FR2661509A1 - METHOD FOR CONTROLLING MEASUREMENTS CARRIED OUT IN A TESTING AND SORTING APPARATUS OF MINIATURE ARTICLES - Google Patents

Abstract

The present invention relates to a method for monitoring measurements made in an apparatus for testing and sorting miniature articles. According to this process: a reference article having stable characteristics over time is positioned permanently in one of the cells; - in a first step, at least one of the characteristics of said article is tested N times (N> 1) to determine the mean and the standard deviation of said characteristic; - for the mean and the standard deviation of said characteristic, at least two framing values are determined; - in a second step, said characteristic of the reference article is tested N 'times (N'> 1) in order to determine, again, the mean and the standard deviation of said characteristic; - comparing the determined mean and standard deviation with their respective framing values; - if the mean and the standard deviation are outside the framing values, either stop the testing and sorting apparatus, or start the second step again for checking; - if the mean and the standard deviation are within the framing values, the measurements are continued by resuming the process at the level of the second step. Application in particular to capacitors.

Description

METHOD OF CONTROLLING MEASUREMENTS CARRIED OUT

IN AN APPARATUS FOR TESTING AND

SORTING MINIATURE ARTICLES

  The present invention relates to a method for controlling the measurements made in an apparatus for testing and sorting miniature articles such as capacitors, ferrites

or the like.

  The devices for testing and sorting miniature articles such as chip capacitors are well known to those skilled in the art. As shown in FIGS. 1 and 2, these devices essentially comprise, mounted on a frame 1, a transfer device consisting of a cylindrical bowl 2 whose axis, in the embodiment shown, is inclined relative to the axis of the frame This cylindrical bowl 2 receives a round plate 3 whose periphery is provided with cells 4 intended to receive articles 5 to be measured and sorted The articles are brought into the bottom of the bowl by means of a feeder-type bowl-vibrator not shown in Figure 1 The items that fall into the bowl take therefore place in the cells of the plate and are rotated under the contacts 6/7 of a measuring head 8 shown schematically in Figure 2 to use the same sorting device for the measurement of size articles or fo In the same way, the plate 3 is removably fixed by means of a screw 9 in the bottom of the bowl 2. Depending on the results obtained at the measuring head 8, the articles are sent into sorting boxes mounted. inside the drawer 10 using

  an ejection system not shown.

  The calibration of the test and sorting apparatus is carried out systematically according to a defined periodicity, for example every six months. However, the currently used capacitor test and sorting devices do not make it possible to detect a certain number of defects such as poor contact between the measuring clamps and the workpiece due to oxidation of the contacts, defective positioning of the workpiece

  under the contacts or a wrong measurement of the device.

  The object of the present invention is to provide a method for controlling the measurements made in a device for testing and sorting miniature articles such as capacitors.

  which overcomes the disadvantages mentioned above.

  Consequently, the subject of the present invention is a method for controlling measurements made in a device for testing and sorting miniature articles comprising a transfer device provided with cells for receiving the articles to be tested and sorted, characterized in that : one permanently places, in one of the cells, a reference article presenting stable characteristics in time; in a first step, N times (N> 1) is tested at least one of the characteristics of said article to determine the mean and the standard deviation of said characteristic; for the average and the standard deviation of said characteristic, at least two framing values are determined; in a second step, it tests N 'times (N'> 1) said characteristic of the reference article to determine, again, the mean and the standard deviation of said characteristic; the average and standard deviation determined are compared with their respective framing values; if the mean and the standard deviation are outside the framing values, either the test and sorting device is stopped, or the second step for checking is repeated; if the mean and the standard deviation are within the framing values, the measurements are continued by repeating the procedure

at the level of the second stage.

  The process described above is a permanent and dynamic process which makes it possible to follow the drift or the dispersion of the

  measurements of the measuring head over time.

  Thus, the standard deviations make it possible to follow a possible dispersion in the measurements of the reference article due to a poor quality of the contacts, a bad positioning of the reference article under the contacts of the reference article.

  measurement or failure of the sorting and testing equipment.

  Preferably, according to another feature of the present invention, for the mean and the standard deviation, two pairs of framing values, called monitoring limits and control limits, are determined, and when the average and / or the The standard deviation is outside the control limits, the test and sorting device is stopped, and when the mean and / or standard deviation are outside the monitoring limits, the procedure is repeated. level of the second step and the test and sorting apparatus is stopped if the mean and / or the standard deviation are again outside the monitoring limits. According to a preferred embodiment, for certain characteristics such as the capacity of the capacitors, the average control values are all calculated from a mean standard deviation determined once and for all following a set of measurements made on

the reference article.

  Preferably, the monitoring limits of the average are substantially equal to LSS f M + 2 C

LIS M 2

  o M represents the average obtained during the first step and C the standard deviation, while the control limits of the mean are substantially equal to

LSC M + 3

LIC _ M 3 S-

  o M represents the average obtained during the first step and

o the standard deviation.

  Other features and benefits of this

  invention will appear on reading the description of a mode

  preferred embodiment of the method, with reference to the accompanying drawings in which: Figure 1 already described is a schematic front view of a test and sorting apparatus on which can be implemented the method of the present invention ; Figure 2 already described is a partially exploded sectional view of the cylindrical bowl used in the apparatus of Figure 1; Figures 3 and 4 show control charts obtained during the implementation of the method of the present invention. The present invention will be described with reference to a capacitor testing and sorting apparatus which performs capacitor testing and sorting based on loss angle capacitance and tangent measurements. Those skilled in the art that the measurement control method according to the present invention can be applied to test and sorting apparatus dealing with other types of miniature articles and carrying out measurements on other types of characteristics. . According to the present invention, for the implementation of the method, a reference capacitor whose capacitance and loss angle tangent values are stable over time is used for each range of capacitors to be sorted. This reference capacitor is positioned permanently in one of the cells of the turntable 3 shown in Figure 2 This capacitor is measured at each of its passages under the measuring head 8 by the test and sorting device, namely every 14 seconds in the test and sorting apparatus on which the present invention has been implemented. Before its use in a capacitor test and sorting apparatus, the capacitance and tangent characteristics of the loss angle of the reference capacitor have been measured for several months on a type of apparatus used under the best measurement conditions. These measures are based on a standard deviation which gives the measurement dispersion of the device.

  measurements follows a Gaussian law of mean M and standard deviation <-.

  These values are therefore used to determine the two framing values that will be used during the implementation of the

  process according to the present invention.

  In the context of the present invention, two pairs of averaging values, namely, monitoring limits and control limits, are used. More specifically, the monitoring limits are determined so that at a minimum moment T if the average of the measured sample is within these limits, there is a 95% chance that the test and sorting device will be set correctly. The monitoring limits are such that: LSS (upper limit of monitoring) M + 2 Due

  LIS (lower limit of surveillance) = M 2 Z-

  Similarly, the control limits are such that if the average of the sample measured at a time T is within these limits, there is a 99.8% chance that the test and sorting apparatus will be properly adjusted. are such that LSC (upper limit of control) j M + 3 LIC (lower limit of control) = M 3 Thus, once these values determined for the capacitor-sample corresponding to a range of capacitors, the method of control of Measurements made in a capacitor test and sorting apparatus can be implemented during the testing and sorting of a batch of capacitors. In fact, the beginning of each batch of capacitors to be sorted is measured, for example, the first 20 capacitance and tangent values of the loss angle of the reference capacitor On the basis of these first 20 values, the calculation of the average value is performed using a known type of processing system, such as a computer system. in c apacity and average value in tangent of the loss angle Knowing, depending on the measurement ranges, the value of the standard deviation capacity and the value of the tangent standard deviation, it is possible to establish control charts. and tangent, as shown in FIGS. 3 and 4 These values thus established serve as reference values for comparing the next values of the sets of these

measures.

  Following the establishment of the control charts, we begin to measure, for example, 20 capacitance and tangent values of the capacitor loss angle. Using these values, the average capacitance value is again calculated. and tangent of the loss angle as well as the capacity deviation and the tangent standard deviation. The averages and standard deviations thus determined are compared to their respective framing values. If a mean value or standard deviation is beyond the upper or lower monitoring limits, then the test and sorting device has a 2, 5% chance of being correctly set. second set of 20 measurements If again one of the mean values or one of the standard deviations is beyond the upper or lower limits of monitoring, then we have 6.25 per 10,000 chances that the test and sorting device be correctly

  As a result, the machine stops automatically.

  On the other hand, if one of the average values or one of the standard deviations is outside the upper or lower limits of control, there is a one in 1000 chance that the device is correctly adjusted In this situation,

  the device stops automatically.

  During the sorting, the above measurements are repeated and all the N 'measurements, namely all the measurements in the embodiment described, the average value in capacitance, the average value in tangent of the loss angle, the standard deviation capacity and the tangent tangential deviation of the loss angle and these values are compared to the values

defined frameworks.

  For the values indicated on the previously established control charts, the device stops automatically when values outside the frame values are detected and the device is operated according to the type of values coming out of the values. Thus, if the average value in capacitance or in tangent of the angle of loss is outside the framing limits, the position of the reference capacitor is verified and the operation of the test apparatus is checked and If, on the contrary, the standard deviation in capacitance or tangent of the loss angle is outside the limits, the reference capacitor, its position under the measuring contacts, the quality of the contacts and

also the test and sorting device.

  Once all the checks have been made, new measurements are taken If the average value or the standard deviation has returned within the limits, the apparatus is again well adjusted, otherwise one acts on another parameter

  until the device is properly tuned.

  An example of values that can be obtained using the measurement control method will be given below.

  performed in a capacitor test and sorting apparatus.

  These values were obtained when measuring capacitors belonging to the range 2 n F to 2 On F of type I. In this case, the upper and lower limits of control were determined once and for all, during the measurement during several words and are at lp F relative to CM while the upper and lower limits of monitoring lie at -0, Gp F of CM (the standard deviation capacity corresponding to the dispersion of the measurement of the device is a value constant for each range of capacitors and = 0, 33 p F in the range 2 n F to 2 On F of type I) Then, after the first 20 measurements, the average capacitance CM is determined which in this case equals 3 , 1595 n F Similarly, the upper and lower limits of control and monitoring for the standard deviation capacitance are fixed once and for all for a range of capacitors, the mean of the standard deviation being obtained after the first 20 measures in the ca s represented, Etm = 0.00150, LSC = 0.002, LIC = 0.0007,

LSS = 0.0017, LIS = 0.0009.

  With respect to the control board established on the tangent of the loss angle for the same batch of capacitors, the following values are obtained: Tangent M = 2; LSC = 7.44; LIC = O; LSS = 5.63; LIS = O Standard deviation M = 1.815; LSC = 3.1; LIC = 1,1 In this case, only LSC and LIC for the standard deviation are set for a range of capacitors. The other limits are obtained during the 20

  first measurements using the formulas given above.

  Different values obtained for different samples of 20 measurements will be given below. These values are placed on the two control charts so as to show their dispersion with respect to the average values: CM I sample 2 nd sample 3 rd sample 4 th sample th sample 6 th sample 3,15953 n F

3.15940

3.15965

3.15960

3.15965

3.15935

and C

0.00150

0.00171

0.00208

0.00159

0.00177

0.00146

tm M 2.0 2.0 1.7 1.8 1.7 1.9 ettg

1.8150

1.8296

1.6815

1.4586

1.3820

1.7168

Claims (6)

  A method for controlling the measurements made in a device for testing and sorting miniature articles comprising a transfer device provided with cells for receiving the articles to be tested and sorted, characterized in that: in one of the cells, a reference article having stable characteristics over time; in a first step, N times (N> 1) is tested at least one of the characteristics of said article to determine the mean and the standard deviation of said characteristic; for the average and the standard deviation of said characteristic, at least two framing values are determined; in a second step, it tests N 'times (N'> 1) said characteristic of the reference article to determine, again, the mean and the standard deviation of said characteristic; the average and standard deviation determined are compared with their respective framing values; if the mean and the standard deviation are outside the framing values, either the test and sorting device is stopped, or the second step for checking is repeated; if the mean and the standard deviation are within the framing values, the measurements are continued by repeating the procedure at the level of the second stage.   Method according to Claim 1, characterized in that, for the mean and the standard deviation, two pairs of frame values, called monitoring limits and control limits, are determined, and in that when the average and / or the standard deviation are out of the control limits, the test and sorting device is stopped while when the mean and / or standard deviation are outside the monitoring limits, we start again the process at the second stage and stop the test and sorting device if the average and / or it   the standard deviation are again outside the monitoring limits.   Process according to any one of claims 1   and 2, characterized in that the average framing values are calculated from a mean standard deviation determined once and for all following a set of measurements performed on the reference article.   Process according to any one of the claims   1 to 3, characterized in that the monitoring limits of the mean are substantially equal to LSS M + 2 LIS M 2   M representing the average obtained during the first step, and the average standard deviation or standard deviation determined during the first stage.   Process according to any one of the claims i   at 3, characterized in that the control limits of the mean are substantially equal to LS C M + 3 LIC M 3 6   M representing the average obtained during the first step, and <the average standard deviation or the standard deviation determined during the first step; first stage.   Process according to any one of the claims i   to 5, characterized in that the miniature article is a capacitor and in that the measured characteristics are the   capacity and the tangent of the angle of loss.