Classifications

• • G—PHYSICS
  • G07—CHECKING-DEVICES
  • G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
  • G07D5/00—Testing specially adapted to determine the identity or genuineness of coins, e.g. for segregating coins which are unacceptable or alien to a currency
• • G—PHYSICS
  • G07—CHECKING-DEVICES
  • G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
  • G07D2205/00—Coin testing devices
  • G07D2205/001—Reconfiguration of coin testing devices
  • G07D2205/0011—Reconfiguration of coin testing devices by downloading test parameters, e.g. remotely

Definitions

• the invention relates to a method for calibrating a coin validator.
• a coin validator has the task of examining inserted coins for properties which the coins to be accepted should have.
• the properties include, for example, the material, the dimensions such as thickness and diameter, the transmission for light, the formation of the embossed edge and image, the weight, the hardness, etc.
• the materials are typically tested using inductive coils, the field of which corresponds to the material the coin interacts. This results in a typical damping in the inductive sensors, the extent of which contains a statement about the material or the material composition.
• Known electronic coin validators are able to check a number of different coin values. They have a microprocessor with a programmable memory for recording reference values to be compared with the measured values. In order to meet tolerances, it is customary to provide an upper and a lower reference value for each coin type, whereby a so-called acceptance band is formed. Before a coin validator reaches the user, the reference values must be saved in accordance with the coin set to be checked. Although it is conceivable to calculate the reference values mathematically, it has been shown in practice that this method is not acceptable. The mechanical or electrical properties of a coin validator are subject to more or less strong, mostly manufacturing-related fluctuations, which find their way into the measuring signals emitted by the measuring probe. It is therefore necessary to determine and program the reference values device-specifically.
• test coins Selected real coins, the properties of which are to be checked in the desired distribution within the acceptance band, are thrown into the device to be verified. The reference values are determined and stored with the aid of the measurement signals obtained. Since test coins wear out over time, new ones have to be used again and again. This turns out to be cumbersome and difficult. It is also known to use so-called tokens or test disks instead of test coins, which have analog physical properties and which are produced specifically for test purposes. However, the processes for producing tokens are also relatively complex.
• the known method requires fewer test coins or disks, it does not do without a minimum of coins or disks.
• a further disadvantage is that the reference values are read into the memory during the test phase.
• coin validators it is usually not yet known for which currencies and, accordingly, for which coins they are used. It is therefore left to a later manufacturing step to calibrate the device in the manner described when the set of coins to be accepted has become known to a currency.
• the invention is therefore based on the object of specifying a method for calibrating a coin validator which further simplifies the calibration procedure.
• the invention uses a test coin or caliber disc different from the coin to be accepted. It is checked in the coin checking section of a coin acceptor to be verified.
• the parameter value obtained from the measurement signal of the caliber disc is entered directly into the programmable memory of the coin validator. With several caliber disks, several parameter values are generated accordingly. This process can take place, for example, as the last stage in production. As will be shown later, the insertion of a single caliber disc may be sufficient to generate a parameter value. This process step is applied regardless of which coin types are to be checked later, for all produced coin validators.
• the coin validators are standardized in this way with a view to later adjustment.
• the coin validator can be calibrated.
• the parameter value is transferred to a computer, which uses the parameter value from a correlation function to determine at least one reference value for a coin. It goes without saying that at least one correlation function must be present for each coin type to be checked. If an acceptance band is specified, correlation functions are formed for the lower and upper band limits.
• the correlation function which is stored in an external memory, for example of the computer, is determined from the relation of the size of a set of parameter values and the size of a set of measured values of at least one standard coin for at least one coin type, the values being checked by multiple checks in the coin checking section Majority of coin acceptors can be won.
• coin tests are carried out in a number of coin validators.
• a correlation function is calculated mathematically by comparing the measured values for caliber disks and standard coins. According to one embodiment of the invention, this can be done, for example, by determining from mean values which are formed from the measured values for the individual coin validators when the test and standard coins are checked several times.
• the correlation function can be mathematically formed using a so-called Taylor series.
• a correlation function can therefore be determined for each coin type for a specific type of caliber disk.
• the correlation functions can be stored in a database. If a specific coin validator is now to be verified, the parameter value stored in its memory is transferred to a computer.
• the computer searches the associated correlation function from the database after having entered the type of coin for which the calibration is to be carried out. Since, as already mentioned, not a single reference value is used, but rather a so-called reference or acceptance band, it can also be specified whether the acceptance band should be wide or narrow.
• the computer calculates the upper and lower limit of the reference band and then transfers these reference values into the programmable memory of the coin validator. The calibration is now complete.
• determining the correlation function requires a certain amount of effort. As described, a large amount of data can be determined from a large number of coin validators and with the help of caliber disks and test coins, from which the function can then be determined by mathematical approximations. However, this effort is not higher than with the conventional method, because standard parameter values and standard reference values are also required for this. These standard values can also only be determined by taking a large number of measurements with a plurality of devices in order to determine the standard values statistically.
• the calibration method according to the invention can therefore be carried out in the shortest possible time. It is also advantageous that the calibration process does not necessarily have to be part of the manufacturing process, but can be carried out at any later time and also at another location.
• an individual value is considered an outlier if it shows a deviation from the median that occurs less frequently on average than every twenty-fifth throw-in.
• the numbers X and Y are now also used for lower throw-in numbers, e.g. 5.
• the Taylor series for the correlation function is calculated from the comparison of the mean values, using the mathematical method of the so-called minimal quadratic deviation.
• a function that calculates the reference values from the mean values of the caliber discs This function should also determine the measured reference values (using standard coins) as precisely as possible. In general, however, a somewhat different value FKal will result from the function of the caliber values for the measured reference value YMü:
• the method of minimal square deviation now tries to adapt YMü and FKal as precisely as possible, in such a way that the square of YMü - FKal, which is available for each of the 50 devices and is summed over all 50 devices, is made as small as possible.
• the condition described cannot be used alone to determine the correlation function.
• the Taylor series approximately represents the function sought.
• the parameters must be specified which describe the dependence of the function on certain powers of the caliber disc values. In principle, the number of these parameters is infinite, and the approximation is to allow only a few of them to be non-zero. Which of these is to be decided for each coin and each measured value.

Landscapes

• Physics & Mathematics (AREA)
• General Physics & Mathematics (AREA)
• Testing Of Coins (AREA)

Applications Claiming Priority (2)

Publications (2)

Family

ID=6475548

Family Applications (1)

Country Status (3)

Cited By (4)

Citations (3)

• 1992
  • 1992-12-17 DE DE4242639A patent/DE4242639C2/de not_active Expired - Fee Related
• 1993
  • 1993-12-03 EP EP93119509A patent/EP0602474B1/fr not_active Expired - Lifetime
  • 1993-12-03 DE DE59307568T patent/DE59307568D1/de not_active Expired - Lifetime
  • 1993-12-03 ES ES93119509T patent/ES2109414T3/es not_active Expired - Lifetime

Patent Citations (3)

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events