SU1112326A1 - Device for checking and selecting articles by reliability - Google Patents

Abstract

1. DEVICE FOR CONTROL AND SELECTION. RELIABILITY PRODUCTS, containing an input unit connected by an input to an input terminal of a device, an output — to a first input of a first classifier connected by a first output to a first input of a first accumulator, which, in order to increase the reliability of control and reliability, is inserted into the device classifier, second drive, block of input and program block connected by inputs with corresponding first inputs of second classifier, second drive and output block and with second inputs of input block first the second classifier and the first accumulator, the second output of the first classifier is connected to the second input of the second classifier, the third input of which is connected to the first output of the first accumulator connected by the second output to the first input of the output unit connected by the output to the output terminal of the device, the second input to the output of the second accumulator, connected by a second input to the output of the second classifier.

Description

2, the device according to claim 1, characterized in that the first classifier contains a first memory block connected by a first input to the first input of the boundary limit generator and to the second input of the classifier, the second input to the first input of the classifier, the third input to the calculator output, the first output — to the first input of the expiration analyzer, connected by the output to the first output of the classifier; the second input — to the output of the boundary limit generator, connected by the input to the second output of the memory unit, connected by the third output to the second output of the classifier, and the fourth output - with the input of the calculator.

3. The device of claim 1, wherein the second classifier contains a normalization unit connected by the first input to the first input of the classifier and to the first input of the unit of measurement error settings, the second input to the second input of the unit of error settings and to the second input classifier, output - to the input of the second memory unit, connected via the comparison unit to the first input of the migration analyzer, the second input of which is connected to the output of the unit of measurement error settings, connected to the output to the classifier output, rety input coupled to a third input of the normalization block.

The invention relates to the field of control and technical diagnostics, a group production process and can be used to control the quality and quality of highly reliable products used in the manufacture of high-quality electronic equipment (CEA), computing devices and automation equipment. A device for controlling the reliability of products is known, which contains a sequentially included parameter input unit, a classification unit, a storage unit and a decision unit. The disadvantage of this device is the limited scope of its application only in production, which is characterized by a constant stable process and unchanged raw material characteristics. The closest technical solution to the invention is a device for controlling and selecting products for reliability, containing an input unit connected by an input to an input terminal of a device, an output — to a first input of a first classifier connected by a first output to a first input of a first accumulator 2. The disadvantages of the known device are the reliability of the control is low due to the impossibility of assessing the individual reliability of products in the lot, as well as the impossibility of operatively controlling the stability of the technological process Mass production due to the obvious periodicity, high labor intensity and the duration of the test cycle of product samples for reliability. In addition, products that have been tested for reliability cannot be used further. The classification of products in a controlled batch into suitable and unsuitable products is carried out by an indirect method, namely, according to the classifying rule, developed on the basis of the results of tests for the reliability of other products of this batch. The aim of the invention is to increase the reliability of the control and its reliability. The goal is achieved by the fact that a device for controlling and selecting products for reliability, containing an input unit connected by an input to an input terminal of a device, an output to a first input of a first classifier connected by a first output to a first input of a first drive, a second classifier, a second drive , output unit is a software unit connected to the input with the corresponding first inputs of the second classifier, the second accumulator and the output unit and with the second inputs of the input unit of the first classifier and the first The second output of the first classifier is connected to the second input of the second classifier, the third input of which is connected to the first output of the first drive connected by the second output to the first input of the output unit connected to the output terminal of the device, the second input to the output of the second drive connected to the second input with the release of the second classifier. The first classifier contains the first memory block, the first primary input with the first input of the boundary limit generator and the second input of the classifier, the second input with the first input of the classifier. The third input - with the output of the calculator, the first output - with the first input of the analyzer of fitness for life, connected by the output to the first output of the classifier, the second input to the output of the boundary limit generator, connected by the input to the second output memory block, connected by the third output to the second output of the classifier, and the fourth exit - with the input of the evaluator. The second classifier contains the normalization unit connected to the first input of the classifier and the first input of the unit of measurement error settings, the second input to the second input of the unit of measurement error settings and the second input of the classifier connected to the second memory block connected through the unit comparison with the first input of the migration analyzer, the second input of which is connected to, the output of the block of measurement error settings, connected to the output of the classifier, the third input of which is connected to t etim input unit normalization. FIG. 1 shows a block diagram of a device for controlling and selecting products according to reliability levels; in fig. 2 is a block diagram of a first classifier; FIG. 3 is a block diagram of a second classifier; FIG. 4 and histograms), which illustrate the distribution of parameters during the first and second test effects, respectively. The device contains a data input unit 1, first classifier 2, first drive 3, second classifier 4, second drive 5, output block 6, program block 7, memory block 8 of the first classifier, calculator, shaper 10 limit limits, 11-year analyzer, block 12 rationing, block 13 of the second classifier memory, block 14 of measurement error settings, block 15 comparison., Analyzer 16 migrations. The normalization unit 12 is connected to the first input of the classifier 4 and the first input of the block 14 of measurement error settings, the second input to the second input of the block 14 of measurement error settings and the second input of the classifier 4, the third input to the first input of the classifier 4, the output to the input the second memory block connected via the comparison block 15 to the first input of the analyzer 16 migrations, the second input of which is connected to the output of the block 14 of measurement error settings. The input unit 1 is connected by an input to the input terminal of the device, an output to the first input of the first classifier 2 connected by the first output to the first input of the first drive 3. Program block 7 connected by inputs to the corresponding first inputs of the second classifier 4, second drive 5 and the bis output block second the inputs of input unit 1, the first classifier 2, and the first accumulator 3, the second output of the first classifier 2 is connected to the first input of the second classifier 4, the second input of which is connected to the first output of the first an ite 3 connected by a second output to a first input of output unit 6 connected by an output to an output terminal of a device; a second input to an output of a second storage unit 5 connected by a second input to an output of a second classifier 4. A first memory block 8 connected by a first input to a first input of a forirovator 10 boundary limits and the second input of the classifier 2, J1 by the second input - with the first input of the classifier 2, the third input with the output of the calculator 9, the fourth output - with the input of the calculator 9, the first output - with the first input of the analyzer 11 and, connected to the first output of the classifier 2, the second input to the output of the imaging unit 10 boundary limits connected to the second output of the memory block 8 connected to the third input to the second output of the classifier 2. The device works as follows. For the control of reliability, a statically presented controllable batch of products with a volume of N pcs. and measuring the uncorrelated diagnostic parameters (x, y, etc.) of each product with at least two test effects (measurement conditions), for example, under normal conditions (x, y) and with different input effects, at reduced (x, y) or increased power, temperature, or other. The measurement data of the parameters of the products in the sequence determined by a predetermined sequence of arrival of the products to the measurements is entered; using the data input unit 1 (Fig. 1) at the command of the program unit 7 to the classifier 2 and recorded in the prescribed order in the storage unit of the first class of fixator 8 (Fig. 2). The sequence number of the parameter measurement result at each test action corresponds to the product serial number. After completing the data recording operation by the calculator 9 of classifier 2 for the measurement results of each parameter obtained at each established test action, expectation value and standard deviation. , according to the measurements of the parameter X at the first test exposure (Fig, 3), M (x) and 6 (x.) are determined, according to the measurement data at the second test exposure, M (x) ibh are determined. According to the measurement data of the parameter y, the values of M (y), d y and M (y), d y, etc. are determined, respectively. The calculated results are recorded in the memory block of the classifier 8, and then, at the command of unit 7, are transmitted to the driver 10 boundary limits. . Simultaneously from block 7, the set value of the coefficient K is introduced into the imaging unit 10, which provides the required limitation of the limits of deviations of diagnostic parameters, which determines a certain level of reliability of products. The shaper of 10 boundary limits, based on the entered value K, sets specific values of the boundary limits for all diagnostic parameters measured at the test effects received, based on the obtained values of standard deviations relative to the corresponding values of mathematical expectation. Thus, for a parameter, the boundary limits are set for the values obtained during the first test action M (x) kdx, and for the second test action M (x) -k (jx. For the parameter y, the boundary limits M (y) k (jy and M ( i) ikdy. In the boundary limit setting order, these data are fed into the 11-life analyzer, which in the same order requests data from the memory of the classifier 3 of the measurement data of the corresponding diagnostic parameters obtained under the respective test effects, and selects the values, and , and products (the sequence number of the parameter measurement result corresponds to the serial number of the product) on all parameters falling within the established limits on all parameters at the same time and not falling into them. Selection of the results of the values of parameters within the established limits, which to a certain extent determines the constancy the behavior of the parameters in the accepted measurement conditions is the sampling of products according to the first criterion. So, for the parameters x and y of the product, measured under two test effects, the obtained values of which satisfy the first selection criterion, the following conditions are met (Fig. 4 and 5)

/

| lAUVk6xl x- (fAU1-kdxl,

lft UVk6x4 x, lMCxVk6xl3 and

| MCI yu1 Ch; 1 1HNB1 ||,, (ij) -k (5yb

For products that do not meet the requirements of the first selection criterion, the life analyzer 11 generates a code signal for a sign of non-conformity together with a code signal for the product number.

The data on the conformity and non-conformity of each product to the lot in the order of their arrival is transferred by the block 11 to the accumulator 3, which stores it. After this, block 7 gives the command to start the classifier 4. On this command, block 12 of the second classifier's normalization (Fig. 3) requests data from the first accumulator 3, as well as values of the mean and standard deviation obtained for each parameter with the appropriate test effects from the storage unit of classifier 8, and performs data processing in terms of reducing them to relative values, for example, for the parameter x

x -MU1

X, -MS

 01H.

, --- dx 01I,

and so on for all available values for each parameter, measured with installed test effects of 1 x.

The obtained results are recorded in the memory block of the classifier 13, the end of the operation is recorded by the program block 7, which gives the command to transfer data from the FIRST drive 3 to the block 14 of the error setpoints of the classifier 4. At the same time, the data from the memory block of the second classifier 13 is supplied to the block 15 compare .

The unit 14 of measurement error settings for the measurement result of each parameter arriving from the first storage unit 3 with the appropriate test action determines the measurement range and determines the relative limit measurement error i inherent in this range.

12326-8

At the same time, the sequence number of the measured measurement error value corresponds to the sequence number of the measurement result under consideration, and, consequently, the product number. The measurement ranges of measuring instruments and the inherent limiting measurement errors lie in the block 1 of 14, which considers the measurement results that only meet the requirements of the first selection criterion. The value of the measurement error that it receives enters the analyzer 16 migrations

15 classifier 4.

At the same time, block 15 compares this classifier by the input relative values of each parameter measured at

20 test effects with the same sequence numbers for the same parameter determine the difference between them. So, for the x-th item parameter, the value of the difference is 5 s. -Th x-

  U M h O N

for the parameter y -,.

The value of the difference is determined for products suitable for the first criterion. The result

0 from the comparison unit 15 is fed to the analyzer 16 migrations of the second classifier 4.

The migration analyzer 16 for each product determines

5 stability of carrying out the same parameter for different accepted test effects within the established boundary limits by comparing the corresponding value of the input difference with the magnitude of the measurement error of the considered data of this parameter.

.

With a high degree of stability of the parameter behavior in the measurement conditions under consideration, the difference value should not exceed the measurement error. This requirement is the second criterion for selecting a product, namely: for the results of measuring the parameter k. Of the item selected by the first criterion, meeting the requirement of the second selection criterion, the condition must be fulfilled for the results

5 measurements of the parameter y, respectively; ± bn 4mU; This comparison is performed by block 16 for all available values

91

controlled products that meet the requirements of the first selection criteria. If the result and, consequently, the product does not meet the requirement of the second selection criterion, the analyzer 16 migration generates a code signal of a sign of product non-conformity.

The data on the conformity or non-compliance of the second criterion of each batch product in the previously established order with block 16 transmits to the second accumulator 5, which stores it.

After the completion of these operations, the control program of block 7 issues a command to start the output block 6.

Unit 6 output requests information from the first drive 3 and vtov

1232610

storage device 3 and provides data on product numbers that do not meet the selection criteria, which correspond only to the first selection criteria and correspond to the first and second

selection criteria simultaneously. As a result, the controlled batch of products is selected into groups characterized by the corresponding

10 levels of reliability.

Thus, the proposed device ensures that the parameters that migrate when the ambient conditions change, with a large value. migration is greater than the error of measuring devices, which significantly increases the reliability of monitoring the reliability of products.

four

F

J

h

//

.2

12

14

L7

five

sixteen

15

N I

N

, / X1M (X) H f1 (X) M (X)

N

ABOUT

M1U} -Kb (U)

FIG. four

N

0

/ fl (y) ff {y) -f (6y M1U) K6y

FIG. five

Claims (3)

1. DEVICE FOR CONTROL AND SELECTION. RELIABILITY PRODUCTS, containing an input unit connected by an input to the input terminal of the device, an output - with the first input of the first classifier, connected by the first output to the first input of the first drive, which differs * in order to increase the reliability of control and its reliability, in the device introduced a second classifier, a second drive, an output unit and a program unit connected by inputs to the corresponding first inputs of the second classifier, a second drive and an output unit and with the second inputs of the input unit of the first class the identifier and the first drive, the second output of the first qualifier is connected to the second input of the second qualifier, the third input of which is connected to the first output of the first drive, connected by the second output to the first input of the output unit, connected by the output to the output terminal of the device, the second input to the output of the second drive, connected the second input with the output of the second classifier. Figure 1 2. The device according to π. 1, the first classifier comprising a first memory block connected by a first input to the first input of the boundary limiter and the second input of the classifier, the second input with the first input of the classifier, the third input with the output of the calculator, the first output with the first input of the shelf life analyzer, connected by the output to the first output of the classifier, the second input - with the output of the boundary limiter, connected by the input to the second output of the memory block, connected by the third output to the second output of the class sifificator, and the fourth output - with the input of the computer. 3. The device according to π. 1, the second classifier comprising a standardization unit connected by the first input to the first input of the classifier and to the first input of the unit for setting the measurement errors, the second input to the second input of the unit for setting the errors and the second input of the classifier, the output to the input the second memory block connected through the comparison unit to the first input of the migration analyzer, the second input of which is connected to the output of the block of settings of the measurement errors, connected by the output to the output of the classifier, the third input of which oedinen the third input unit normalization.