JP2009290885A - A/d converter and calibration device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To calibrate a time error generated among a plurality of analog-to-digital (A/D) converters. <P>SOLUTION: A calibration device includes: an error calibration value calculating section 70a that calculates a time error calibration value used for an arithmetic operation to calibrate a time error, on the basis of a sample value resulting from sampling a test signal used by first and second A/D converters to calculate the time error; a read section 76 that reads the sample value from a storage section storing the sample value obtained by sampling a measured signal which is an analog signal to be measured; and an error calibration section 70b that calibrates the time error generated in the second A/D converter when the measured signal is sampled, on the basis of the sample value read by the read section 76 and the time error calibration value. Through this configuration, time error calibration can be made by the arithmetic operation. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a semiconductor device test apparatus, and more particularly to an A / D conversion apparatus and a calibration apparatus incorporated in a semiconductor device test apparatus.

  FIG. 1 shows a block diagram of a conventional A / D converter 101 that converts an analog signal into a digital signal. The A / D converter 101 includes an analog signal input unit 11, A / D converters 13a and 13b, a sampling clock signal generator 15, a reference clock signal generator 17, a delay circuit 24, and an interleave processing unit 19. The interleave processing unit 19 includes a multiplexer 29 and a storage unit 21.

  An analog signal 50 is input to the analog signal input unit 11. The input analog signal 50 is sampled and converted into a digital signal by the A / D converters 13a and 13b that alternately perform sampling operations. The digital signals output alternately from the A / D converters 13 a and 13 b are sequentially rearranged by the multiplexer 29 and stored in the storage unit 21.

  The sampling clock signal generator 15 generates sampling clock signals 56 a and 56 b for alternately sampling the A / D converters 13 a and 13 b based on the reference clock signal 54 generated from the reference clock signal generator 17. The delay circuit 24 calibrates the timing of the sampling operation of the A / D converters 13a and 13b, and is disposed on the transmission path of the sampling clock signals 56a and 56b generated from the sampling clock signal generator 15.

  FIG. 2 shows a conventional A / D conversion device 102 having a plurality of analog signal input units. The A / D converter 101 includes an A / D converter (13a to 13d), a reference clock signal generator 17, and a storage unit (21a to 21d) corresponding to a plurality of analog signal input units (11a to 11d). Prepare.

  Analog signals (50a to 50d) are input to the analog signal input units (11a to 11d), respectively. The input analog signal is converted into a digital signal by each of the A / D converters (13a to 13d). The converted digital signal is stored in each of the storage units (21a to 21d).

  FIG. 3A is a block diagram showing the interleaving process. The interleaving process is a process of sequentially rearranging sample values obtained by alternately sampling the two A / D converters 13 a and 13 b using the interleave processing unit 19. By the interleaving process, it is possible to obtain the same sample value as when sampling at a higher sampling rate than the sampling rate of one A / D converter. In the interleaving process, sampling clock signals 56a and 56b (see FIG. 3B) that are out of phase with each other are supplied to the A / D converter, thereby alternately sampling the A / D converter.

  Interleaving is a method of sequentially rearranging digital signals output from a plurality of A / D converters as described above. In the case of interleaving processing, the sampling timings at which each A / D converter actually performs a sampling operation based on the sampling clock signal must be equidistant. However, in reality, a time error occurs with respect to a desired sampling timing due to the difference in the characteristics of each A / D converter and the characteristics of the transmission path of the sampling clock signal. Therefore, time error calibration is required. Conventionally, as described with reference to FIG. 1, the time error is calibrated by providing a variable delay element in the path for guiding the sampling clock signals 56a and 56b to each A / D converter.

  The conventional A / D conversion device 101 shown in FIG. 1 performs only the interleaving process in which the A / D converters 13a and 13b are alternately sampled, and cannot perform other processes.

  In the conventional A / D converter 102 shown in FIG. 2, the A / D converters used for processing the analog signals input from the respective analog signal input units are fixed in advance. In addition, the conventional calibration of the time error was performed by the delay circuit. However, this complicates the calibration of the time error, and the range in which the time error can be calibrated depends on the performance of the delay circuit. Highly accurate calibration (calibration) was not possible.

  Therefore, an object of the present invention is to provide an A / D conversion apparatus, a calibration apparatus, and a semiconductor device test apparatus that can contribute to solving at least one of the above problems. This object is achieved by a combination of features described in the independent claims. The dependent claims define further advantageous specific examples of the present invention.

  In order to solve the above problems, a first embodiment of the present invention is an A / D conversion device that samples an analog signal and converts it into a digital signal, an analog signal input unit for inputting the analog signal, and an analog signal A plurality of A / D converters that sample the analog signals input by the input unit and convert them to digital signals, a sampling clock signal for averaging processing that causes the plurality of A / D converters to perform a sampling operation, Sampling clock signal generator for supplying one of interleave processing sampling clock signals for alternately sampling A / D converters to a plurality of A / D converters, and sampling based on sampling clock signals for averaging processing Averaging processing for averaging digital signals output from the operated A / D converter If, to provide an A / D converter, characterized in that it comprises the interleaving processing unit for interleaving a digital signal output from the sampling operation by the A / D converter on the basis of the interleave process sampling clock signal.

  Another aspect of the first aspect further includes a mode designation signal generator that generates a mode designation signal that designates either the averaging process or the interleave process, and performs averaging based on the mode designation signal. Either the processing unit or the interleave processing unit may be selected. Further, a reference clock signal generator for generating a reference clock signal is further provided, and when the averaging process is designated by the mode designation signal, the sampling clock signal generator is an averaging process sampling clock synchronized with the reference clock signal. When the signal is supplied to each of the plurality of A / D converters and the interleave processing is designated by the mode designation signal, the sampling clock signal generator performs sampling for interleave processing having different phases based on the reference clock signal A clock signal may be supplied to each of the plurality of A / D converters.

  The averaging processing unit converts the digital signals output from the plurality of A / D converters into voltage values based on voltage conversion coefficients for converting the digital signals into voltage values, and converts each converted voltage value. May be added together. Further, the interleave processing unit may convert digital signals output from the plurality of A / D converters into voltage values based on voltage conversion coefficients, and rearrange the converted voltage values alternately. In addition, a plurality of storage units for storing digital signals output from each of the plurality of A / D converters are further provided, and the averaging processing unit and the interleave processing unit perform processing based on the digital signals stored in the storage unit. May be.

  The plurality of A / D converters include a first A / D converter and a second A / D converter, and the second A / D converter should sample the timing sampled by the first A / D converter. An error calculating unit that calculates a time error that is a time lag between a predetermined timing and a timing at which the second A / D converter is actually sampled, and a second A / D converter based on the calculated time error. From an error calibration value calculation unit for calculating a time error calibration value used for an operation for calibrating a time error, and a storage unit storing a digital signal obtained by sampling a signal under measurement which is an analog signal to be measured When reading the signal under measurement based on the digital signal read from the storage unit and the time error calibration value by the readout unit that reads the digital signal It may further comprise a error correction unit for performing an operation of calibrating the time error occurring in the 2A / D converter.

  A second aspect of the present invention is an A / D conversion device that samples an analog signal and converts it into a digital signal. The analog signal input unit inputs an analog signal, and the analog signal is sampled into a digital signal. An adder that adds the digital signal output from the first A / D converter to be converted and the digital signal output from the second A / D converter that samples and converts the analog signal to a digital signal; A multiplexer that alternately inputs a digital signal output from the D converter and a digital signal output from the second A / D converter and sequentially outputs them, an adder output value, or an output value of the multiplexer Provided is an A / D conversion device comprising a selector for selecting one of them.

  A third aspect of the present invention is an A / D conversion device that samples an analog signal and converts it into a digital signal, and a plurality of analog signal input units to which different analog signals are respectively input, and the analog signal is sampled Then, a plurality of A / D converters for converting into digital signals, and an analog signal is converted into any one or a plurality of A / D converters according to the contents of the digital processing of the analog signal. There is provided an A / D conversion device comprising an analog signal distributor for distributing to a converter. For example, the content of processing may be the number of A / D converters used for interleaving processing or the number of A / D converters used for averaging processing.

  Another aspect of the third mode is that an analog signal distributor is provided corresponding to each of a plurality of analog signal input units, and an analog signal distributor is input from one analog signal input unit. The signal may be distributed to a plurality of A / D converters. In addition, a plurality of sampling clock signals for averaging processing for performing sampling operations in synchronization with a plurality of A / D converters and a plurality of sampling clock signals for interleaving processing for alternately performing sampling operations for a plurality of A / D converters are used. A sampling clock signal generator to be supplied to the A / D converter, and an averaging processor for averaging the digital signal output from the A / D converter that has been sampled based on the sampling clock signal for averaging And an interleaving unit that interleaves a digital signal output from the A / D converter that has performed the sampling operation based on the interleaving sampling clock signal.

  Further, it further comprises a mode designation signal generator for generating a mode designation signal for designating either one of the averaging process or the interleave process, and based on the mode designation signal, either the averaging processor or the interleave processor Either one of them is selected, and based on the selected processing, a distribution control signal that specifies that the analog signal is distributed to any one or a plurality of A / D converters of the plurality of A / D converters. A distribution control signal generator for supplying to the distributor may be further provided. Further, the apparatus further includes a reference clock signal generator for generating a reference clock signal, and when the averaging process is designated by the mode designation signal, the sampling clock signal generator is an averaging process sampling clock synchronized with the reference clock signal. When the signal is supplied to each of the plurality of A / D converters and the interleave processing is designated by the mode designation signal, the sampling clock signal generator performs sampling for interleave processing having different phases based on the reference clock signal A clock signal may be supplied to each of the plurality of A / D converters.

  The averaging processing unit converts the digital signals output from the plurality of A / D converters into voltage values based on voltage conversion coefficients for converting the digital signals into voltage values, and converts each converted voltage value. You may perform the calculation which adds. In addition, the interleave processing unit may convert digital signals output from the plurality of A / D converters into voltage values based on voltage conversion coefficients, and alternately rearrange the converted voltage values. In addition, a plurality of storage units for storing digital signals output from each of the plurality of A / D converters are further provided, and the averaging processing unit and the interleave processing unit perform processing based on the digital signals stored in the storage unit. May be.

  The plurality of A / D converters include a first A / D converter and a second A / D converter, and the second A / D converter should sample the timing sampled by the first A / D converter. An error calculating unit that calculates a time error that is a time lag between a predetermined timing and a timing at which the second A / D converter is actually sampled, and a second A / D converter based on the calculated time error. From an error calibration value calculation unit that calculates a time error calibration value used for an operation to calibrate a time error, and a storage unit that stores sample values obtained by sampling a signal under measurement that is an analog signal to be measured When the sampled signal is sampled based on the sample value read from the storage unit and the time error calibration value by the reading unit that reads the sample value, and the second A / It may further comprise a error correction unit for performing an operation of calibrating the time error occurring in the transducer.

  The fourth aspect of the present invention occurs between a first A / D converter that samples an analog signal and converts it to a digital signal, and a second A / D converter that samples the analog signal and converts it to a digital signal. A calibration device for calibrating an error, wherein the second A / D converter actually samples with respect to the timing at which the second A / D converter should sample with respect to the timing at which the first A / D converter samples. Based on a sample value obtained by sampling a test signal used to calculate a time error that is a time lag with respect to the calculated timing, an error calculation unit that calculates a time error, and based on the calculated time error, An error calibration value calculation unit for calculating a time error calibration value used in an operation for calibrating the time error of the second A / D converter, and an analog signal to be measured. Based on the sample value and the time error calibration value read from the storage unit by the reading unit that reads the sample value from the storage unit in which the sample value obtained by sampling the signal under measurement is stored Thus, a calibration apparatus is provided that includes an error calibration unit that performs an operation of calibrating a time error generated in the second A / D converter when the signal under measurement is sampled.

  In another aspect of the fourth mode, the error calibration unit performs discrete Fourier transform on the sample value of the signal under measurement read from the storage unit by the readout unit, and a discrete Fourier transform value obtained by the discrete Fourier transform and a time error The time error may be calibrated by calculating based on the calibration value. The error calculation unit calculates the gain and offset of the first A / D converter and the second A / D converter, and the error calibration value calculation unit calculates the first A / D converter and the second A / D based on the gain. A gain calibration value calculation unit for calculating the gain calibration value of the converter, and an offset calibration value calculation unit for calculating the offset calibration value of the first A / D converter and the second A / D converter based on the offset, Based on the sample value of the signal under measurement read from the storage unit by the reading unit, the gain calibration value, and the offset calibration value, the error calibration unit is configured to obtain the gains of the first A / D converter and the second A / D converter and A gain / offset calibration unit for calibrating the offset may be provided. The gain / offset calibration unit may perform an operation of adding the offset calibration value to a value obtained by multiplying the sample value of the signal under measurement sampled by the second A / D converter by the gain calibration value.

  The fifth aspect of the present invention occurs between a first A / D converter that samples an analog signal and converts it to a digital signal, and a second A / D converter that samples the analog signal and converts it to a digital signal. An error calibration method for calibrating an error, the second A / D converter actually sampling with respect to a timing sampled by the first A / D converter and a predetermined timing to be sampled by the second A / D converter A step of calculating a time error, which is a time lag with respect to timing, a step of calculating a time error calibration value used for calibrating the time error based on the time error, and sampling a signal under measurement to be measured. Calibrating a time error generated when the signal under measurement is sampled based on the obtained sample value and the time error calibration value; and Providing an error calibration method, characterized in that it comprises.

  According to another aspect of the fifth mode, the error calibration method includes a step of calculating gains and offsets of the first A / D converter and the second A / D converter, and the gain and offset based on the calculated gain and offset. Based on the step of calculating the gain calibration value and the offset calibration value used for the calculation for calibrating the offset, the sample value obtained by sampling the signal under measurement, the gain calibration value and the offset calibration value, the first A / D And calibrating the gain and offset of the converter and the second A / D converter.

  According to a sixth aspect of the present invention, a predetermined timing that the second A / D converter should sample with respect to the timing sampled by the first A / D converter using the arithmetic unit, and the second A / D converter A recording medium that records a program for calibrating a time error, which is a time lag with respect to a timing at which actual sampling is performed, and the program is based on a module that causes a computing device to calculate a time error, and the calculated time error, Based on the module for calculating the time error calibration value used for the operation for calibrating the time error of the second A / D converter, the sample value obtained by sampling the signal under measurement to be measured, and the time error calibration value, A program for calibrating the time error, comprising a module for calibrating the time error generated when the signal under measurement is sampled. Providing readable recording medium and arithmetic unit.

  According to another aspect of the sixth aspect, the program includes a module that causes the arithmetic device to calculate the gain and the offset, and a gain calibration value and an offset calibration that are used for an operation that calibrates the gain and the offset based on the calculated gain and offset. Based on the module for calculating the value, the sample value obtained by sampling the signal under measurement, the gain calibration value, and the offset calibration value, the gain and offset of the first A / D converter and the second A / D converter are calculated. And a module for calibrating.

  According to a seventh aspect of the present invention, there is provided a semiconductor device test apparatus for testing a semiconductor device that outputs an analog signal. The pattern generator generates a semiconductor device input signal for testing the semiconductor device, and the pattern generator A performance board that gives the semiconductor device input signal to the semiconductor device, an analog signal input unit that inputs the analog signal output from the semiconductor device, and a digital signal that samples the analog signal input from the analog signal input unit A plurality of A / D converters to be converted, an averaging processing sampling clock signal for performing a sampling operation in synchronization with the plurality of A / D converters, and an interleaving process for alternately sampling a plurality of A / D converters Sampling one of the sampling clock signals A sampling clock signal generator for supplying a clock signal to a plurality of A / D converters, and an average for averaging the digital signals output from the A / D converters sampled based on the sampling clock signal for averaging processing There is provided a semiconductor device test apparatus comprising: a processing unit; and an interleaving unit that interleaves a digital signal output from an A / D converter that performs a sampling operation based on an interleaving sampling clock signal. .

  Another aspect of the seventh aspect of the present invention includes a plurality of analog signal input units and an analog signal that distributes the analog signal to any one or a plurality of A / D converters of the plurality of A / D converters. And a distributor. The plurality of A / D converters include a first A / D converter and a second A / D converter, and the second A / D converter should sample the timing sampled by the first A / D converter. An error calculating unit that calculates a time error that is a time lag between a predetermined timing and a timing at which the second A / D converter is actually sampled, and a second A / D converter based on the calculated time error. From an error calibration value calculation unit that calculates a time error calibration value used for an operation to calibrate a time error, and a storage unit that stores sample values obtained by sampling a signal under measurement that is an analog signal to be measured A reading unit that reads a sample value, and a second A / A when the signal under measurement is sampled by the reading unit based on the sample value read from the storage unit and the time error calibration value. It may further comprise a error correction unit for performing an operation of calibrating the time error occurring in the transducer.

  The above summary of the invention does not enumerate all the necessary features of the present invention, and sub-combinations of these feature groups can also be the invention.

  As is apparent from the above description, according to the present invention, an A / D conversion apparatus that efficiently uses a plurality of A / D converters can be provided.

1 shows a conventional A / D converter. 1 shows an A / D conversion device including a plurality of conventional analog signal input units. The block diagram which shows an interleave process. 1 shows an A / D conversion apparatus according to a first embodiment. 2 shows a sampling clock signal generator. 2 shows a timing chart of a sampling clock signal generated by a sampling clock signal generator provided in an A / D converter. 4 shows another embodiment of an A / D converter. An A / D conversion device provided with a plurality of analog signal input parts which is a 2nd embodiment is shown. One form of the analog signal distributor with which an A / D converter provided with a plurality of analog signal input parts is provided is shown. The distribution form of an analog signal distributor is shown. The timing chart of the sampling clock signal which the sampling clock signal generator with which the A / D converter which is 2nd Embodiment is provided is shown. The time error of the sampling timing which arises between two A / D converters is shown. An A / D conversion apparatus provided with the calibration apparatus which is 3rd Embodiment is shown. The block diagram of a calibration apparatus is shown. 1 shows a semiconductor device test apparatus for testing a semiconductor device that outputs an analog signal, including an A / D conversion apparatus according to a first embodiment and a calibration apparatus according to a third embodiment. The semiconductor device test apparatus which tests the semiconductor device which outputs the analog signal provided with the A / D conversion apparatus which is 2nd Embodiment, and the calibration apparatus which is 3rd Embodiment is shown.

  Hereinafter, the present invention will be described through embodiments of the invention. However, the following embodiments do not limit the claimed invention, and all combinations of features described in the embodiments are the solution of the invention. It is not always essential to the means.

  FIG. 4 is a block diagram of the A / D conversion apparatus 100 according to the first embodiment. The A / D converter 100 includes an analog signal input unit 10, A / D converters 12a and 12b, a sampling clock signal generator 14, a reference clock signal generator 16, a processing unit 18, storage units 20a and 20b, and a mode. A designation signal generator 22 is provided. The processing unit 18 includes an interleave processing unit 18a and an averaging processing unit 18b.

  In the present embodiment, the interleave processing unit 18a included in the processing unit 18 performs an interleaving process that alternately rearranges sample values obtained by alternately sampling two A / D converters (12a and 12b). . By the interleaving process, the A / D conversion apparatus 100 can obtain the same sample value as when sampling at a sampling rate that is twice the sampling rate of one A / D converter 12a (or 12b).

  The averaging processing unit 18b performs a process of averaging the sample values obtained by simultaneously sampling the two A / D converters (12a and 12b). By this averaging process, a resolution that is twice the quantization resolution of one A / D converter (or 12b) can be obtained. For example, when two A / D converters having a quantization resolution of 10 bits (range: −512 to +511) are used and simultaneously sampled, the digital signals output from the respective A / D converters are added. -1024 to +1023 data, and a resolution of almost 11 bits can be obtained. When an analog signal having an amplitude of −1.024 V to +1.023 V is sampled, the quantization resolution is 1 mV with one 10-bit A / D converter, but with two 10-bit A / D converters, The quantization resolution is 0.5 mV.

  The mode designation signal generator 22 generates a mode designation signal 58 that designates each component in the A / D converter 100 to operate in either the averaging process mode or the interleave process mode. The mode designation signal 58 is supplied to the processing unit 18 and the sampling clock signal generator 14. In the processing unit 18, either the interleaving processing unit 18 a or the averaging processing unit 18 b is selected based on the mode designation signal 58.

  Further, the sampling clock signal generator 14 is based on the reference clock signal 54 supplied from the reference clock signal generator 16 and the mode specifying signal 58 supplied from the mode specifying signal generator 22. A sampling clock signal for operating each of 12a and 12b) is generated. When the mode designating signal 58 is a signal designating an averaging processing mode, the sampling clock signal generator 14 generates an averaging processing sampling clock signal for simultaneously sampling the A / D converters (12a and 12b). Supply to each of the A / D converters (12a and 12b).

  When the mode designating signal 58 is a signal designating the interleave processing mode, the sampling clock signal generator 14 outputs an interleaving sampling clock signal for alternately sampling the A / D converters (12a and 12b) to the A / D converter. Supply to each of the D converters (12a and 12b).

  The analog signal 50 is sampled by the A / D converters (12a and 12b) that perform the sampling operation according to the sampling clock signal generated from the sampling clock signal generator 14, and the sample values thereof are A / D converters (12a and 12b). Is digitized. When the sampling clock signal generator 14 generates a sampling clock signal for averaging processing, the analog signal 50 is sampled in synchronization with the reference clock signal 54. When the sampling clock signal generator 14 generates a sampling clock signal for interleaving processing, the analog signal 50 is sampled alternately. The sample values digitized by the A / D converters (12a and 12b) are stored in the storage units (20a and 20b).

  The storage unit (20a and 20b) is preferably provided for each A / D converter (12a and 12b). The sample value of the analog signal 50 stored in the storage unit (20a and 20b) uses either one of the interleave processing unit 18a and the averaging processing unit 18b selected based on the mode designation signal 58. Processed.

  The interleave processing unit 18a performs an interleave process for alternately rearranging the sample values. For example, the interleave processing unit 18a converts the digitized sample value stored in the storage units (20a and 20b) into a voltage value based on a voltage conversion coefficient that converts a digital signal into a voltage value. Each voltage value is rearranged alternately. Moreover, the averaging process part 18b performs the averaging process which averages a sample value. For example, the averaging processing unit 18b converts the digitized sample value stored in the storage unit (20a and 20b) into a voltage value based on the voltage conversion coefficient, and adds each converted voltage value. Match.

  FIG. 5 is a block diagram illustrating one embodiment of the sampling clock signal generator 14. In this embodiment, the mode designation signal 58 is a binary signal expressed by two voltage values of Hi (logical value “1”) and Lo (logical value “0”). Hi designates the interleave processing mode, and Lo designates the averaging processing mode. The sampling clock signal generator 14 generates an averaging processing sampling clock signal and an interleaving sampling clock signal based on the reference clock signal 54.

  When the mode designation signal 58 is Hi, the sampling clock signal generator 14 alternately switches the A / D converters (12a and 12b) based on the phase of the reference clock signal 54 generated from the reference clock signal generator 16. A sampling clock signal for interleaving processing to be sampled is output. When the mode designation signal 58 is Lo, the sampling clock signal generator 14 is controlled by the A / D converters (12a and 12b) based on the phase of the reference clock signal 54 generated from the reference clock signal generator 16. A sampling clock signal for averaging processing synchronized with the reference clock signal 54 is output so as to perform the sampling operation at the same time.

  FIG. 6A shows a timing chart of the averaging sampling clock signal output from the sampling clock signal generator 14 shown in FIG. When the mode designation signal 58 is Lo which designates the averaging processing mode, the sampling clock signals (56a and 56b) are output in synchronization with the reference clock signal 54. The A / D converters (12a and 12b) perform a sampling operation based on the sampling clock signals (56a and 56b). The sampling timing of this sampling operation is synchronized with the reference clock signal 54.

  FIG. 6B is a timing chart of the sampling clock signal for interleaving processing output from the sampling clock signal generator 14 shown in FIG. When the mode designating signal 58 is Hi designating the interleave processing mode, the sampling clock signal 56a is output by dividing the reference clock signal 54 by 1/2, and the sampling clock signal 56b is divided by 1/2. The signal phase is further shifted by a half phase. The A / D converters (12a and 12b) perform a sampling operation based on the sampling clock signals (56a and 56b). The sampling timing of this sampling operation is synchronized with the reference clock signal 54.

  In general, an A / D converter requires a certain time (hereinafter referred to as sampling time) in order to convert an analog signal into a digital signal. Therefore, the maximum sampling rate at which one A / D converter can sample in a fixed time is determined for each A / D converter. Therefore, interleaving processing in which a plurality of A / D converters are operated alternately becomes effective. By the interleaving process, an analog signal can be converted into a digital signal at a sampling rate higher than the sampling rate of each A / D converter without increasing the sampling rate of each A / D converter. For example, in the case of the averaging process, the reference clock signal generator 16 cannot make the period of the reference clock signal 54 shorter than the sampling time. However, in the case of interleaving, a plurality of A / D converters are operated alternately, so that the reference clock signal generator 16 can make the period of the reference clock signal 54 shorter than the sampling time of each A / D converter.

  FIG. 7 shows another embodiment. The A / D converter 110 includes an analog signal input unit 10, A / D converters (12a and 12b), a sampling clock signal generator 14, a reference clock signal generator 16, a storage unit 20, and a mode designation signal generator 22. , An adder 26, a multiplexer 28, and a selector 30. The multiplexer 28 is provided as the interleave processing unit 18a described with reference to FIG. The adder 26 is provided as the averaging processing unit 18b described with reference to FIG. In FIG. 7, the configuration given the same reference numeral as in FIG. 4 has the same or similar function as the corresponding configuration in FIG. 4.

  An analog signal 50 is input from the analog signal input unit 10. The input analog signal 50 is sampled by the A / D converters (12a and 12b) that perform the sampling operation based on the sampling clock signals (56a and 56b) supplied from the sampling clock signal generator. The sampling clock signal generator 14 generates either an averaging processing sampling clock signal or an interleave processing sampling clock signal based on the mode specifying signal 58 supplied from the mode specifying signal generator 22. The digital signal output from each A / D converter is output to each of the adder 26 and the multiplexer 28.

  The adder 26 and the multiplexer 28 operate based on the reference clock signal 54 generated from the reference clock signal generator 16. The adder 26 adds each of the digital signals output from the A / D converters (12a and 12b) at the timing of the reference clock signal 54. The multiplexer 28 alternately selects the A / D converters (12a and 12b) at the timing of the reference clock signal 54. Therefore, the multiplexer 28 receives the digital signal generated by the A / D converters (12a and 12b). The data can be sequentially output to the selector 30.

  The digital signal output from the adder 26 and the multiplexer 28 is selected by the selector 30 that operates based on the mode specifying signal 58 supplied from the mode specifying signal generator 22, and either one is output to the storage unit 20. The For example, when the mode designating signal 58 is a signal designating the averaging processing mode, the selector 30 selects the digital signal output from the adder 26 and outputs it to the storage unit 20. When the mode designation signal 58 is a signal designating the interleave processing mode, the selector 30 selects the digital signal output from the multiplexer 28 and outputs it to the storage unit 20. In this embodiment, since the averaging process and the interleaving process are performed by the adder 26 and the multiplexer 28, the storage unit 20 stores the digital signal subjected to the averaging process or the interleaving process. Therefore, the averaging process and the interleave process can be performed at high speed by the form described with reference to FIG. With the above configuration, it is possible to provide an A / D conversion apparatus 110 that can selectively perform averaging processing and interleaving processing for simultaneously sampling a plurality of A / D converters.

  FIG. 8 shows a block diagram of an A / D conversion device 120 according to the second embodiment. The A / D converter 120 includes a plurality of analog signal input units (10a to 10d), A / D converters (12a to 12d), a sampling clock signal generator 14, a reference clock signal generator 16, a processing unit 18, and a storage. Unit (20a to 20d), a mode designation signal generator 22, an analog signal distributor 32, and a distribution control signal generator 34. The processing unit 18 includes an interleave processing unit 18a and an averaging processing unit 18b. In FIG. 8, the components denoted by the same reference numerals as those in FIG. 4 perform the same or similar functions and operations as the corresponding components in FIG.

  In the present embodiment, the interleave processing unit 18a included in the processing unit 18 performs an interleaving process in which sample values obtained by alternately sampling four A / D converters (12a to 12d) are rearranged. . By the interleaving process, the A / D conversion apparatus 100 can obtain the same sample value as when sampling at a sampling rate four times the sampling rate of one A / D converter 12a (12b, 12c, or 12d). .

  The averaging processing unit 18b performs a process of averaging the sample values obtained by simultaneously sampling the four A / D converters (12a to 12d). By this averaging process, a resolution four times the quantization resolution of one A / D converter 12a (12b, 12c or 12d) can be obtained.

  The mode designation signal generator 22 generates a mode designation signal 58 that designates each component in the A / D converter 120 to operate in either the averaging processing mode or the interleave processing mode. The mode designation signal 58 is supplied to the processing unit 18 and the sampling clock signal generator 14. In the processing unit 18, either the interleaving processing unit 18 a or the averaging processing unit 18 b is selected based on the mode designation signal 58.

  The distribution control signal 60 is a signal that designates the distribution destination A / D converters (12a to 12d) of the analog signals (50a to 50d) input to the analog signal distributor 32. In the present embodiment, the distribution control signal generator 34 has a one-to-four distribution designating distribution of one analog signal to four A / D converters and one A / D conversion of one analog signal. A distribution control signal 60 is generated that designates either one-to-one distribution that designates distribution to a container. The analog signal distributor 32 converts the input analog signals (50a to 50d) to any of the A / D converters (12a to 12d) based on the distribution control signal 60 supplied from the distribution control signal generator 34. Distribute.

  For example, when the distribution control signal 60 specifies 1 to 4 distribution and the analog signal distributor 32 selects the analog signal 50a, the analog signal 50a is distributed to the A / D converters (12a to 12d). . Furthermore, when the interleave processing mode is designated by the mode designation signal 58, the analog signal 50a is interleaved using sample values obtained from the four A / D converters (12a to 12d). Therefore, sample values sampled at a sampling rate four times the sampling rate of one A / D converter 12a (12b, 12c or 12d) can be obtained. At this time, the remaining analog signals 50b, 50c, and 50d are not distributed to the A / D converter.

  When the averaging processing mode is designated by the mode designation signal 58, the analog signal 50a input from the analog signal input unit 10a is obtained by obtaining sample values obtained from the four A / D converters (12a to 12d). And averaged. Therefore, a resolution four times the quantization resolution of one A / D converter 12a (12b, 12c or 12d) can be obtained. At this time, the remaining analog signals 50b, 50c, and 50d are not distributed to any A / D converter.

  When the distribution control signal 60 specifies one-to-one distribution, the analog signals (50a to 50d) are distributed to each of the four A / D converters (12a to 12d). By this distribution, each A / D converter (12a to 12d) can sample the analog signals (50a to 50d).

  Further, the sampling clock signal generator 14 is based on the reference clock signal 54 supplied from the reference clock signal generator 16 and the mode specifying signal 58 supplied from the mode specifying signal generator 22. 56d). When the mode designation signal 58 designates the averaging processing mode, the sampling clock signal generator 14 outputs an averaging processing sampling clock signal for simultaneously sampling the four A / D converters (12a to 12d). It supplies to each of A / D converter (12a-12d).

  When the mode designation signal 58 designates the interleave processing mode, the sampling clock signal generator 14 converts the A / D sampling clock signal for alternately operating the four A / D converters (12a to 12d). Supply to each of the converters (12a-12d).

  The analog signals (50a to 50d) distributed to the A / D converters (12a to 12d) are sampled by the A / D converters (12a to 12d) that perform the sampling operation based on the sampling clock signals (56a to 56d). The sample value is digitized. The sample values digitized by the A / D converters (12a to 12d) are stored in the storage units (20a to 20d).

  The interleave processing unit 18a performs an interleave process for alternately rearranging the sample values. For example, the interleave processing unit 18a converts a digitized sample value stored in the storage unit (20a to 20d) into a voltage value based on a voltage conversion coefficient for converting a digital signal into a voltage value. Each voltage value is rearranged alternately. Therefore, an analog voltage value can be obtained at intervals of the sampling period. Moreover, the averaging process part 18b performs the averaging process which averages a sample value. For example, the averaging processing unit 18b converts the digital signal stored in the storage units (20a to 20d) into a voltage value based on the voltage conversion coefficient, and adds the converted voltage values. With the configuration shown in FIG. 8, the A / D converter 100 converts the input analog signals (50 a to 50 b) into one or more A / D conversions based on the mode designation signal 58 and the distribution control signal 60. Can be selectively processed using a vessel (12a-12b).

  FIG. 9 is an example of an embodiment of the analog signal distributor 32. The analog signal distributor 32 includes a multiplexer 28a that distributes an input signal and an output signal 4: 1 and a multiplexer 28b that distributes the signal 2: 1. Based on the distribution control signal 60, the analog signal distributor 32 can distribute the analog signals (50 a to 50 b) to each of the A / D converters (12 a to 12 d) on a one-to-one basis, or on a one-to-four basis. it can. For example, when the distribution control signal 60 designating the one-to-one distribution is input to the analog signal distributor 32, the multiplexer 28b converts the input signals from the input terminals A group (A1 to A4) to A / D. Output to each of the converters (12a to 12d). Since the A group is connected to each of the analog signal input units (10a to 10d) in a one-to-one relationship, the analog signals (50a to 50d) are one-to-one in the A / D converters (12a to 12d). Distributed to each.

  In addition, when the distribution control signal 60 for designating the one-to-four distribution is input to the analog signal distributor 32, the multiplexer 28b performs A / D conversion on the input signal from the input terminal group B (B1 to B4). Output to each of the devices (12a to 12d). The group B is connected to the multiplexer 28a, and the multiplexer 28a outputs any one of the analog signals 50a, 50b, 50c, or 50d input from each of the analog signal input units (10a to 10d). Therefore, any one of the analog signals (50a to 50d) is distributed to each of the A / D converters (12a to 12d). Moreover, it is preferable that the analog signal distributor 32 can be further distributed arbitrarily by changing the number and configuration of multiplexers.

  FIG. 10A shows a distribution method for distributing an analog signal to an A / D converter on a one-to-one basis. (See FIG. 9 together with FIG. 10) When the distribution control signal 60 is a signal designating one-to-one distribution, the input terminal group A (A1 to A4) of the multiplexer 28b is selected and the analog signal input unit (10a to 10a) is selected. The analog signals (50a to 50d) input from 10d) are distributed one-to-one to the A / D converters (12a to 12d).

  FIG. 10B shows a distribution method for distributing the analog signals to the A / D converters on a one-to-four basis. (See FIG. 9 together with FIG. 10) When the distribution control signal 60 is a signal designating one-to-four distribution, the input terminal group B (B1 to B4) of the multiplexer 28b is selected, and the multiplexer 28a has four input terminals. By selecting one of the analog signals, any one of the analog signals 50a, 50b, 50c, and 50d input from the analog signal input units (10a to 10d) is converted into four A / D converters ( 12a to 12d). For example, in the example shown in FIG. 10B, the input terminal A of the multiplexer 28a is selected, and the input terminal group B (B1 to B4) of the multiplexer 28b is selected. At this time, the analog signal 50a is distributed to the A / D converters (12a to 12d). Further, by selecting each of the input terminals B, C, and D of the multiplexer 28a, any one of the corresponding analog signals 50b, 50c, and 50d is distributed to the A / D converters (12a to 12d). In another embodiment, it is preferable that a desired distribution such as a one-to-two distribution can be performed as shown in FIG.

  FIG. 11A shows a timing chart of the sampling clock signal for averaging output from the sampling clock signal generator 14. In this embodiment, the mode designation signal 58 is a binary signal expressed by two voltage values, Hi (logical value “1”) and Lo (logical value “0”). Hi designates generation of a sampling clock signal for interleaving processing, and Lo designates generation of a sampling clock signal for averaging processing. When the mode designation signal 58 is Lo, the sampling clock signals (56a to 56d) are output in synchronization with the reference clock signal 54. The A / D converters (12a and 12b) perform a sampling operation based on the sampling clock signals (56a and 56b). The sampling timing of this sampling operation is synchronized with the reference clock signal 54.

  FIG. 11B shows a timing chart of the sampling clock signal for interleaving processing output from the sampling clock signal generator 14. When the mode designation signal 58 is Hi, the sampling clock signal 56a is output by dividing the reference clock signal 54 by 1/4, and the sampling clock signals 56b, 56c and 56d are divided by 1/4 by the reference clock signal 54. The phase of the sampling clock signal 56a obtained in this way is further shifted by ¼ phase. The A / D converters (12a and 12b) perform a sampling operation based on the sampling clock signals (56a and 56b). The sampling timing of this sampling operation is synchronized with the reference clock signal 54.

  In general, an A / D converter requires a certain time (hereinafter referred to as sampling time) in order to convert an analog signal into a digital signal. Therefore, the maximum sampling rate at which one A / D converter can sample in a fixed time is determined. Therefore, interleaving processing in which a plurality of A / D converters are operated alternately becomes effective. With the above configuration, when performing interleave processing, it is possible to select an A / D converter to be used according to the sampling rate. For example, when testing a semiconductor device having a high sampling rate, four A / D converters are used for testing. For example, when testing a semiconductor device having a low sampling rate, one A / D converter is used. In this case, a plurality of semiconductor devices can be tested simultaneously. Therefore, since the A / D converter to be used can be selected according to the characteristics of the semiconductor device, the semiconductor device can be efficiently tested.

  In addition, when performing an averaging process, an A / D converter can be selected according to the resolution required for measurement. For example, when testing a semiconductor device that needs to measure an analog signal with high resolution, four A / D converters are used for testing. For example, when testing a semiconductor device that may have a low resolution, one A / D converter is used. In this case, a plurality of semiconductor devices can be tested simultaneously. Therefore, since the A / D converter to be used can be selected according to the characteristics of the semiconductor device, the semiconductor device can be efficiently tested. Therefore, the A / D converter can be selected according to the processing content necessary for testing the semiconductor device under test.

  Next, a calibration apparatus that calibrates errors such as time errors occurring between a plurality of A / D converters will be described. In general, when a sampling operation is performed using a plurality of A / D converters, a desired sampling timing can be obtained depending on a difference in characteristics of each A / D converter and a characteristic of a transmission path of a sampling clock signal. A time error will occur.

  FIG. 12 shows a time error in sampling timing that occurs between two A / D converters, the first A / D converter and the second A / D converter. In general, when a plurality of A / D converters are used and the A / D converters are alternately sampled at a desired time interval, the sampling clock signal is different due to the difference in the characteristics of each A / D converter. Therefore, there is a difference in the time from the input to the A / D converter until the actual sampling. This deviation is defined as a time error τ. In this figure, interleaving processing is performed using two A / D converters, but the time taken to actually perform the sampling operation after inputting the sampling clock signal is a characteristic of each A / D converter. Therefore, there is a time error τ. When processing a signal under measurement using sample values obtained by alternately sampling a plurality of A / D converters, the signal under measurement cannot be accurately reproduced unless the sampling timing is equal. Therefore, it is necessary to calibrate the time error τ.

  FIG. 13 shows an A / D conversion device 130 in the third embodiment. The interleave processing unit 18a in the present embodiment includes a calibration device 70. The functional blocks denoted by the same reference numerals as those in FIG. 4 have the same functions as the functional blocks described in FIG. The calibration device 70 calibrates an error generated between a plurality of A / D converters and outputs an output signal 90. For example, the calibration device 70 calibrates a time error, a gain error, and an offset error. Therefore, the A / D conversion apparatus 100 can output an output signal 90 in which errors generated between a plurality of A / D converters are calibrated.

  FIG. 14 shows a detailed configuration of the calibration apparatus 70. The calibration device 70 includes a calibration value calculation unit 70a, an error calibration unit 70b, and a reading unit 76. The calibration value calculation unit 70 a includes an error calculation unit 72 and an error calibration value calculation unit 74. The error calibration value calculation unit 74 includes a time error calibration value calculation unit 74a, a gain calibration value calculation unit 74b, and an offset calibration value calculation unit 74c. The error calibration unit 70 b includes a gain / offset calibration unit 78 and a time error calibration unit 80.

  The reading unit 76 reads the sample value stored in the storage unit 20a or the storage unit 20b and outputs it to either the calibration value calculation unit 70a or the error calibration unit 70b. The error calculation unit 72 calculates an error that occurs between the plurality of A / D converters based on the sample value supplied from the reading unit 76. For example, the error calculation unit 72 calculates a time error, a gain error, and an offset error that occur between a plurality of A / D converters, and outputs them to the error calibration value calculation unit 74. For example, the error calculator 72 may calculate the error calibration value based on the sample value of the test signal used for calculating the error calibration value. For example, the test signal is preferably a known waveform such as a sine wave or a cosine wave. Further, for example, the error calculation unit 72 may calculate an error calibration value based on a conversion value obtained by Fourier transform of the sample value.

  The error calibration value calculation unit 74 calculates an error calibration value based on the error supplied from the error calculation unit 72 and outputs the error calibration value to the calibration value calculation unit 70a. For example, the error calibration value calculation unit 74 outputs an error calibration value used to calibrate each error to the error calibration unit 70b based on the time error, gain error, and offset error supplied from the error calculation unit 72. . Based on the time error supplied from the error calculation unit 72, the time error calibration value calculation unit 74a calculates a time error calibration value 88a used for a calculation for correcting the time error. Based on the gain error supplied from the error calculation unit 72, the gain calibration value calculation unit 74b calculates a gain error calibration value 88b used for the calculation for calibrating the gain error. Based on the offset error supplied from the error calculation unit 72, the offset calibration value calculation unit 74c calculates an offset error calibration value 88c that is used for calculation to calibrate the offset error. Therefore, the calibration value calculation unit 70a can calculate an error calibration value used for an operation for calibrating an error generated between a plurality of A / D converters.

  The error calibration unit 70b calibrates an error generated between the plurality of A / D converters based on the error calibration value supplied from the calibration value calculation unit 70a, and outputs an output signal 90. The gain / offset calibration unit 78 calibrates the gain error based on the gain error calibration value 88b. Further, the gain / offset calibration unit 78 calibrates the offset error based on the offset error calibration value 88c. The time error calibration unit 80 calibrates the time error based on the time error calibration value 88a. Therefore, the error calculation unit 70b can calibrate an error that occurs between a plurality of A / D converters. Since the calibration device 70 can calibrate an error generated between a plurality of A / D converters by calculation, the error can be calibrated with high accuracy.

  When the interleave processing unit 18 a is selected by the mode designation signal 58, the gain / offset calibration unit 78 calibrates the gain error and offset error of the sample value and outputs them to the time error calibration unit 80. The time error calibration unit 80 calibrates the time error and outputs an output signal 90. When the averaging processing unit 18b is selected by the mode designation signal 58, the gain / offset calibration unit 78 calibrates the gain error and the offset error and outputs them to the averaging processing unit 18b. The averaging processing unit averages the sample values (84a and 84b) of the signal under measurement calibrated for gain error and offset error.

  An example of the calculation method of the error calibration value in the calibration value calculation unit 70a will be described below. In order to calculate an error calibration value based on an error between a plurality of A / D converters, a test signal is input to the two A / D converters 12a and 12b to be error calibrated to perform a sampling operation. . For example, the test signal is a sine wave sin (2π · f · t). Here, f represents an arbitrary frequency, and t represents time.

の形に表すことができる。ここで、Ａ１及びＡ２はゲインを示し、Ｂ１及びＢ２はオフセットを示す。また、各々のＡ／Ｄ変換器のサンプリング動作の時間間隔を２Ｔｓ（ｆ＝１／２Ｔｓ）、Ａ／Ｄ変換器１２ａとＡ／Ｄ変換器１２ｂのサンプリングタイミングが、Ｔｓ＋τ時間の位相ずれがあるとき（時間誤差τ）、次式が成り立つ。図１４において、時間誤差τを時間誤差８６ａ、ゲインＡ１及びＡ２をゲイン８６ｂ、オフセットＢ１及びＢ２をオフセット８６ｃとする。

従って、時間誤差τは

となる。誤差算出部７２は、時間誤差τを時間誤差校正値算出部７４ａに出力する。誤差算出部７２は、ゲインＡ１及びＡ２をゲイン校正値算出部７４ｂに出力する。また、誤差算出部７２は、オフセットＢ１及びＢ２をオフセット校正値算出部７４ｃに出力する。 Sample values (82a and 82b) of the test signals sampled and digitized by the respective A / D converters are stored in the storage units (20a and 20b). The reading unit 76 reads the sample values (82a and 82b) of the test signal from the storage units (20a and 20b), and outputs them to the error calculation unit 72. The error calculation unit 72 performs discrete Fourier transform on the sample values (82a and 82b) of the input test signal. The results obtained from this discrete Fourier transform are

It can be expressed in the form of Here, A1 and A2 indicate gain, and B1 and B2 indicate offset. Further, the time interval of the sampling operation of each A / D converter is 2Ts (f = 1 / 2Ts), and the sampling timing of the A / D converter 12a and the A / D converter 12b has a phase shift of Ts + τ time. When (time error τ), the following equation holds. In FIG. 14, time error τ is time error 86a, gains A1 and A2 are gain 86b, and offsets B1 and B2 are offset 86c.

Therefore, the time error τ is

It becomes. The error calculation unit 72 outputs the time error τ to the time error calibration value calculation unit 74a. The error calculation unit 72 outputs the gains A1 and A2 to the gain calibration value calculation unit 74b. Further, the error calculation unit 72 outputs the offsets B1 and B2 to the offset calibration value calculation unit 74c.

  The time error calibration value calculation unit 74 a calculates a time error calibration value 88 a based on the time error τ supplied from the error calculation unit 72. For example, the time error calibration value calculation unit 74a calculates the time error calibration value 88a by the following method.

δはデルタ関数

この２つの波形をインターリーブ処理した場合のフーリエ変換は、（３）（４）を用いて

と表すことができる。 Assume that two sampling timings are expressed with the subscripts of even and odd. The phase of the sampling timing odd is assumed to be shifted by Ts + τ time with respect to the phase of the sampling timing even, and the following equation is established if each sampling timing is Fourier-transformed as P _even and P _odd .

δ is the delta function

The Fourier transform when these two waveforms are interleaved is processed using (3) and (4).

It can be expressed as.

  When the time error τ occurs between the two A / D converters (12a and 12b), the term having the same frequency as the sampling frequency (1 / 2Ts) in the Fourier transform has no time error τ. And the same frequency term as the sampling frequency (1 / 2Ts). The time error calibration value calculator 74a calibrates the time error so that the same frequency term as the sampling frequency in the Fourier transform matches the frequency term when the time error τ does not occur.

となる。 Here, the value of the term of the sampling frequency (1 / 2Ts) when the time error τ has occurred can be obtained by substituting k = 1 into the equation (5).

It becomes.

となる。

が得られる。この時間誤差校正値８８ａを、サンプリングタイミングＰ_ｏｄｄでサンプリング動作するＡ／Ｄ変換器のサンプル値をフーリエ変換した結果に乗じることで時間誤差の校正ができる。 The term of the sampling frequency (1 / 2Ts) when the time error τ does not occur in the sampling timing of the two A / D converters (12a and 12b) is

It becomes.

Is obtained. The time error calibration can be performed by multiplying the time error calibration value 88a by the result of Fourier transform of the sample value of the A / D converter that performs the sampling operation at the sampling timing _Pod .

  The gain calibration value calculation unit 74b calculates the gain calibration value 88b based on the input gain 86b. There are several methods for calculating the gain calibration value 88b. For example, the reciprocal of the gain 86b may be used. If the gain 86b is “A1”, the gain calibration value 88b is “1 / A1”.

  The offset calibration value calculation unit 74c calculates an offset calibration value 88c based on the input offset 86c. There are several methods for calculating the offset calibration value 88c. For example, in the present embodiment, the offset calibration value 88c is a value obtained by changing the sign of the offset 86c calculated by the error calculation unit 72. For example, if the offset 86c is “B1”, the offset calibration value 88c is “−B1”. With the above method, the calibration value calculation unit 70a calculates the time error calibration value 88a, the gain calibration value 88b, and the offset calibration value 88c.

  Next, sample values (84a and 84b) obtained by sampling the signal under measurement output from the semiconductor device 98, a time error calibration value 88a and a gain calibration value 88b calculated in advance by the calibration value calculation unit 70a. The error calibration unit 70b that calibrates the error based on the offset calibration value 88c will be described.

  The reading unit 76 reads sample values (84a and 84b) of the signal under measurement obtained by sampling the signal under measurement from the storage units (20a and 20b), and outputs them to the gain / offset calibration unit 78. The gain / offset calibration unit 78 receives the sample of the signal under measurement input based on the sample values (84a and 84b) of the signal under measurement supplied from the reading unit 76, the gain calibration value 88b, and the offset calibration value 88c. Calibrate the gain and offset of the values (84a and 84b). There are several calibration methods. For example, the following method is used in this embodiment.

  If the sample value is D (t), the gain and offset calibrated value is D '(t), the gain calibration value 88b is "G", and the offset calibration value 88c is "O", the following holds.

D ′ (t) = G · D (t) + O
By performing this calculation on the sample values (84a and 84b) of the signal under measurement, the gain and offset included in each sample value can be calibrated. The sample values (84a and 84b) of the signal under measurement in which the gain and offset calibration unit 78 has calibrated the gain and offset are input to the time error calibration unit 80.

  The time error calibrating unit 80 performs conversion by adding a time component to the sample values (84a and 84b) of the signal under measurement whose input gain and offset are calibrated. This conversion is a conversion that reproduces an input analog signal by adding a time component to the sample values (84a and 84b) of the signal under measurement. Since the sample value output from the A / D converter does not include a time component, it is necessary to include a time component in order to reproduce a waveform from the sample value. For example, this transformation is a Fourier transformation, and in this embodiment, a discrete Fourier transformation is used.

The time error is calibrated based on the result of the discrete Fourier transform and the time error calibration value 88a. There are several time error calibration methods. For example, in the present embodiment, the following method is used.

  Therefore, the time error calibration unit 80 can output the output signal 90 calibrated for gain, offset, and time error by this calculation. The calibration device 70 that calibrates the gain, offset, and time error described above is preferably realized by an arithmetic device that performs predetermined processing by a program. For example, the arithmetic device may be a computer such as a workstation that performs predetermined processing by a program. By realizing the calibration device 70 using an arithmetic device, it is not necessary to provide hardware such as a variable resistor or a delay circuit for error calibration. Since calibration is performed by calculation, it does not depend on the characteristics of an element such as a delay circuit, so that highly accurate calibration can be performed. Therefore, a recording medium storing a program for realizing the calibration device 70 using an arithmetic device is provided. Further, since the error is calibrated based on the sample value, it is possible to calibrate the error included in the sample value measured by the A / D converter that converts the conventional analog signal into the digital signal. With the above configuration, it is not necessary to provide elements such as a delay circuit and a variable resistor as time error, gain, and offset calibration means, and error calibration can be facilitated. In addition, since the range in which the error can be calibrated does not depend on the performance of an element such as a delay circuit, highly accurate calibration can be performed.

  FIG. 15 shows a semiconductor device test apparatus for testing a semiconductor device that outputs an analog signal according to the fourth embodiment. This semiconductor device test apparatus includes a waveform digitizer 95 having an arithmetic unit 97 and an A / D converter 100, a recording medium 38, a pattern generator 91, a waveform shaper 92, a comparator 93, and a performance having a semiconductor device contact portion 94. A board 96 is provided. In this embodiment, the calibration device 70 is realized using an arithmetic device 97 based on a program recorded on the recording medium 38. Here, the arithmetic unit 97 is preferably a general-purpose computer such as a workstation.

  A semiconductor device 98 that outputs an analog signal to be tested is placed on the semiconductor device contact portion 94. For example, the semiconductor device contact portion 94 may be a socket that is electrically connected to an input / output terminal of the semiconductor device 98. The pattern generator 91 generates a semiconductor device input signal 42 to be supplied to the semiconductor device 98 and outputs it to the waveform shaper 92. The pattern generator 91 outputs the theoretical value to be output from the semiconductor device 98 to the comparator 93. The waveform shaper 92 shapes the semiconductor device input signal 42 according to the characteristics of the semiconductor device 98 and outputs it to the semiconductor device contact portion 94. The semiconductor device contact portion 94 outputs the semiconductor device input signal 40 supplied from the waveform shaper 92 to the semiconductor device 98. The semiconductor device 98 outputs an analog signal 50 to the semiconductor device contact portion 94 based on the input semiconductor device input signal 40. The semiconductor device contact portion 94 outputs an analog signal 50 that is a signal under measurement to the A / D converter 100. The A / D conversion apparatus 100 converts the analog signal 50 supplied from the semiconductor device contact portion 94 into a digital signal.

  The A / D conversion apparatus 100 is the A / D conversion apparatus 100 according to the first embodiment described with reference to FIG. 4, and processes the input analog signal 50 in either an averaging process or an interleave process. Furthermore, the A / D conversion apparatus 100 includes a calibration apparatus 70 that calibrates time errors that occur between the plurality of A / D converters shown in the third embodiment, so that an error-corrected output signal is provided. 90 can be output. The comparator 93 determines the quality of the semiconductor device 98 based on the error-corrected output signal 90 and the theoretical value supplied from the pattern generator 91, and outputs a determination signal 52. With this configuration, one test apparatus can perform two processes, an averaging process and an interleave process. Also, it is possible to realize a semiconductor device test apparatus that can easily calibrate a time error generated between a plurality of A / D converters.

  FIG. 16 shows a semiconductor device test apparatus for simultaneously testing a plurality of semiconductor devices that output analog signals according to the fifth embodiment. This semiconductor device test apparatus includes a waveform digitizer 95 having an arithmetic unit 97 and an A / D converter 100, a recording medium 38, a pattern generator 91, a waveform shaper 92, a comparator 93, and semiconductor device contact portions (94a to 94d). ). In this embodiment, the calibration device 70 is realized using an arithmetic device 97 based on a program recorded on the recording medium 38. Here, the arithmetic unit 97 is preferably a general-purpose computer such as a workstation.

  The pattern generator 91 generates a semiconductor device input signal 42 to be supplied to the semiconductor devices (98a to 98d) and outputs it to the waveform shaper 92. Further, the pattern generator 91 outputs theoretical values to be output from the semiconductor devices (98a to 98d) to the comparator 93. The waveform shaper 92 shapes the semiconductor device input signal 42 supplied from the pattern generator 91 according to the characteristics of the semiconductor devices (98a to 98d), and outputs it to the semiconductor device contact portions (94a to 94d). The semiconductor device contact portions (94a to 94d) supply the semiconductor device input signal 40 supplied from the waveform shaper 92 to each of the semiconductor devices (98a to 98d). The semiconductor devices (98a to 98d) output analog signals (50a to 50d) to the semiconductor device contact portions (94a to 94d) based on the input semiconductor device input signal 40. The semiconductor device contact portions (94 a to 94 d) output analog signals (50 a to 50 d) to the A / D conversion device 100.

  The A / D conversion device 120 is the A / D conversion device 120 according to the second embodiment described with reference to FIG. 8, and the input analog signals (50a to 50d) are either averaged or interleaved. Process with. Further, the A / D converter 120 includes a calibration device 70 that calibrates a time error generated between the plurality of A / D converters shown in the third embodiment, so that an error-corrected output signal is provided. (90a to 90d) can be output. The comparator 93 determines the quality of the semiconductor device 98 based on the error calibrated output signals (90a to 90d) and the theoretical values supplied from the pattern generator 91, and outputs the determination signals (52a to 52d). To do. Since the A / D converter 120 described with reference to FIG. 8 includes the analog signal distributor 32, one or a plurality of A / D converters used for processing are selected according to the processing contents of the analog signal. Can be used. Moreover, two processes of an averaging process and an interleave process can be performed with one test apparatus. In addition, it is possible to realize a semiconductor device test apparatus that can easily calibrate a time error generated between a plurality of A / D converters.

  As is clear from the above description, the single A / D converter 120 performs an averaging process for simultaneously sampling a plurality of A / D converters and an interleaving process for alternately sampling a plurality of A / D converters. realizable. Further, the A / D converter used for processing can be selectively changed according to the processing content of the analog signal. Further, calibration of a time error occurring between a plurality of A / D converters can be performed using an arithmetic device. Therefore, the time error can be calibrated without using a delay circuit.

  As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the above-described embodiment. It is apparent from the description of the scope of claims that embodiments with such changes or improvements can be included in the technical scope of the present invention.

10 (10a, 10b, 10c, 10d) ... Analog signal input unit 12 (12a, 12b, 12c, 12d) ... A / D converter 14 ... Sampling clock signal generator 16 ... Reference clock signal generator 18 ... Processing unit 18a ... Interleave processing unit 18b ... Averaging processing unit
20 (20a, 20b, 20c, 20d) ... storage unit 22 ... mode designation signal generator 24 ... delay circuit 26 ... adder
28 (28a, 28b) ... multiplexer 30 ... selector 32 ... analog signal distributor 34 ... distribution control signal generator 38 ... recording medium 40, 42 ... semiconductor device input signal 50 (50a, 50b, 50c, 50d) ... analog signal 52 ... Determination signal 54 ... Reference clock signals 56a, 56b, 56c, 56d ... Sampling clock signal 58 ... Mode designation signal 60 ... Distribution control signal 70 ... Calibration device 70a ... Error calculation unit 70b ... Error calibration unit 72 ... Error calculation unit 74 ... Error calibration value calculation unit 74a ... Time error calibration value calculation unit 74b ... Gain calibration value calculation unit 74c ... Offset calibration value calculation unit 76 ... Reading unit 78 ... Gain / offset calibration unit 80 ... Time error calibration units 82a, 82b ... Test Sample values 84a, 84b of the signal ... Sample values 86a of the signal under measurement ... During error τ
86b ... Gain 86c ... Offset 88a ... Time error calibration value 88b ... Gain calibration value 88c ... Offset calibration value 90 (90a, 90b, 90c, 90d) ... Output signal 91 ... Pattern generator 92 ... Waveform shaper 93 ... Comparator 94 ... Semiconductor device contact part 95 ... Waveform digitizer 96 ... Performance board 97 ... Arithmetic unit 98 ... Semiconductor device 100 A / D converter

Claims (20)

An A / D converter that samples an analog signal and converts it into a digital signal,
An analog signal input unit for inputting the analog signal;
A plurality of A / D converters that sample the analog signal input by the analog signal input unit and convert it into the digital signal;
One of an averaging processing sampling clock signal that causes the plurality of A / D converters to perform a sampling operation in synchronism and an interleaving processing sampling clock signal that causes the plurality of A / D converters to perform sampling operation alternately is used A sampling clock signal generator for supplying to a plurality of A / D converters;
An averaging processor for averaging the digital signal output from the A / D converter that has been sampled based on the sampling clock signal for averaging;
An A / D conversion apparatus comprising: an interleave processing unit that performs interleaving processing on the digital signal output from the A / D converter that has performed a sampling operation based on the sampling clock signal for interleaving processing. A mode designation signal generator for generating a mode designation signal for designating one of the averaging process and the interleave process;
2. The A / D conversion apparatus according to claim 1, wherein either the averaging processing unit or the interleaving processing unit is selected based on the mode designation signal. A reference clock signal generator for generating a reference clock signal;
When averaging processing is designated by the mode designation signal, the sampling clock signal generator supplies the averaging processing sampling clock signal synchronized with the reference clock signal to each of the plurality of A / D converters. Supply
When interleave processing is specified by the mode specifying signal, the sampling clock signal generator converts the interleave processing sampling clock signals having different phases from each other based on the reference clock signal to the plurality of A / D converters. The A / D converter according to claim 2, wherein the A / D converter is supplied to each of the A / D converters. A plurality of storage units for storing digital signals output from each of the plurality of A / D converters;
4. The A / D conversion apparatus according to claim 1, wherein the averaging processing unit and the interleaving processing unit perform processing based on the digital signal stored in the storage unit. 5. The plurality of A / D converters include a first A / D converter and a second A / D converter,
A time lag between a predetermined timing to be sampled by the second A / D converter and a timing at which the second A / D converter is actually sampled with respect to a timing sampled by the first A / D converter. An error calculation unit for calculating a time error;
Based on the calculated time error, an error calibration value calculation unit that calculates a time error calibration value used for a calculation for calibrating the time error of the second A / D converter;
A reading unit that reads out the digital signal from the storage unit in which a digital signal obtained by sampling a signal under measurement that is an analog signal to be measured is stored;
The reading unit calibrates the time error generated in the second A / D converter when the signal under measurement is sampled based on the digital signal read from the storage unit and the time error calibration value. The A / D conversion device according to claim 4, further comprising: an error calibration unit that performs calculation. An A / D converter that samples an analog signal and converts it into a digital signal,
An analog signal input unit for inputting the analog signal;
The digital signal output from the first A / D converter that samples the analog signal and converts it to the digital signal, and the second A / D converter that samples the analog signal and converts it to the digital signal. An adder for adding the digital signal;
A multiplexer that alternately inputs and sequentially outputs the digital signal output from the first A / D converter and the digital signal output from the second A / D converter;
A selector that selects either the output value of the adder or the output value of the multiplexer. An A / D converter that samples an analog signal and converts it into a digital signal,
A plurality of analog signal input units to which different analog signals are respectively input;
A plurality of A / D converters that sample the analog signal and convert it to the digital signal;
An analog signal distributor that distributes the analog signal to any one of the plurality of A / D converters or to the plurality of A / D converters according to the contents of the digital processing of the analog signal. An A / D converter characterized by the above. The plurality of A / D converters are provided corresponding to the plurality of analog signal input units, respectively.
The A / D converter according to claim 7, wherein the analog signal distributor distributes the analog signal input from one analog signal input unit to a plurality of the A / D converters. . One of an averaging processing sampling clock signal that causes the plurality of A / D converters to perform a sampling operation in synchronism and an interleaving processing sampling clock signal that causes the plurality of A / D converters to perform sampling operation alternately is used as the above-mentioned A sampling clock signal generator for supplying to a plurality of A / D converters;
An averaging processor for averaging the digital signal output from the A / D converter that has been sampled based on the sampling clock signal for averaging;
The interleaving processing unit further comprising: an interleaving processing unit that interleaves the digital signal output from the A / D converter that has been sampled based on the sampling clock signal for interleaving processing. A / D converter. A mode designation signal generator for generating a mode designation signal for designating one of the averaging process and the interleave process;
Based on the mode designation signal, either the averaging processing unit or the interleaving processing unit is selected,
Based on the selected processing, a distribution control signal that specifies distribution of the analog signal to any one of the plurality of A / D converters or to the plurality of A / D converters is performed. The A / D conversion apparatus according to claim 9, further comprising a distribution control signal generator to be supplied to the converter. A calibration device that calibrates an error between a first A / D converter that samples an analog signal and converts it to a digital signal, and a second A / D converter that samples the analog signal and converts it to a digital signal. And
A time lag between a predetermined timing to be sampled by the second A / D converter and a timing actually sampled by the second A / D converter with respect to the timing sampled by the first A / D converter. Based on a sample value obtained by sampling a test signal used to calculate a certain time error, an error calculation unit for calculating the time error;
Based on the calculated time error, an error calibration value calculation unit that calculates a time error calibration value used for a calculation for calibrating the time error of the second A / D converter;
A reading unit that reads the sample value from a storage unit that stores a sample value obtained by sampling the signal under measurement that is an analog signal to be measured;
Calculation for calibrating the time error generated in the second A / D converter when the signal under measurement is sampled by the reading unit based on the sample value and the time error calibration value read from the storage unit A calibration device comprising: an error calibration unit that performs the following.   The error calibration unit performs discrete Fourier transform on the sample value of the signal under measurement read from the storage unit by the readout unit, and converts the discrete Fourier transform value obtained by the discrete Fourier transform and the time error calibration value. The calibration apparatus according to claim 11, wherein the time error is calibrated by performing calculation based on the calibration. The error calculation unit calculates a gain and an offset of the first A / D converter and the second A / D converter,
The error calibration value calculator is
A gain calibration value calculator for calculating gain calibration values of the first A / D converter and the second A / D converter based on the gain;
An offset calibration value calculator that calculates offset calibration values of the first A / D converter and the second A / D converter based on the offset;
The error calibration unit is configured to perform the first A / D converter and the first calibration based on the sample value of the signal under measurement read from the storage unit by the reading unit, the gain calibration value, and the offset calibration value. The calibration apparatus according to claim 11, further comprising: a gain / offset calibration unit that calibrates the gain and the offset of a 2A / D converter.   The gain / offset calibration unit performs an operation of adding an offset calibration value to a value obtained by multiplying the sample value of the signal under measurement sampled by the second A / D converter by the gain calibration value. The calibration device according to claim 13. An error calibration method for calibrating an error generated between a first A / D converter that samples an analog signal and converts it to a digital signal and a second A / D converter that samples the analog signal and converts it to a digital signal. And
A time lag between a predetermined timing to be sampled by the second A / D converter and a timing at which the second A / D converter is actually sampled with respect to a timing sampled by the first A / D converter. Calculating a time error;
Calculating a time error calibration value for use in a calculation for calibrating the time error based on the time error;
Calibrating the time error generated when the signal under measurement is sampled based on a sample value obtained by sampling the signal under measurement to be measured and the time error calibration value. And error calibration method. The error calibration method is:
Calculating gains and offsets of the first A / D converter and the second A / D converter;
Calculating a gain calibration value and an offset calibration value used for an operation for calibrating the gain and the offset based on the calculated gain and the offset;
Based on the sample value obtained by sampling the signal under measurement, the gain calibration value, and the offset calibration value, the gain and the offset of the first A / D converter and the second A / D converter The error calibration method according to claim 15, further comprising the step of calibrating. A time between a predetermined timing at which the second A / D converter should sample and a timing at which the second A / D converter has actually sampled with respect to the timing at which the first A / D converter has sampled using the arithmetic unit. A recording medium recording a program for calibrating a time error that is a deviation of
A module for causing the arithmetic unit to calculate the time error;
A module for calculating a time error calibration value used for a calculation for calibrating the time error of the second A / D converter based on the calculated time error;
A sample value obtained by sampling the signal under measurement to be measured, and a module for calibrating the time error generated when the signal under measurement is sampled based on the time error calibration value. A recording medium readable by an arithmetic unit in which a program for calibrating the time error is recorded. The program is
A module for causing the arithmetic unit to calculate a gain and an offset;
A module for calculating a gain calibration value and an offset calibration value used for an operation for calibrating the gain and the offset based on the calculated gain and the offset;
Based on the sample value obtained by sampling the signal under measurement, the gain calibration value, and the offset calibration value, the gain and the offset of the first A / D converter and the second A / D converter are calculated. The recording medium according to claim 17, further comprising a module for calibrating. A semiconductor device test apparatus for testing a semiconductor device that outputs an analog signal,
A pattern generator for generating a semiconductor device input signal for testing the semiconductor device;
A performance board for providing the semiconductor device with the semiconductor device input signal output from the pattern generator;
An analog signal input unit for inputting the analog signal output from the semiconductor device;
A plurality of A / D converters that sample the analog signal input from the analog signal input unit and convert it into the digital signal;
Sampling of one of an averaging processing sampling clock signal that causes the plurality of A / D converters to perform a sampling operation synchronously and an interleaving processing sampling clock signal that causes the plurality of A / D converters to perform sampling operations alternately A sampling clock signal generator for supplying a clock signal to the plurality of A / D converters;
An averaging processor for averaging the digital signal output from the A / D converter that has sampled based on the averaging sampling clock signal;
A semiconductor device test apparatus, comprising: an interleave processing unit that performs interleaving processing on a digital signal output from an A / D converter that has performed a sampling operation based on the sampling clock signal for interleaving processing. The plurality of A / D converters include a first A / D converter and a second A / D converter, and the second A / D converter samples at a timing sampled by the first A / D converter. An error calculation unit that calculates a time error that is a time lag between a predetermined timing and a timing at which the second A / D converter is actually sampled;
Based on the calculated time error, an error calibration value calculation unit that calculates a time error calibration value used for a calculation for calibrating the time error of the second A / D converter;
A reading unit that reads the sample value from a storage unit that stores a sample value obtained by sampling the signal under measurement that is an analog signal to be measured;
The reading unit calibrates the time error generated in the second A / D converter when the signal under measurement is sampled based on the sample value and the time error calibration value read from the storage unit. The semiconductor device test apparatus according to claim 19, further comprising: an error calibration unit that performs calculation.

Images