JPS60188858A - Effective value meter - Google Patents

Abstract

PURPOSE:To measure the effective value of an AC distorted wave in a short time by making the measured signal directly to numerical data and calculating the effective value by arithmetic operation. CONSTITUTION:An effective value meter is constituted mainly of an A/D converter 12 and a CPU14 that controls the converter and fetches in output signals. The CPU14 outputs sweep data successively to a D/A converter 3, and at the same time, detects pulse signals of a maximum value detecting circuit 8 and registers a measured signal constituting period to a built-in memory, and calculates a measured signal period tecm. Then, the number of times of sampling is determind by dividing the period tecm by sampling period tCLK and this is used after increasing it by integer times to improve the accuracy of measurement. The specified number of times of sampling is made by a sampling circuit consisting of a converter 12, sample hold circuit 10 and a delay circuit 11, and measured signals are recorded in a memory as a numerical value data string. Gain correction of an amplifier for a gain adjusting 9 is made by operating the numerical value data and the effective value of a measured signal Vrms is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to the inspection of electrical characteristics of ICs and the like in a production line, and particularly to an effective value meter that measures the effective value of AC distorted waves at high speed.

[Background of the invention]

Conventional effective value meters that cover the video band use thermocouples, which has the disadvantage of taking a long time to measure the 811I. In other words, the measurement current heats the thermocouple contacts, and this temperature rise causes heat to be generated in the pair.This method uses a method to stabilize the DC electromotive force, so there is a stabilization time required for the contacts to reach thermal equilibrium.
The measurement time was long. The response time of commercially available products is 1.5 to 7 seconds (Model 3400A: YHP, Model 93: Bunton.

Model 8920A: flake) was not suitable as a production line effective value meter, and the measurement time was long.

[Purpose of the invention]

An object of the present invention is to provide an effective value meter that can measure the effective value of AC distorted waves in a short time.

[Outline of the invention]

Conventionally, the amount of electricity was converted to heat, converted to an effective value, and then converted back to the amount of electricity using a thermocouple, which required a long response time.

There were some drawbacks. On the other hand, the present invention directly converts the amount of electricity into numerical data using an ADC and then calculates the effective value through arithmetic processing, so there is no need to convert it into heat, and the drawbacks of the prior art described above can be overcome. In addition, using the frequency analysis method,
．． No. 1 period tecm [Embodiment of the invention] Maximum value detection circuit that outputs a pulse signal, 9 inputs and fixes the CPU output voltage value, and outputs a sample 8/HIO output signal with CLK from the delay circuit 11: k S - ADC113 that converts (T) into numerical data: DIi is AI)C
12 Output data: The square of the input data with DIi as input, or the output data: DOi Now, the frequency components Fp, Fq
and Fr. The flow yard is shown in Figure 2. The operation is the first step (#1 to ##) of measuring the signal to be measured.
9 and effective value: Second step to calculate t rms (#1~
+14). The time chart is shown in Figure 3.

Reference data: Dui %D Sequentially output to AC3, same F
When this is detected, the relevant measured signal configuration frequency]: Entry table with built-in ti [Memo + JAii = l to n)
] and calculate the measured signal period; ecm.

Figure 4 shows the one-to-one correspondence of sweep data: D''t+-1+ D A C3 output 1 voltage, same (VC04 input terminal: rj, VCO4j vibration frequency: fI and the relevant measured signal configuration frequency j: ti) .

FIG. 5 shows the EnI-IJ table. Below, 1+11 one" ) I -” of Masaru.

5 1 3 Bad condition & no goirin. Therefore, if there is no record, measurement is not possible and troubleshooting is required. The above period in #9: t
Calculate the least common multiple period t ecm of i.

Sweeping of the VCO 4 oscillation frequency is started from time Ts in FIG. Since the maximum value detection circuit 8 pulse signal is detected at time 'rl)l'I'Q and Tr, the corresponding configuration period is stored in memory A1 of the entry table as "pt Same period as memo 1JA2" Q + and same period as memo 1JA3 :Record tr. The sweep is completed at time Te, ending the first step.

2nd step First, according to Figure 6, ADC12°8
The operation of the sunblink circuit consisting of /HIQ and delay circuit 11 will now be described. At time T1, the CPU 14 starts the sampling operation 1 by raising the CLK-i 11" and outputs the signal to the ADC 12.
, S/HIO, etc. The S/HIC captures the input signal value: k r ('I'1)i at the rising edge of CLK, fixes this value, and outputs it. On the other hand, CL via the delay circuit 11
K later rises to '1'' at time T2 and is applied to the ADC 12. Upon receiving CLK at time T2, the ADC 12 lowers the conversion end signal (hereinafter abbreviated as EOC) to the CPU 14.
At the same time, the above analog value kr(T
Start the digital conversion of t). ADC1 at time T3
2 completes the digital conversion, starts up 'BOC, and turns on CI.
) Notify U14 of completion of conversion. At the same time as the CPU 14 receives the completion of the conversion, it lowers the CLK to '0''. Since the ADC 12 output data is the matrix circuit 13 manual data: DIi, the output data: DOi is the square of the analog value k F (TI). Equal.CPU1
4 records the above output data: DOi in the memory 1JBi after falling CLK.

This is the end of the first Sunblink, and the second Sunblink is subsequently started from time T4. Similarly, the operation of each part after time T4 is the first repetition, and time Tt-T4 is the sunblink period t CLK. The details will be explained below.

410 in FIG. 2 indicates the number of sampling times: S. S is the least common multiple period: tecmi Zumbling period: t c
Divide by LK, and multiply this by a positive integer to increase measurement accuracy. #11 performs consecutive S same sunblinks,
The signal to be measured is recorded in the memory Bi as a numerical data string. The memory Bi is shown in FIG. #12 calculates the additive average of the above numerical data: v75. #13 is the additive average: Calculate the square root of Te. #14 corrects the AMP9 gain to obtain the effective value of the signal under measurement: r rms.

After time Te in FIG. 3, the CPU 14 operates at the least common multiple period:
t ecm and number of sunblinks: S, time TI, T
Perform 8 consecutive sunblinks with 48 lengths. Continue adding average value near, square root, 7 and rms value: r rms
All measurements are completed at time Tf.

In summary, the total time required for the measurement is Tf longer than time Ts.

This example targets the measurement of video band (DC to 6MHz) signals, and the VCO oscillation frequency sweep range is 20 to 3'OM.
Hz, IP filter passing frequency: 20 MHz Sunblink period: t In this example, where CLK is set to 10 seconds, a DAC is used as the lamp voltage generator and is controlled by the CPU, so that the lamp voltage is digitized and output from the CPU. The S/H and ADC for reading and viewing on the U are no longer required, making the circuit simpler, smaller, and cheaper. In addition, by adjusting the gain of the signal under test using AIIP, the input range of AI can be used to its fullest and most effective.
Despite being 8-bit, we were able to secure a measurement accuracy of approximately ±0.4% using a high-speed ADC. Furthermore, by using a matrix circuit to replace the square wave 'JL, the measurement time was reduced by 70% compared to the conventional method.

〔Effect of the invention〕

According to the present invention, since the signal to be measured is directly converted into numerical data and the effective value is calculated through arithmetic processing, the measurement time can be shortened compared to the effective value calculation method using a thermocouple. Furthermore, since the numerical data acquisition time is set to a multiple of the period of the signal to be measured, numerical data can be obtained evenly over one cycle, and the data acquisition time can be shortened to the theoretical minimum time for measurement. The speed can be increased.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of an embodiment of the present invention, Fig. 2 is a flow chart, Fig. 3 is a time chart, Fig. 6 is a detailed time chart of sampling operation, Fig. 4 is a sweep data, VCO, FIG. 5 is an explanatory diagram of the contents of the entry table, and FIG. 7 is an explanatory diagram of the contents of memo 1JBi. 1...Target 1++ constant signal 4...■CO5...Mixer 8...Local maximum value detection circuit 12...A'DC13.
...Matrix circuit 14...CPU 1 Figure 2Il! l Figure 3 v Figure 4 Figure 5 Figure 7

Claims (1)

[Claims] 1. It consists of an analog-to-digital converter and a computer equipped with means for controlling the analog-to-digital converter and capturing its output signal, and converts the signal under measurement into the analog-to-digital converter.
An effective value meter characterized in that it has a built-in block for performing field processing on n time-series numerical data ζ obtained by digital conversion with a digital converter, and obtains an effective value of a 1111 constant signal.