JP4369626B2 - Coin identification sensor and coin identification method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a coin identification sensor provided in an ATM, a change machine, a vending machine, a ticket machine, or a coin identification method using the coin identification sensor. The present invention relates to a coin identifying sensor and a coin identifying method capable of accurately identifying the type of a coin based on determination.
[0002]
[Prior art]
Recently, many coin identification sensors and coin identification methods have been proposed for identifying the type and authenticity of coins using the magnetic or electrical characteristics of coins. This type of coin identifying sensor includes a pair of independent cores facing each other with a gap for passing coins, a transmission coil wound around one core in order to generate a magnetic field in the gap, In order to detect a change in magnetic flux of the magnetic field accompanying the passage, a receiving coil wound around the other core is provided. As a coin identifying method using the coin identifying sensor, there is known a method for identifying the type of a coin based on a change in output voltage of the receiving coil.
[0003]
[Problems to be solved by the invention]
However, in the identification of coins using the conventional coin identification sensor or coin identification method, there are the following problems.
1) Since each coil is wound around an independent core, the magnetic coupling is weak, and the magnetic characteristics of the coin cannot be clearly shown in the output waveform of the receiving coil.
2) Since a coin is identified from a slight difference in output waveform, a highly accurate detection circuit is required, resulting in an increase in cost.
3) Since the change in the output voltage of the receiving coil depends on the material and thickness of the coin, there is a possibility that a modified coin crafted with a thickness corresponding to the material may be mistakenly recognized as a genuine coin.
[0004]
[Means for Solving the Problems]
The present invention has been created in view of the above circumstances and has been created for the purpose of solving these problems, and is a coin identifying sensor used for identifying a coin, wherein the coin identifying sensor is a coin. A core that forms a loop-shaped magnetic path with a gap for passage, an excitation coil wound around the core to generate a magnetic field in the gap, and a magnetic flux change of the magnetic field accompanying the passage of coins. A detection coil wound around the core for detection, and a natural frequency / natural period specifying means for specifying a natural frequency or a natural period depending on the material of the coin based on an output waveform of the detection coil. It is a feature. In other words, by winding an exciting coil and a detecting coil around a core formed in a loop shape with a gap for passing coins, magnetic energy generated by applying the exciting coil is concentrated in the gap and magnetically coupled. As a result, the difference in coin material and thickness can be clearly expressed in the output waveform of the detection coil. Moreover, since the natural frequency or natural period depending on the material of the coin can be specified, the material of the coin can be accurately determined without being affected by the thickness of the coin. Based on this, it becomes possible to accurately identify the type of coin.
The natural frequency / natural period specifying unit includes a pulse voltage applying unit that applies a pulse voltage to the excitation coil, and a natural period measuring unit that measures a natural period of a response waveform output from the detection coil. It is characterized by. In other words, since the natural period of the coin (the amount depending on the natural frequency) is measured based on the pulse response waveform of the detection coil without sweeping the excitation frequency, the detection circuit can be simplified, Detection accuracy can be increased.
Further, the coin identification sensor includes data comparison means for comparing the natural period data measured by the natural period measurement means or the natural frequency data calculated based on the natural period data with the determination data. It further comprises a material determining means for determining the material of the coin based on the result. That is, since the coin identification sensor includes the material determination means, it is not necessary to perform material determination at the upper control unit on the apparatus body side, and as a result, the processing load on the upper control unit can be reduced. The versatility of the sensor can be improved.
The coin identifying sensor further includes a real part magnetic permeability specifying means for specifying a real part magnetic permeability depending on a thickness of the coin based on an output waveform of the detection coil. That is, since the real part magnetic permeability depending on the thickness of the coin can be specified, it is possible to accurately determine the thickness of the coin.
The real part magnetic permeability specifying means includes pulse voltage applying means for applying a pulse voltage to the exciting coil, and peak voltage measuring means for measuring a peak voltage of a response waveform output from the detection coil. It is a feature. That is, since the peak voltage depending on the magnetic permeability is measured based on the pulse response waveform of the detection coil, the real part magnetic permeability (coin thickness) can be specified with a simple detection circuit.
Further, the coin identifying sensor is a thickness calculating means for calculating the thickness of the coin using the peak voltage data measured by the peak voltage measuring means and a proportional constant prepared in advance corresponding to the material of the coin. Is further provided. That is, since the coin identification sensor includes the thickness calculating means, it is not necessary to perform the thickness calculation at the upper control unit on the apparatus body side, and as a result, the processing load on the upper control unit can be reduced. The versatility of the identification sensor can be enhanced.
In addition, the coin identification sensor includes a data comparison unit that compares the material data determined by the material determination unit and the thickness data calculated by the thickness calculation unit with the identification data; The apparatus further comprises coin identifying means for identifying the type. In other words, since the material and thickness are individually determined and the type of coin is identified based on both determination results, this is compared to the case where the coin type is identified based on a single detection data that depends on the material and thickness. Thus, the identification accuracy can be greatly improved.
Further, the coin identification sensor is based on a coin passage detection means arranged at a predetermined interval in the coin passage direction with respect to the core, a detection signal of the coin passage detection means, and an output waveform of the detection coil. And a size determining means for determining the size of the coin. That is, since the means for determining the size of the coin is further provided, the identification accuracy can be further improved by adding the determination data to the coin identification data.
Moreover, this invention is a coin identification method using the sensor for coin identification of Claims 1 thru | or 8, Comprising: This coin identification method is a pulse voltage application process which applies a pulse voltage to the said excitation coil, and the said detection coil. A natural period measurement step for measuring the natural period of the response waveform output from the data, and a data comparison step for comparing the measured natural period data or the natural frequency data calculated based on the natural period data with the determination data And a material determination step of determining the material of the coin based on the data comparison result. In other words, since the natural period of the coin (a numerical value depending on the natural frequency) is measured based on the pulse response waveform of the detection coil without sweeping the excitation frequency, the detection circuit can be simplified, A highly accurate material determination can be performed based on the measurement result.
The coin identifying method includes a peak voltage measuring step of measuring a peak voltage of a response waveform output from the detection coil, a measured peak voltage data, and a proportional constant prepared in advance corresponding to the material of the coin. And a thickness calculating step for calculating the thickness of the coin. That is, since the peak voltage depending on the real part permeability is measured based on the pulse response waveform of the detection coil, the real part permeability (coin thickness) can be specified with a simple detection circuit.
The coin identification method includes a data comparison step of comparing the material data determined in the material determination step and the thickness data calculated in the thickness calculation step with identification data, and the type of coin based on the data comparison result And a coin identifying step for identifying the item. In other words, since the material and thickness are individually determined and the type of coin is identified based on both determination results, this is compared to the case where the coin type is identified based on a single detection data that depends on the material and thickness. Thus, the identification accuracy can be greatly improved.
[0005]
DETAILED DESCRIPTION OF THE INVENTION
Next, one embodiment of the present invention will be described with reference to the drawings. In the drawings, 1 is a coin identifying sensor, and the coin identifying sensor 1 identifies the type of the coin C and the like based on the sensor unit 3 provided in the coin path 2 and the output waveform of the sensor unit 3. It comprises an identification unit 4. The sensor unit 3 includes a core 5 that forms a loop-shaped magnetic path with a gap G for passing coins, an excitation coil 6 wound around the core 5 in order to generate a magnetic field in the gap G, In order to detect the magnetic flux change of the magnetic field accompanying the passage of the coin C, it is configured to include a detection coil 7 wound around the core 5. The basic principle will be described.
[0006]
The distance l _{g of the} gap G is made narrower (l >> l _g ) than the total magnetic path length l of the magnetic circuit configured in the sensor unit 3, and the core 5 is made of a high permeability material such as ferrite or amorphous. Then, the relative permeability μ _s of the magnetic core becomes μ _s >> μ ₀ , and the magnetic energy of the magnetic circuit is concentrated in the gap G. When an AC voltage is applied to the excitation coil 6 and an AC magnetic field H = H ₀ e ^iωt is generated in the gap G, the change in magnetic flux detected by the detection coil 7 is B = B ₀ e ^{i (ωt−δ)} . And the permeability μ defined by B = μH is
μ = B / H = μ′−iμ ″
μ ′ = (B ₀ / H ₀ ) cos δ, μ ″ = (B ₀ / H ₀ ) sin δ
μ ″ / μ ′ = tan δ (sometimes defined as ≡1 / Q)
Tan δ is an amount related to energy loss, and δ is called a loss angle.
Contrast the frequency dependence of the complex permeability μ = μ′−iμ ″ with the harmonic vibration model,
μ′⇔ (ω ₀ ² −ω ² ) / {(ω ₀ ² −ω ² ) ² + 4k ₀ ² ω ² }
μ ″ ⇔2kω / {(ω ₀ ² −ω ² ) ² + 4k ₀ ² ω ² }
a state where the k ₀ ≒ _{ω 0} shown in FIG. 2 _{(k 0} is the damping constant, omega ₀ is the natural frequency). This relationship is expressed by the Kramers-Kronig relationship and shows good agreement with the experiment.
FIG. 3 shows measured values of various coins C. The actually measured values are in good agreement with the shape of the vibration model when k ₀ ≈ω ₀ . If the coin C is inserted into the gap G and the natural frequency ω ₀ is measured, an amount specific to the material of the coin C can be obtained. In addition, the natural frequency of white copper containing Ni, which is a ferromagnetic material, is close to 1.5 KHz, and the real permeability μ ′ (the real part of the complex permeability) decreases depending on the thickness. .
FIG. 4 shows an example of the change in the real part permeability μ ′ with respect to the frequency of Al and Ni by changing the plate thickness. As shown in FIG. 4, the natural frequency ω ₀ is an amount inherent to the material of the coin C, and the magnitude of the real part permeability μ ′ when ω = ω ₀ depends on the thickness of the coin C. I understand that.
That is, the complex magnetic permeability μ varies depending on the excitation frequency ω, and the frequency ω at which the imaginary part permeability μ ″ is zero and the real part permeability μ ′ is maximum depends on the material of the coin C. and define this frequency and the natural frequency omega _0. then, if the material of the coin C is the same a natural frequency omega ₀ is constant, reduced in proportion to the real part permeability mu 'is the thickness at that time. Therefore, the sensor unit 3 provided in the coin identification sensor 1 of the present invention allows the coin C to pass through the gap G of the magnetic circuit, the natural frequency ω ₀ due to the passed coin C, and the real part permeability at that frequency. By measuring μ ′, the material and thickness of the coin C can be identified.
[0007]
Next, a method for measuring the natural frequency ω ₀ that is a means for realizing the above principle will be described. As a method of measuring the natural frequency ω _0, a method of sweeping the excitation frequency of the excitation coil 6 is the most fundamental, but a method of examining the pulse response waveform of the detection coil 7 was invented as a method of easily obtaining the same effect. .
FIG. 5B shows a response waveform of the detection coil 7 when a pulse waveform as shown in FIG. 5A is applied to the excitation coil 6. The period T of the transient characteristic is determined by the natural frequency ω ₀ ,
ω ₀ = 2π / T
It is. Therefore, by measuring the response waveform period T and the positive and negative peak voltages v ₁ and v ₂ , the material and thickness of the coin C passing through the gap G of the sensor unit 3 can be identified.
[0008]
Next, a measurement circuit that realizes the above-described measurement method will be described. The measurement circuit applies a pulse voltage of 20 V and 2 μ width every 1 msec to the exciting coil 6 to obtain the response waveform period T and positive and negative peak voltages v ₁ and v ₂ of the detection coil 7.
The period T is an amount that depends on the above-mentioned natural frequency ω ₀ inherent to the material. In the case of a material having the same natural frequency ω ₀ , the loss amount is proportional to the thickness.
ω ₀ = 2π / T
Calculated natural frequency omega _0, the peak voltage _v 1, _{v 2} and the linear equation of the thickness _{t c t c = K 1 (} ω 0) v 1 + K 2 (ω 0) of the positive and negative when the _{v 2}
Can be approximated by K ₁ and K ₂ are proportional constants, which are obtained in advance according to the type of material, and a table as shown in FIG. 6 is created as a reference value. The type (material) of the coin C is estimated from ω ₀ obtained _first , and the thickness t _c is calculated from the corresponding proportional constants K ₁ and K ₂ .
[0009]
FIG. 7 is a block diagram of the identification unit 4 that implements the above identification method using a microcomputer (hereinafter referred to as a microcomputer). Pulses as shown in (1) and (3) (enlarged view) in FIG. 8 are applied to the exciting coil 6. The response waveforms of the detection coil 7 at that time are shown in (2) and (4) (enlarged view) in FIG. The detection coil 7 is connected to a zero cross circuit 9 and positive and negative peak hold circuits 11 and 12 via a band pass filter 8, and the zero cross circuit 9 detects a rising edge where the voltage of the response waveform becomes zero. The output of the zero cross circuit 9 is shown in (7) of FIG. In order to obtain the period T, the period detection circuit 13 sets the SR flip-flop with the first zero-cross pulse, resets the SR flip-flop with the second zero-cross pulse, and generates a gate pulse as shown in (8) in FIG. obtain. The pulse width of the gate pulse is a period T. A reference clock for time measurement is counted by the counter 14 while the gate is open. The count value of the counter 14 is proportional to the period T. The count value is taken into the microcomputer 15 and the counter 14 is cleared. Further, the outputs of the positive and negative peak hold circuits 11 and 12 are taken into the microcomputer 15 via the built-in A / D converter, and the positive and negative positive and negative peak voltages v ₁ and v ₂ are obtained. The material (type) and thickness of the coin C are determined from the cycle T and the positive and negative peak voltages v ₁ and v ₂ .
[0010]
Next, the procedure of identification processing will be described based on the flowchart shown in FIG. After measuring the period T as described above, the natural frequency ω ₀ is calculated, and the material is determined from the natural frequency ω ₀ . The material judgment is a set S of types of coins C according to the material.
S = {A, B, C,... N}
Whether _{(<ε | | ω 0 -ω} ξ) is against natural frequency omega _xi] of any material ξ (ξ∈S), matches the natural frequency omega ₀ of measurement, including the range of tolerance epsilon belonging to If it matches with any material ξ, it is determined that the material is ξ, and if it does not match any material ξ, it is determined that it is a fake coin.
Following the material determination, after measuring the positive and negative peak voltages v ₁ and v ₂ as described above, the proportional constants K _1ξ and K _2ξ corresponding to the determined material ξ are referred to from the table, and the calculation formula t _c = K _1ξ v ₁ + K _2ξ v ₂
Is used to calculate the thickness t _c .
After obtaining the material ξ and the thickness t _c , the material ξ and the thickness t _c are compared with the reference value for type determination to determine the type and authenticity of the coin C. The determination result is transmitted to the upper control unit as an identification signal represented by a 16-bit code as shown in FIG. The upper 2 bits of the 16-bit code are true / false codes, the lower 8 bits are thickness codes (hexadecimal notation in 0.01 mm units), and the remaining bits are type codes. As the identification signal communication method, any method such as serial, parallel, and PWM output can be used.
[0011]
Next, the measurement of the size of the coin C by the coin identifying sensor 1 will be described with reference to FIG. When a size determination function is added to the coin identification sensor 1, as shown in FIG. 11A, the passage detection sensor unit 16 (optical type) is placed on the upper side of the coin passage with a predetermined interval L. A sensor, an eddy current sensor, a sensor equivalent to the sensor unit 3). FIG. 11B shows output waveforms of the sensor units 3 and 16 when a coin C having a predetermined size passes through the coin passage 2. The time T _D when the output voltage v ₁ of the sensor unit 3 changes is between the passing speed V and the diameter D of the coin C: D = V · T _D
The relationship holds. If the time from passing the output of the sensor unit 16 to the output voltage v ₁ of the sensor 3 changes with T _P, the coin passing speed V is because it is L / T _P, the diameter D of the coin C,
D = L · (T _D / T _P )
Is required. The obtained diameter data is transmitted to, for example, the host control unit and used as secondary identification data.
[0012]
In the structure configured as described above, the coin identifying sensor 1 used for identifying the coin C includes a core 5 that forms a loop-shaped magnetic path with a gap G for passing coins, and a magnetic field in the gap G. An excitation coil 6 wound around the core 5 and a detection coil 7 wound around the core 5 to detect a change in magnetic flux of the magnetic field accompanying the passage of a coin C, Based on the output waveform of the detection coil 7, the magnetic natural frequency ω ₀ or the natural period T of the coin C is specified. That is, by winding the exciting coil 6 and the detecting coil 7 around the core 5 formed in a loop shape with the gap G for passing coins, the magnetic energy generated by the application of the exciting coil 6 is changed to the gap G. As a result, the difference in the material and thickness of the coin C can be clearly expressed in the output waveform of the detection coil 7. Moreover, since the natural frequency ω ₀ that depends on the material of the coin C can be specified, the material of the coin C can be accurately determined without being affected by the thickness of the coin C, and thus the material determination thereof. Based on the result, the type of the coin C can be accurately identified.
[0013]
In addition, since a pulse voltage is applied to the excitation coil 6 and the natural period T of the response waveform output from the detection coil 7 is measured, it is not necessary to sweep the excitation frequency, and as a result, the detection circuit is simplified. The detection accuracy can be improved.
[0014]
The coin identifying sensor 1 compares the measured natural period T or the natural frequency ω ₀ calculated based on the natural period T with determination data, and based on the comparison result, the material of the coin C Therefore, it is not necessary to perform material determination in the host control unit, and as a result, not only the processing load on the host control unit can be reduced, but also the versatility of the coin identification sensor 1 can be improved.
[0015]
Further, the coin identifying sensor 1 specifies the real part permeability μ ′ depending on the thickness of the coin C on the basis of the output waveform of the detection coil 7, so that the thickness of the coin C can be accurately determined. it can.
[0016]
Further, since the peak voltages v ₁ and v ₂ related to the real part permeability μ ′ are measured based on the pulse response waveform of the detection coil 7, the thickness of the coin C can be specified with a simple detection circuit. It becomes possible.
[0017]
The coin identifying sensor 1 calculates the thickness of the coin C using the measured peak voltages v ₁ and v ₂ and a proportional constant prepared in advance corresponding to the material of the coin C. It is not necessary to perform the thickness calculation in the control unit, and as a result, not only the processing load of the host control unit can be reduced, but also the versatility of the coin identifying sensor 1 can be improved.
[0018]
Further, the coin identifying sensor 1 individually determines the material and thickness, and identifies the type of the coin C based on both determination results. Therefore, the coin identifying sensor 1 is based on a single detection data depending on the material and thickness. Compared with the case of identifying the type of coin C, the identification accuracy can be greatly improved.
[0019]
Further, the coin identifying sensor 1 has a passage detection sensor unit 16 disposed at a predetermined interval in the coin passage direction with respect to the core 5, a detection signal of the sensor unit 16, and the detection coil 7. Since the size of the coin C is determined based on the output waveform, the identification accuracy can be further improved by adding the size determination data to the coin identification data.
[Brief description of the drawings]
FIG. 1A is a front view of a main part of a coin identification sensor, and FIG. 1B is a side view of the main part.
FIG. 2 is a graph showing the relationship between magnetic permeability and natural frequency.
FIG. 3 is a table showing measured values of natural frequencies of various coins.
FIG. 4 is a graph showing the relationship between material and natural frequency, and the relationship between thickness and real part permeability.
5A is a waveform diagram showing an excitation waveform of an excitation coil, and FIG. 5B is a waveform diagram showing a detection waveform of a detection coil.
FIG. 6 is an explanatory diagram of a table that defines the relationship among materials, natural frequencies, and proportional constants.
FIG. 7 is a block diagram illustrating a configuration of an identification unit.
FIG. 8 is a waveform diagram showing an operation of an identification unit.
FIG. 9 is a flowchart illustrating a processing procedure of an identification unit.
FIG. 10 is an explanatory diagram showing the contents of an identification signal.
11A is a side view showing the arrangement of the coin identifying sensor unit and the passage detecting sensor unit, and FIG. 11B is a waveform diagram showing the output of each sensor unit.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Coin identification sensor 2 Coin passage 3 Sensor part 4 Identification part 5 Core 6 Excitation coil 7 Detection coil 8 Band pass filter 9 Zero cross circuit 11 Peak hold circuit 12 Peak hold circuit 13 Period detection circuit 14 Counter 15 Microcomputer 16 Sensor part C Coin G gap

Claims (11)

A coin identifying sensor used for identifying a coin, the coin identifying sensor including a core that forms a loop-shaped magnetic path with a gap for passing coins, and to generate a magnetic field in the gap, Based on the excitation coil wound around the core, the detection coil wound around the core to detect the magnetic flux change of the magnetic field accompanying the passage of the coin, and the output waveform of the detection coil, the material of the coin A coin identification sensor comprising: a natural frequency / natural period specifying means for specifying a dependent natural frequency or natural period. The natural frequency / natural period specifying unit according to claim 1, wherein the natural frequency / natural period specifying unit includes a pulse voltage applying unit that applies a pulse voltage to the exciting coil, and a natural cycle measuring unit that measures a natural period of a response waveform output from the detection coil. A coin identification sensor comprising: 3. The data comparison according to claim 1, wherein the coin identification sensor compares the natural period data measured by the natural period measuring unit or the natural frequency data calculated based on the natural period data with the determination data. A coin identifying sensor, further comprising: means; and material determining means for determining the material of the coin based on the data comparison result. 4. The coin identifying sensor according to claim 1, further comprising: real part magnetic permeability specifying means for specifying real part magnetic permeability depending on a thickness of the coin based on an output waveform of the detection coil. Coin identification sensor. 5. The real part magnetic permeability specifying means according to claim 1, wherein the real part magnetic permeability specifying means is a pulse voltage applying means for applying a pulse voltage to the exciting coil, and a peak voltage measuring means for measuring a peak voltage of a response waveform output from the detecting coil. And a coin identifying sensor. 6. The coin identifying sensor according to claim 1, wherein the coin identifying sensor calculates a coin thickness using peak voltage data measured by the peak voltage measuring means and a proportional constant prepared in advance corresponding to the material of the coin. A coin identifying sensor, further comprising a thickness calculating means. 7. The coin identifying sensor according to claim 1, wherein the coin identifying sensor includes data comparing means for comparing the material data determined by the material determining means and the thickness data calculated by the thickness calculating means with identification data, and a data comparison result. And a coin identifying means for identifying a coin type based on the coin identification sensor. 8. The coin identification sensor according to claim 1, wherein the coin identification sensor is a coin passage detection unit disposed at a predetermined interval in the coin passage direction with respect to the core, a detection signal of the coin passage detection unit, and the detection coil. A coin discriminating sensor, further comprising: a size determining means for determining the size of the coin based on the output waveform of the coin. 9. A coin identifying method using the coin identifying sensor according to claim 1, wherein the coin identifying method includes a pulse voltage applying step of applying a pulse voltage to the exciting coil, and a response waveform output from the detecting coil. A natural period measuring step for measuring the natural period of the data, a data comparison step for comparing the measured natural period data or the natural frequency data calculated based on the natural period data with the determination data, and the data comparison result And a material determining step for determining the material of the coin. The coin identification method according to claim 9, wherein the coin identification method includes a peak voltage measurement step of measuring a peak voltage of a response waveform output from the detection coil, and a proportionality prepared in advance corresponding to the measured peak voltage data and the material of the coin. And a thickness calculation step of calculating the thickness of the coin using a constant. 11. The coin identifying method according to claim 9, wherein the coin identifying method includes a data comparing step for comparing the material data determined in the material determining step and the thickness data calculated in the thickness calculating step with identification data, and a data comparison result. A coin identifying method, further comprising: a coin identifying step for identifying a coin type based on the coin identifying process.

Images