enable a measurement of coin diameter even when the length of the sensor is shorter than the diameter.

In Figure 1 b, the length of the optical sensor array 2b and the position of the sensor array relative to the coin path are chosen such that for at least the largest diameter coins of said predetermined range the coins will not obscure the entire sensor array. The diameter of the coin can be assessed based on the portion of the array that is not obscured.

In Figure 1 c, a shorter sensor array is used, which is positioned so that the smallest coins will obscure a portion of the array. If it is assumed that the coin will always sit on the coin rail then the diameter of the coin can be assessed even though the sensor array does not span the entire width of the coin as in Figure 1 b.

The device of Figure 1 d is for identifying coins having a coin path in free fall, or along a sloping slide surface. Two optical sensor arrays are used to detect opposite edges of a coin falling along the coin path.

In these prior art systems the linear array would typically be used in combination with an electromagnetic inductor system. The combination of diameter and magnetic characteristics can be matched against stored data to identify the coins. Thus, various systems exist that can validate coins. However, whilst these systems can accurately identify and validate coins when the coins are passed past the sensors in single file, difficulties arise when coins overlap and/or are closely spaced so that they bounce off each other. Overlapping coins produce different readings from the inductor and the overlap can make it harder to detect the diameter. When coins bounce off one another this can change the speed and the direction of travel along the coin path, leading to spurious or inaccurate readings. Hence, genuine coins might be rejected or wrongly identified, and counterfeit or unacceptable coins might be incorrectly accepted.

Viewed from a first aspect, the present invention provides a sensor apparatus for a coin validator, the apparatus comprising: first and second sensor arrays placed on a coin path, wherein the sensor arrays are linear optical arrays extending transverse to the direction of movement of coins along the coin path and wherein the arrays are offset from one another in the direction of movement of the coins.

The sensor arrays are placed on the coin path in order that, in use, coins passing along the coin path will pass between the arrays and a light source and cast a shadow onto at least some of one or both the sensor arrays. Since the arrays are offset from one another along the direction of coin movement then they form a sensor capable of obtaining a greater amount of information about the coin than conventional sensors. An ordinary linear array cannot distinguish between overlap and bounce back, whereas the offset array arrangement can. The offset in the longitudinal direction along the coin path allows the sensor arrays to detect a coin in different positions along the coin path, which will correspond to the position of the coin at different times during its movement along the coin path. This enables the sensor array to provide data relating not only to the dimensions of the coin but also to the velocity of the coin.

When overlapping and bouncing coins can be dealt with more effectively the sensor apparatus can process coins at a greater speed. This means that with the use of this sensor apparatus the number of coins per minute can be increased compared to prior art systems.

Preferably, the arrays are offset from one another in the transverse direction, i.e.

perpendicular to the direction of coin movement. The offset in the transverse direction means that the sensor arrays extend over a greater length across the coin path than if they were aligned.

The degree of longitudinal and transverse offset can vary whilst still enabling

advantages to be obtained. Preferably the first array and the second array are located relatively close together in the longitudinal direction, perhaps with an offset of one to three array widths between the centres of the arrays. This means that the coin measurements at different positions along the coin path will show the coin at two points relatively close together in time, compared to a greater degree of offset. The effects of any bouncing or variation in coin speed will hence be minimised. Ideally the first and second arrays are preferably placed so that the extent of the arrays overlaps in the transverse direction. This means that the arrays can provide a reading that extends continuously. Preferably, the overlap is small compared to the length of the array, for example the overlap may be less than one tenth of the length, preferably less than one twentieth. The overlap may be limited based on the resolution of the array, for example the degree of overlap may be equivalent to a certain number of pixels. In a preferred embodiment the overlap is a single pixel. By reducing the degree of overlap the combined length of the two arrays will extend over a greater length, hence covering a greater potential range of coin diameters.

The use of offset arrays enables the sensor array to detect and discriminate between overlapping coins and coins that are bouncing off one another. In a preferred embodiment the sensor apparatus has a control system that receives data from the sensor arrays.

Since the offset in the direction of travel allows the coin velocity to be assessed, including the direction of movement, then a bouncing coin can be identified since it will travel in the opposite direction to the usual direction of travel, or it will have a reduced velocity compared to the expected velocity. A lower than expected coin speed can be caused by retardation due to contact with a preceding coin. A more significant contact with a preceding coin can cause the coin to bounce back.

The control system may be arranged to determine coin speed and direction of travel based on data from the two sensor arrays and on the time of the data. For example, the control system may be arranged to determine the direction of travel based on which array is first affected by the coin. The control system may be arranged to determine coin speed based on the difference between the times at which the coin is detected by the first and second arrays, and on the distance between the arrays. The determination of coin speed may also take into account the degree of obscuration of the sensor array(s) by the coin, and the control system may be arranged to calculate the position of the coin relative to the sensor. This can enable accurate velocity calculations even when different parts of the coin are detected by the two offset arrays.

The prior art does not provide a control system operating in this manner.

In a preferred embodiment the control system is arranged to determine that data from the sensor arrays relates to a bouncing coin when the coin is travelling in the opposite direction to the usual direction of coins, and/or when the coin speed is lower than an expected speed. The data relating to a bouncing coin could be ignored if it is determined that the coin has already been accurately measure and can be identified. Alternatively a bouncing coin may be rejected.

Preferably, the control system is arranged to determine when coins overlap. More preferably, the control system is arranged to determine the degree of overlap.

The degree of overlap may be determined based on a measurement or calculation of the diameter of the coins and a measurement or calculation of the combined length of the overlapped coins in the direction of movement along the coin path. The sensor array may be used to measure or calculate diameter in any conventional fashion, for example based on a measurement of the maximum degree of obscuration of the linear array or arrays. The combined length of the overlapped coins may be determined based on the coin speed and the time for which the coins obscure an array or both arrays. If the coin diameters and the combined length are known then the degree of overlap can be determined.

Alternatively or in addition the control system may be arranged to use geometrical considerations to determine the degree of overlap.

The degree of overlap can be utilised to determine if other measurements in a broader coin validation system are accurate. For example, an inductor measurement on the leading edge of the first coin of two overlapping coins may still provide accurate information about the first coin despite the overlap, if the second overlapping coin is sufficiently far from the point of measurement to have no effect or a negligible effect. Similarly, an inductor measurement on the trailing edge of the second coin of two overlapping coins may be accurate, depending on the degree of overlap. Thus, the control system may be arranged to determine when inductor measurements, or other coin measurements, will give accurate results. The control system may be arranged to provide data on the required timing for inductor readings, for example the data may be provided to a broader control system that controls a coin validation system, including other sensor devices.

The degree of overlap may also be utilised by the control system to determine if coins can be physically separated by mechanical systems. Since the sensor apparatus will typically be used as a part of a coin validating and sorting device it is beneficial to know if overlapping coins can be sorted by other parts of the larger device, as well as being able to identify the coins.

In a particularly preferred embodiment the control system is arranged to both determine if a coin is bouncing and also to determine the degree of overlapping for overlapped coins. This can enable the sensor apparatus to be used in a coin validator to more accurately accept or reject coins.

By distinguishing between bouncing coins and overlapping coins, and processing the measurements of the coins accordingly, the sensor apparatus can be used to more accurately sort the coins. In prior art systems similar readings might be provided by bouncing coins and overlapping coins, meaning that coins would need to be rejected to avoid inaccuracies. When overlapping and bouncing coins can be distinguished this means that overlapping coins can be identified and hence subsequently validated and sorted, because the control system can indicate the degree of overlap and this permits suitable inductor and diameter readings to be selected to identify the overlapping coins.

The sensor apparatus preferably includes a third sensor array. The third sensor array may be located along the coin path in alignment with one of the first and second arrays, i.e. at the same distance along the coin path, whilst being offset from the aligned array in the transverse direction. The use of a third array enables the sensor array to measure a greater range of coin diameters. The third array may be spaced apart from both of the first and second arrays in the offset direction, overlapping with neither array. Alternatively the third array may be spaced apart from the aligned array and overlapped with the other of the first and second arrays, in the transverse direction.

Viewed from a second aspect, the present invention provides a method of coin validation comprising the use of first and second sensor arrays placed on a coin path to detect coins on the coin path, wherein the sensor arrays are linear optical arrays extending transverse to the direction of movement of coins along the coin path and wherein the arrays are offset from one another in the direction of movement of the coins.

Preferably, the arrays are also offset from one another in the transverse direction.

In preferred embodiments the method includes determining data relating to the dimensions of the coin and also to the velocity and/or speed of the coin.

The first and second arrays may be located as discussed above in relation to the first aspect

Preferably, the method comprises determining coin speed and direction of travel based on data from the two sensor arrays and on the time of the data. For example, the direction of travel may be determined based on the position of the array that is first affected by the coin. The method may include determining coin speed based on the difference between the times at which the coin is detected by the first and second arrays, and on the distance between the arrays. The determination of coin speed may also take into account the degree of obscuration of the sensor array(s) by the coin, and the control system may be arranged to calculate the position of the coin relative to the sensor.

In a preferred embodiment the method includes determining that a coin is a bouncing coin when the coin is travelling in the opposite direction to the usual direction of coins, and/or when the coin speed is lower than an expected speed.

Preferably, the method includes a step of determining when coins overlap, and more preferably a step of determining the degree of overlap. This may include measuring or calculating the diameter of the coins and the combined length of the overlapped coins in the direction of movement along the coin path. The sensor array may be used to gather data to measure diameter in any conventional fashion, for example by measurement of the maximum degree of obscuration of the linear array or arrays. The combined length of the overlapped coins may be determined based on the coin speed and the time for which the coins obscure an array or both arrays. If the coin diameters and the combined length are known then the degree of overlap can be determined based on the difference between the sum of the diameters and the combined length.

Alternatively or in addition the control system may be arranged to use geometrical considerations to determine the degree of overlap.

The degree of overlap is preferably utilised to determine if other measurements in a broader coin validation system are accurate. Thus, the method may comprise determining when inductor measurements, or other coin measurements, will give accurate results and hence may comprise a step of providing data on the required timing for inductor readings.

In a particularly preferred embodiment the method includes determining if a coin is bouncing and also determining the degree of overlapping for overlapped coins.

In a further aspect, the invention provides a coin validator comprising a sensor apparatus as described above. Preferably the coin validator further comprises an inductor device for measurement electromagnetic properties of the coin. A control system of the coin validator, which may include or be linked to the control system for the sensor apparatus, may be arranged to use data from the sensor apparatus to determine when the inductor measurements will be accurate.

The inductor device may be an inductor core with a C-shape, preferably a C-shape with elongated tips, wherein the tips have a length as long as a coin guide of the coin validator in order to extend across the full range of sizes of coins that may pass along the coin guide. This arrangement of inductor core allows an accurate measurement of the coin properties.

In a preferred embodiment the validator includes a coin guide. The coin guide preferably has a vertical extent when in use, such that coins may pass along the coin guide rolling along their circumference. With this arrangement one of the first and second arrays may be a lower array positioned at a lower part of the coin path or below the coin path thereby enabling measurement of the coin diameter.

A preferred embodiment is a coin validator with a coin guide is for guiding a coin over the sensor arrays and includes windows covering the sensor arrays. The windows should permit passage of light for the photodiodes, and protect the photodiodes from dirt that may enter the validator via the coin path.

In another aspect, the invention provides a coin validator apparatus comprising a coin path passing through a sensor apparatus, the sensor apparatus comprising a light source for generating one or more beams of light extending across the coin path and an optical detector comprising at least two linear optical array devices arranged adjacent an opposite side of the coin path to the light source, the at least two linear optical array devices extending longitudinally in a direction that is substantially perpendicular to a direction of coin travel and substantially parallel to a plane of a coin travelling along the coin path, wherein the linear optical array devices are positioned offset with respect to each other in the direction of coin travel. The sensor apparatus may include features a described above.

Certain preferred embodiments will now be described by way of example only and with reference to the accompanying drawings in which:

Figures 1 a through 1 d show prior art arrangements of optical sensor arrays;

Figure 2 illustrates the problem of bouncing coins;

Figure 3 shows how sensors offset from each other along the coin path can provide information about bouncing coins;

Figure 4 illustrates another offset sensor array arrangement, along with an exemplary geometric calculation of the degree of overlap for overlapping coins;

Figure 5 illustrates an alternative arrangement of sensor arrays;

Figure 6 shows the sensor arrays of Figure 5 along with a diagram of measurement location and data points that may be used in coin validation;

Figure 7 illustrates a further alternative arrangement of sensor arrays;

Figure 8 shows the sensor apparatus installed in a coin validator system with an inductor;

Figure 9 is a schematic showing power supply and other connections for the arrays;

Figure 10 shows an optical configuration that can be used with embodiments of the sensor array;

Figures 1 1 and 12 show an alternative optical configuration including an inductor;

Figure 13 shows a further alternative optical configuration; and

Figures 14 and 15 show the optical system and inductor along with a coin guide.

As noted above, Figures 1 a through 1 b show known arrangements of sensor arrays. These require single file coins and cannot accurately identify overlapping coins or discriminate between bouncing and overlapping coins. Figure 2 shows the problem of bouncing coins with a sensor array arrangement similar to that of Figure 1 c. A first coin 1 is passing along the coin path in the correct direction, as indicated by the solid arrow 14. When a second coin 2 comes into contact with the first coin 1 at touching point 15 it can bounce backward along the coin path as shown by the dashed outline of the bouncing coin 2' and the dashed arrow 16. The linear sensor array of Figure 2 cannot discriminate between the bouncing coins 2, 2' and overlapping coins.

This issue is addressed when two sensor arrays are provided that are offset in the direction of coin travel. Figure 3 illustrates how this solution arises. Since the two offset sensors will be obscured by the coin at different times then the readings they provide can be processed to determine the direction and speed of travel of the coin. Thus, the bouncing coin 2' can be identified.

Figure 4 illustrates an offset array that utilises this advantageous offset in the direction of travel in combination with an offset transverse to the direction of coin travel. The transverse offset means that the same two sensor arrays A, B that provide readings relating to direction of travel can also be used to provide a continuous measurement path along a line almost equivalent to double the length of a single array.

The linear arrays depicted in Figure 4 and used in the preferred embodiment are ball grid array CMOS linear sensor arrays with an array of n x 1 pixels running transversely to the direction of coin movement. An example of a suitable array is that supplied by Texas Advanced Optoelectronic Solutions Inc., of Piano, Texas, USA and given the designation TSL1401 CS-LF. These arrays consisting of a 128 ^χ 1 array of photodiodes, associated charge amplifier circuitry, and a pixel data-hold function that provides simultaneous-integration start and stop times for all pixels. The pixels measure 63.5 Mm (H) by 55.5 pm (W) with 63.5-pm center-to-center spacing and 8-pm spacing between pixels. It will be appreciated that any other suitable photodiode array or equivalent sensor device could be used. The sensor arrays are located in a coin path and a light source is placed so that coins passing along the coin path cast a shadow on to the sensor arrays. Aside from the particular location of the sensor arrays they are used in generally conventional fashion in order to take a measurement of the shadow cast by the coin.

Since the arrays closely overlap in the transverse direction they provide a combined array of almost twice the size of a single array in this direction. The arrays are overlapped by at least a pixel to ensure that continuous readings are possible.

The offset arrays A, B can produce an image of the coin based on the coin's shadow. A coherent light source such as a laser is used to generate the shadow. Alternatively it can be possible to use a non-coherent light source, but in this case resolution is reduced and so features such as milling detection may not be achievable.

The arrays are positioned so that they cover the entire diameter of the coin, i.e. so that even the largest of the coins to be measured does not obscure the entirety of the arrays.

Where the coin image shows overlapping coins and it is determined by measurement of the direction of travel that the coins are not bouncing, then the data from the sensor arrays is used to determine the degree of overlap. This determination may involve a comparison of the diameters of the coins and the combined coin length. A geometric calculation may also be carried out to calculate the area of overlap, as set out below and illustrated in the lower part of Figure 4.

Example calculation:

The distance between the centres of two coins C1 and C2 is equal to 10 mm. The example coins have equal radii of 10 mm. O is the centre of C2 and P is the centre of C1.

The overlapping area is made up of two equal parts. The idea is to find one part then multiplies it by 2. Since the two coins have equal radii, M is the midpoint of segment OP.

Distance OP = 10 mm so that

d(0,M) = 5 mm

We now use the cosine to find angle BOM.

cos(angle BOM) = d(0,M) / OB = 5 / 10 = 1 / 2

Using arccos function, we obtain,

angle BOM = 60 degrees.

Angle AOB is twice angle BOM, hence,

angle AOB = 120 degrees.

The overlapping area can be calculated by determining the difference between area of sector OBPA and area of triangle AOB. The area of sector OBPA is given by:

sector area = (1 / 2) * r ² * angle BOA = (1/2) 10 ² (120 * pi / 180) = 100 pi / 3

The area of triangle OBA is given by:

area of triangle = (1/2) r * r * sin(120 degrees) = 50 sqrt(3) / 2

The total overlapping area is twice the difference.

overlapping area = 2(100 pi / 3 - 50 sqrt(3) / 2) = 122.8 mm ² (approximated to one decimal place).

The overlapping area calculation can be very useful information to know in order to determine the coin dimensions when one coin is covering the other. It can be used as an indicator of a degree of overlap to be used to see if both coins can be accepted. For example the optical sensors can be combined with a "U or C" shaped ferrite inductor core that takes an electromagnetic reading, for example as disclosed in US 7520374 to precisely determine the coin metals content. Knowledge of the degree of overlap enables the inductor measurement at a known point of non-overlapping. As a consequence overlapping coins, which have been determined to be non-bouncing coins, can then be identified and accepted instead of rejecting them.

Figure 5 illustrates an alternative embodiment. Two offset sensor arrays are used as shown in Figure 4, and in addition there is a third lower array. The third array is aligned with, which aligns with one of the offset arrays in the direction along the coin path, but is offset from both arrays in the transverse direction. This arrangement enables larger coins to be measured without the need for the use of longer arrays. In Figure 5 the large coin 8 casts a shadow over the sensor apparatus but does not obscure the entirety of the uppermost sensor array. The three arrays are positioned in such manner that they can measure the smallest coins and the largest one without obscuring the whole pixels.

The use of multiple shorter arrays can save component cost and also lead to easier maintenance. Typically the lower portions of the sensor apparatus will become dirty more quickly. If the lower part of the sensor apparatus is a separate array then this array can be disconnected and replaced without the need to replace all of the arrays.

Figure 6 provides an example of the identification of inductor measurement points in the context of the sensor array arrangement of Figure 5. A coin travelling along the coin path, from right to left in the Figure, casts a shadow on the first light sensor A and subsequently casts a shadow on the second offset sensor B. The timing of the readings from the offset sensors can be used to determine the direction of travel and also the velocity of the coin. An inductor measurement device is connected with a control system for the sensor apparatus and hence receives signals from the light sensors. These signals are used for calculating a preferred point for inductor measurement and also based on the inductor measurement for calculating a material and or thickness value of the coin. As the coin casts a shadow 8 on the sensor 10, signals from starting and finishing ranges of sampling points are identified.

Figure 6 shows sampling points on a coin profile and also shows parts on the coin profile that will cast a shadow. For ease of reference, the coin image is shown as if the coin slides across the sensors, but it will be appreciated that in practice a coin on a coin rail will roll. The lower chord shows the parts of the coin that will cast a shadow on the lowest sensor array, which is the third array in this embodiment. The centre chord a'-a", which in this example is approximately across the centre of the coin, indicates the limit of parts of the coin that will cast a shadow on the first and second arrays A, B. The upper chord b'-b" indicates the line above which the coin shadow will fall only on the upper sensor B. The core 18 of the coin passes the upper sensor in the first array A.

Using the data from the sensor arrays in combination with readings from an inductor device in the coin validator it is possible to take inductor values at an appropriate point on the rim to ensure an accurate measurement of coin thickness/material. For example, with a single coin it may be desirable to take induction values approximating to the location of the truncated vertical chords extending downward from points b' and b".

The radius value is calculated by matching normals to tangents at the points at which measurements are taken, each normal having a common centre point within the coin. The centre point is determined using a least squares fit.

The sensor apparatus also optically detects and count grooves (milling) 17 in an edge of a coin. This is done by sensing and calculating a dimensional characteristic of a coin along the point where the edge of the coin casts a shadow on the upper sensor array B. A signal processor uses an iterative square root method to perform the calculations.

The sensor apparatus of the Figures would generally used in a larger coin validator system that included not only an inductor device but also other features as explained below. In a typical system, coins, preferably after cleaning, e.g. using a hopper, are singulated by a coin pickup assembly configured to reduce jamming. A coin rail assists in providing separation between coins as they travel past a sensor. The sensor provides a Laser set up and an oscillating electromagnetic field generated on a single sensing core. The Laser set up is composed by one or more diodes, two or more CMOS linear arrays a lens group; the oscillating electromagnetic field is composed of one or more frequency components. The electromagnetic field interacts with a coin, and these interactions are monitored and used to classify the coin according to its physical properties. All frequency components of the magnetic field are phase- locked to a common reference frequency. The phase relationships between the various frequencies are fixed, and the interaction of each frequency component with the coin can be accurately determined without the need for complicated electrical filters.

The inductor sensor takes a measurement of the coin close to the optical array and generally just beyond the optical array, in the direction of coin movement. The inductor will include a core, preferably ferrite, which is curved, such as in a U-shape or in the shape of a section of a torus, and defining a gap is provided with a wire winding for excitation and/or detection. The sensor can be used for simultaneously obtaining data relating to two or more parameters of a coin or other object, such as size and conductivity of the object. Two or more frequencies can be used to sense core and/or cladding properties. Objects recognized as acceptable coins, using the sensor data, are diverted by a controllable deflecting door, to tubes for delivery to acceptable coin bins.

Figure 7 shows a further alternative arrangement of offset sensor arrays. In the arrangement of Figures 5 and 6 the lower and centre arrays are aligned along the direction of coin movement and spaced apart from one another in the transverse direction, with the upper array being offset in the direction of coin movement. In the Figure 7 arrangement the top and bottom sensor arrays are aligned in the direction of coin movement and the centre array is offset. In the transverse direction the sensors are arranged so that both the top and bottom sensor overlap with the centre sensor. The arrangement of Figure 7 can make it easier to detect and identify non-circular coin geometries, such as the UK fifty pence piece.

The sensor arrangement of Figures 5 and 6 is shown in Figure 8 installed in a coin validator system with an inductor core 9. The inductor 9 is a U-shaped ferrite core of the type that is disclosed in US 7520374. The photodiode sensor arrangement 10 is in the coin path preceding the inductor core 9. The photodiodes 10 and inductor 9 are held in a housing 20. For illustrative purposes the inductor core 9 and housing 20 are shown in two parts. Coins pass along the coin path in front of the sensors 10 and between the two ends of the ferrite core 9. The data from the photodiode sensor arrays is used to determine the point at which inductor measurements should be taken, as described above.

Figure 9 shows a schematic of the electrical connections for the photodiode sensor arrays of the preferred embodiment. The three arrays U1 , U2, U3 each provide signals to a microcontroller. A 0.1 \JF bypass capacitor is connected between VDD and ground as close as possible to the arrays. The HOLD pin on the device is normally connected to the SI pin in single-array operation. In multi-array operation of n die, the HOLD pin is used to provide a continuous scan across the n die. Note that there is a single AO signal when used in this mode. Alternately, the individual die may be scanned all at once by connecting the individual SI and HOLD lines and reading the AO signals in parallel.

The photodiodes can be illuminated by a single light source and a suitable arrangement of lenses. However, in preferred embodiments multiple light sources are used, taking the form of laser diodes. Figure 10 shows one arrangement, wherein three laser diodes 1 1 are directed at two lenses 12, 13. For this example the photodiode sensors 10 are laid out as in the embodiment of Figures 5 and 6. Coins pass along a coin path in between the laser diodes 1 1 and photodiodes 10 in the manner shown in Figure 5. The first lens 12, at the lower part of the device as shown in Figure 10, directs light from the lower laser diode to the lower photodiodes. The other lens 13 at the upper part of the device (as in the orientation shown in Figure 10) directs light from the upper two laser diodes toward the upper two photodiode arrays, which are offset from one another both transversely and longitudinally. The lenses 12, 13, photodiode arrays 10 and laser diodes are supported in a housing 20.

Figures 1 1 and 12 show an alternative optical configuration and also show the inductor core 9 as in Figure 8. The housing 20 holds the optical parts and includes additional housing elements (not shown) that support the inductor 9. The first lens 12, once again shown at the lower part of the Figures, is similar to the first lens 12 in the Figure 10 embodiment. However in Figure 1 1 second and third lenses 14a, 14b are used in place of the second lens 13 of Figure

10. The second and third lenses 14a, 14b direct light from the two upper laser diodes 1 1 to the upper photodiodes 10.

Another alternative optical configuration is shown in Figure 13. In this arrangement only a single laser diode 1 1 is used, and via a lens and light guide arrangement 19 light is supplied from this single laser diode to all three of the photodiode arrays 10.

The photodiode arrays 10 in these two optical configurations receive light from the laser diodes 1 1 and a coin passing along the coin path casts a shadow on the photodiodes. Data from the photodiodes 10 is used to detect coins in the coin path as described above.

Figures 14 and 15 show an embodiment including the housing 20 that holds the optical sensor parts 10, 1 1 , 12, 13, 14 and inductor core 9 in side view, along with a coin guide 21 holding a coin 8. The coin 8 passes along the coin path past the photodiodes 10 and inductor 9 enabling information to be gathered about the coin 8 in the manner discussed above. In the perspective view of Figure 15 the coin 8 is not shown, which means that the location of the photodiodes 10 is visible. In this embodiment the photodiodes 10 are behind windows 22 in the coin guide 21. The windows 22 protect the photodiode arrays 10 from dirt and dust.

The features of certain preferred embodiments are set out in the following numbered clauses:

1 . A coin validator apparatus comprising a coin path passing through a sensor apparatus, the sensor apparatus comprising a light source for generating one or more beams of light extending across the coin path and an optical detector comprising at least two linear optical array devices arranged adjacent an opposite side of the coin path to the light source, the at least two linear optical array devices extending longitudinally in a direction that is

substantially perpendicular to a direction of coin travel and substantially parallel to a plane of a coin travelling along the coin path, wherein the linear optical array devices are positioned offset with respect to each other in the direction of coin travel.

2. Apparatus as described in clause 1 , wherein the at least two linear optical array devices are also staggered in a direction substantially perpendicular to the direction of coin travel.

3. Apparatus as described in clause 1 or 2, wherein the at least two linear optical array devices are arranged in a substantially end to end configuration.

4. Apparatus as described in clause 1 , 2 or 3, wherein the at least two linear optical array devices are arranged substantially parallel to one another.

5. Apparatus as described in any preceding clause, wherein the at least two linear optical array devices are n x 1 array devices where n is the number of detection pixels in the array.

6. Apparatus as described in any preceding clause, wherein the sensor apparatus comprises a control system that is arranged to determine if coins are overlapped, and if overlap is determined, to alter a validation process to take account of the overlap.

7. Apparatus as described in any preceding clause, wherein the at least two linear optical array devices are arranged as upper and lower array device elements of a staggered linear array that is positioned above a coin rail that the coins are arranged to roll along, the lower array device element being spaced from the coin rail.

8. Apparatus as described in clause 7, wherein the lower array device element is spaced by at least the length of a linear optical array device from the coin rail.

9. Apparatus as described in clause 7 or 8, wherein the at least two linear optical array devices are positioned at or close to an entrance of an induction sensor.

10. Apparatus as described in any of clauses 7 to 9, wherein the sensor apparatus includes an induction sensor comprising an electromagnetic core and the at least two linear optical array devices are mounted on the induction sensor. 1 1. Apparatus as described in clause 10, wherein optical detection of a coin as it crosses the one or more light beams for the lower array device of the staggered linear array triggers a start point for an induction sensor measurement.

12. Apparatus as described in clause 1 1 , wherein optical detection of a coin as it crosses the one or more light beams for the upper array device of the staggered linear array triggers an end point for the induction sensor measurement.

13. Apparatus as described in clause 12, wherein a loss of optical detection of the coin as it exits the one or more light beams for the upper array device of the staggered linear array triggers a start point for a second induction sensor measurement for the coin.

14. Apparatus as described in clause 13, wherein a loss of optical detection of the coin as it exits the one or more light beams for the lower array device of the staggered linear array triggers a finish point for the second induction sensor measurement of the coin.

15. Apparatus as described in any of clauses 7 to 14, wherein a third linear optical array device is arranged close to the coin rail below the staggered linear array of the upper and lower linear optical array devices.

16. Apparatus as described in clause 15, wherein the third linear optical array device is arranged offset with respect to the lower linear optical array device and aligned with but spaced from the upper linear optical array device.