JPH04110669A - Sampling-type measuring equipment - Google Patents

Abstract

PURPOSE:To eliminate the need for a large-capacity memory and enable measurement signal ranging from a low frequency to a high frequency to be measured by providing a means for selecting a clock signal with a different period and outputting it and then performing sampling in synchronization with a clock signal which is applied to by this means. CONSTITUTION:A CPU 12 adds a control signal S2 to a selector 13 so that a clock signal with an instructed frequency may be selected and then selects a clock signal with a different period and then outputs it. Thus, when signal to be measured is a low frequency, a clock signal with a long period can be selected and output. As a result, sampling is performed with a slow interval by AD conversion means 1 and 3, the number of data to be stored in a memory 6 is not increased even if sampling is performed over several periods of a low- frequency signal to be measured. Also, in the case of the high frequency, a clock signal with a short period can be selected and output and a needed number of sample data can be stored in the memory 6 in a short time, thus enabling the value of signal to be measured to be calculated by an arithmetic unit.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention is a sampling type measuring instrument that measures the value of the signal under test by sampling the signal under test, converting it into digital data, and adding arithmetic operations to this digital value. It is related to.

<Conventional technology> After sampling the voltage or current that is the signal to be measured at a fixed period, converting it into digital data, and adding an exponential averaging operation to this digital value, the effective value or average value of the signal to be measured can be rectified. There are sampling-type measuring instruments that use an exponential averaging method to measure values and active power.

The operating principle of this method will be briefly explained.

First, the measured signal (voltage or current) is sampled at regular intervals to extract instantaneous values) and converted into digital values.

For example, when measuring the average rectified value of the signal under test, the absolute value of the sampling digital value of the signal under test is x(B in equation (1)), and when measuring the effective value, the absolute value of the sampling digital value is If we set the multiplication value as approaches the average rectified value or the true value of (effective value) 2. Therefore, by repeating the calculation of equation (1), the desired measured value can be obtained.

Y(n)=V(n-1)+(1/G)-(x(t
)-y(n-1)) [1) y(n): Exponentialized average calculation result processed up to time t y(n-1): Exponentialized average processed up to time [t-1) Calculation result Xft) : Sampling value at time t (1/G)
Bi-exponential specific constant (1/G<kl), time t
is equivalent to the meaning of the second sampling.

Figure 2 shows y(
It is a figure showing the locus of n). Characteristic (a) is frequency 12
, characteristic (b) is the locus of y(n) in equation (1) when sampling the signal under test with frequency f1, and both are expressed as the exponentization constant 1/G = 1/G1 in equation (1). There is. As is clear from this figure, both reach the true value Y1 with the number of samples n1. However, for the lower frequency f2 ((fl), the value of y(n) includes a large ripple, making it difficult to grasp the accurate true value Y1.

Therefore, 1/G = 1/G2 (< 1/
G1 ), the ripple becomes small as shown in FIG. 2(C), but this shows that a large number of samples n2 are required until the true value Y1 is reached.

Furthermore, power can also be measured using the above-mentioned index calculation method. In this case, the instantaneous value of the voltage at time t (
If v(B) is the sampling value) and 1(t) is the instantaneous value of the current at the same time t, then in equation (1), just substitute X(t)=V(t)-1(t). .

<Problems to be Solved by the Invention> The above is an example of a sampling measuring instrument using an exponential averaging method, but generally the sampling period of any sampling measuring instrument is constant. As a result, when trying to measure a signal under test over a wide band, there is a problem that memory capacity must be increased if the signal under test is a low band.

Let me explain this. With sampling type measuring instruments, as shown in Figure 2, it is necessary to collect data over a certain number of sample data (for example, 1000 points) or over a certain number of cycles regarding the waveform of the signal under test.
The true value Y1 of the signal under test cannot be measured. Therefore, if sampling is performed at a long period, it will take time to collect the necessary amount of data, and a measuring instrument with good responsiveness cannot be realized.

Therefore, a sampling clock with a short cycle is generated to sample the signal under test.

On the other hand, the sampled voltage/current values are -
After a certain amount of data is stored in the memory, the various calculations described above are performed to calculate the measured value.

Therefore, when data of a low-frequency signal under test is collected using a sampling clock with a short period, the memory capacity increases because the number of data that can be taken in one period of the signal under test increases. This will be explained with reference to FIG.

FIG. 3 shows a state in which signals under measurement of frequencies f3 and f4 are sampled at the same period t1. Note that FIG. 3 is an example when measuring the average value, and the signal under measurement f3.
f4 is expressed as an absolute value.

It is assumed that the number of data that can be stored in the memory is n.For one frequency f3, the number of samples is 10 (01 to 010) in one cycle, and data covering several cycles of the waveform is obtained in n samples. be able to. Therefore, from the data obtained in this way, for example, the average value of the signal to be measured at frequency f3 can be calculated using equation (1).

On the other hand, for frequency f4, n pieces of data R1 to Rn
Even if the data is accumulated, it is still less than the data for one cycle of the waveform (see Figure 3), so such n pieces of data R1~
Rn cannot calculate the value of the signal under test f4, and therefore requires a large-capacity memory capable of storing data for several cycles of the waveform.

An object of the present invention is to provide a sampling type measuring instrument that can measure signals under test from low frequencies to high frequencies without requiring a large capacity memory.

<Means for Solving Section U> In order to solve the above problems, the present invention provides a means (12 , 13), AD conversion means (1, 3) that samples the signal under test in synchronization with the clock signal applied by this means and converts it into a digital signal, and stores the output data of the AD conversion means. This method includes a memory, and an arithmetic unit that reads data stored in the memory, performs an exponential averaging operation, and calculates the value of the signal under measurement.

Effect> The means (12, 13) can select and output clock signals of different cycles.

Therefore, when the signal under test has a low frequency, a clock signal with a long period can be selected and output. As a result, since the AD conversion means performs sampling at slow intervals, the number of data stored in the memory does not increase significantly even if sampling is performed over several cycles of the low-frequency signal under test.

Furthermore, when the signal under test has a high frequency, a clock signal with a short period can be selected and output. As a result, the required number of samples of data can be stored in the memory in a short period of time, and the value of the signal under measurement can be immediately calculated by the arithmetic unit.

<Example> Hereinafter, the present invention will be explained in detail using the drawings.

FIG. 1 is a diagram showing an example of the configuration of a sampling type measuring instrument according to the present invention.

Although the present invention can be applied to sampling-type measuring instruments in general, FIG. 1 shows an example of a configuration applied to a power meter.

An overview of the operation of the device shown in FIG. 1 will be explained. The device shown in FIG. 1 has a low (long) sampling frequency (period).

It has three types of clock signals: medium and high (short). When measuring a signal under test in a low band, it is possible to select sampling at a long period, a medium period at a medium band, and a short period at a high band.

If the operator selects the low band, the sampling frequency will be the low frequency f[. The instantaneous values of v and i sampled at this time are directly sent to the arithmetic unit 11, where they are multiplied by (e.multidot.1) and processed by equation (1) in real time. In other words, in this case, since the sampling period is long,
The next sampling data (instantaneous values of v, i) is
Since there is plenty of time until the data is added to 1, within that period, the multiplication and calculation of equation (1) can be performed for each sampled data sent. There is no need to store the sampled data in memory. That's because there isn't.

When the middle or high band is selected, the sampling frequency becomes the middle frequency fo or the high frequency fh. ■ At this time,
The instantaneous value of i is stored in memory 6.7. That is, in this case, since the sampling period is short, there is not enough time to perform the multiplication and calculation of equation (1) on the previously sent sampling data before the next sampling data is added to the calculator 11. It is from. When the set number N has been accumulated, the data is taken into the arithmetic unit 11, and multiplication and arithmetic processing of equation (1) are performed to calculate the effective power.

In FIG. 1, which will be described in detail below, an analog input voltage (2), which is a signal to be measured, is applied to a terminal P1, and an analog input voltage (i), which is a signal to be measured, is applied to a terminal P2. The device of the present invention can measure the effective power, voltage value, and current value without increasing the memory capacity, even if the voltage V and current 1 are low frequency to high frequency signals of several hundred KHz. .

Analog input voltage (2) applied to terminal P1 is guided to sample and hold circuit (hereinafter simply referred to as S/H circuit) 1. Every time the clock signal SC is applied via the selector 13, the S/H circuit 1 samples the instantaneous value V(t) of the analog input voltage V at that time, and instead of the value captured in the previous sampling, It outputs the newly imported value.

An analog-to-digital converter (hereinafter simply referred to as 80C) 3 converts the sampling voltage V(t) into a digital signal De every time it is output from the S/H circuit 1 (such conversion will be described below). , simply referred to as AD conversion), and
For each AD conversion, EOC (End Of Convex)
rsion) signal.

The ADC3 having the functions described above is commercially available and can be easily obtained. AD stated in the claims
The conversion means consists of 1, this S/H circuit, and 80C.

The memory 6 introduces the output data Oe of the ADC 3,
This output data De is stored at the address indicated by the signal ^D1 applied from the address counter 8. That is, the memory 6 stores data obtained by sampling the voltage (2) applied to the terminal P1.

For example, the address counter 8 is set to 1. A numerical value N is set in advance. Then, each time the EOC signal is added, 1 is subtracted from this numerical value N. Therefore, when the 1 and EOC signals are output, the output of the address counter 8 becomes (g), and the data De at that time is stored in the address (g) of the memory 6.

The same component as the component connected to the terminal P1 is also connected to the terminal P2. That is, S/ connected to terminal P2
H circuit 2 corresponds to S/H circuit F#11, and S/H circuit 2
ADC4 connected to ADC4 corresponds to ADC3, and ADC4 connected to
The memory 7 which introduces the output data D: corresponds to the memory 6, and the address counter 9 corresponds to the address counter 8.

Note that the analog input stream 1 is normally converted into a voltage signal V and applied to the S/H circuit FI#I2. I will add an explanation to this. When an analog input current 1 is passed through a shunt resistor R3 (not shown) whose resistance value is known, the voltage Vi=R
3-i occurs. Since the resistance value R3 is already known, this voltage Vi can be calculated by measuring the voltage Vi to 1=Vi/R8
Therefore, the analog input current is measured. In this specification, the voltage signal v1 obtained by converting the analog input current i in this manner is measured as the analog input current 1.

S/H circuit 2 uses the same clock signal SC as S/H circuit 1.
is added, so the analog input current i is sampled at the same time as the sampling of the S/H circuit 1.

The selector 13 introduces clock signals of different frequencies fh, fi, and fl- from the clock generators 15, 16, and 17, selects one of the clock signals of one frequency according to the control signal S2 from the CPU 12, and performs AD. Output to conversion means. Here, it is assumed that there is a relationship of fL<fn<fh.

Note that each of the clock generators 15 to 17 stops outputting a clock signal in response to a termination signal S1 applied from the address counter 9. The memory 7 stores data obtained by sampling the current 1 applied to the terminal P2.

The computing unit 11 includes a multiplier and an exponential averaging computing unit (not shown).
Built-in. Then, upon receiving the end signal S1, two pieces of data sampled at the same time are read out from the data stored in the memories 6 and 7, and the instantaneous power value is calculated by multiplying these pieces of data together using a multiplier. . Furthermore, using an exponential averaging calculator, calculations such as equation (1), which has not been previously described, are added to this instantaneous power value, and the terminal P1. Calculate the voltage V applied to P2 and the active power (also referred to as average power) of the current t.

Note that the arithmetic unit 11 having the functions of a multiplier and an exponential averaging arithmetic unit as described above can be realized using, for example, a signal processor.

The active power thus obtained by the arithmetic unit 11 is transmitted to the CPU 12 which is a central processing unit, and is displayed on the display 18.

The CPU 12 applies a control signal S2 to the selector 13 so that the tarok signal of the frequency specified by the operator of the apparatus in FIG. 1 is selected. CPII 12 also provides general control of the FIG. 1 apparatus.

The operation of the apparatus thus constructed in FIG. 1 will be explained.

In general, the operator of the device shown in Figure 1 must determine whether the frequencies of the signals to be measured v and l are commercial frequencies (low frequency), around 1 Hz (open frequency), or around 100 KHz (high frequency). It can be known in advance. Now, the frequency of input voltage ■ and input current 1 applied to terminals Pl and P2 is 1KH.
Suppose that it is at position z (intermediate frequency).

In this case, the operator uses the intermediate frequency (intermediate period)
The fi clock signal is instructed to the CPU 12.

In response to this, the CPU 12 sends the signal S2 to the selector 13.
In addition to the frequency Il output from the clock generator 16
The clock signal of is selected by the selector 13 and the two ^D
added to the conversion means.

The S/H circuits 1 and 2 sample the analog input voltage V and the input current 1 at the same time every time the clock signal SC from the clock generator 16 is generated.

This sampled voltage v, flows through the ADC 3,
4, and the digital data De is added one after another.

Converted to D;

A preset number N is set in the address counters 8 and 9 by the setting signals SA and SB. Normally, the numbers N set in the two counters 8 and 9 are the same value.

The ADCs 3 and 4 output EOC signals every time they convert the signals from the S/H circuits 1 and 2 into digital data De and Di. Address counters 8 and 9 each output a value obtained by subtracting 1 from N each time the EOC signal is applied.

For example, when the EOC signal is output, the two address counters 8.9 output address signals AD1 having the same value.

Then, the memory 6.7 stores the kth data De(k
), Difk). As data continues to be written into the memory 6.7 in this manner, the output of the address counter 6.7 will eventually become 00...0. When the contents become all zeros in this way, the address counter 6.7 can output the end signal S1.

That is, this end signal S1 means that a preset number N of sampling data of the input voltage V and the input current 1 have been taken into the memory 6.7. In FIG. 1, the generation of the clock signal SC is stopped by this end signal S1. Since the contents of both address counters 8 and 9 are the same, the end signals of the two address counters 8 and 9 are generated at the same time. Therefore, the end signal of either address counter 8 or 9 is output to the clock generators 15 to 1.
7, but in FIG. 1, the address counter 9
A termination signal S1 is added.

Each of the clock generators 15 to 17 stops outputting clock signals starting from the generation of this end signal S1, so new data output from the 8003.4 stops. Therefore, the count values of address counters 8 and 9 remain at the previous all zeros (OO...0), and the contents of memory 6.7 retain the above-mentioned series of data.

In this way, first, the first illustration! Now, input voltage values and input current values at the same time are captured into the memory by the required number N of data at a specified sampling period (Tm=1/fm).

When the series of data acquisition operations is completed, the arithmetic unit 11 receives the end signal S1 and reads two pieces of data sampled at the same time from among the data stored in the memory 6.7. The task of reading out the two pieces of data is easy. That is, the arithmetic unit 11 is
Data may be read from the same address in the memory 6.7 at the same timing.

Then, the arithmetic unit operates the multiplier to calculate X(t)=v −i t for each set of read data.
(2) Perform the calculation ↑ to calculate the instantaneous power value. Note that X(t), v, and i in equation (2) are the instantaneous force value, instantaneous voltage value, and instantaneous current value at time t. Such calculations are performed on N sets of data read from the memory 6.7, for example, and added to the exponential averaging calculator. Then, if we repeat the calculation of equation (1) above many times, we get (1
y(n) in the equation ) approaches the average power, that is, the effective power value W. In other words, y(n)=W=Σ(v-i) is calculated.

1 1 Furthermore, if the arithmetic unit 11 includes a square arithmetic unit, an exponential average arithmetic unit, and a square root arithmetic unit, the effective value of the signal under measurement can be measured. Furthermore, if the arithmetic unit 11 includes an absolute value arithmetic unit and an indexed average arithmetic unit, the average rectified value of the signal under measurement can be measured. For voltage or current measurements like this,
In FIG. 1, it is sufficient to provide one set of an S/8 circuit, an 800, a memory, and an address counter, and there is no need to provide two sets as in the device of FIG.

For example, when measuring the effective value of the voltage V, the instantaneous value (for example, vB) stored in the memory 6 is read out and squared to obtain X(t) = Vt2.Then, the value of this X(t) is By substituting (1) into equation (1), we can perform the exponential averaging operation.

When measuring the average rectified value, read the instantaneous value νt stored in the memory and calculate this absolute value X1t)=lVt
get l. Then, by substituting this value of x(t) into equation (1) and repeating the indexed average calculation, the herald is obtained.

Measured values such as active power, effective value, and average value obtained in this way are sent to the CPU 12 and displayed as measured values on the display 18.

In addition, when the input voltage ■ applied to terminals Pi and P2 and the frequency of input 1st stream i are about 100 KHz (high frequency),
The operator instructs the CPt112 to send a tarokk signal with a high frequency (short period>fh).In response, the CPI
I12 applies the signal $2 to the selector 13, and the tarok signal of frequency fh output from the clock generator 75 is selected by the selector 13 and applied to the two AD conversion means. The subsequent operation is the same as in the case of the sampling frequency fn, so its explanation will be omitted.

If measurement is to be started again after the measurement is completed with N pieces of data stored in the memory, a reset signal (not shown) is applied from the CPt 112 to the address counters 8 and 9. Then, the contents are sent to the setting signal SA again.

The same operation as described above is repeated using the numerical value N specified by SB.

Furthermore, when the frequency of the signal under test applied to the terminals Pi and P2 is low, the operator instructs a clock signal of low frequency f[. In this case, unlike the above case, the sampled instantaneous value of v,1 is directly sent to the arithmetic unit 11, and multiplication and processing of equation (1) are performed in real time. In this case, the reason why the sampling data is not stored in the memory has already been described.

In the case of this low frequency, even if sampling data is stored in memory, the amount of data is small, so it can be stored without adding memory, and as mentioned above, power values, effective values, average values, etc. Measurements can be taken.

In the above description, an example with three types of sampling frequencies has been described, but the present invention is not limited to these values.

Further, it goes without saying that the number N placed in the address counter 8.9 is a necessary and sufficient value for sampling.

Effects of the Present Invention> As described above, according to the present invention, when the signal under test has a low frequency, it is possible to select and output a tallock signal with a long period. As a result, since sampling is performed at slow intervals, the number of data stored in the memory does not increase even if sampling is performed over several cycles of the low-frequency signal under test.

Note that in this case, the measurement takes time because sampling is performed over several periods of the low-frequency signal under test, but even if sampling is performed using a short-cycle clock signal (for example, clock frequency fh), accurate measurement values can be obtained. For this purpose, it is necessary to accumulate sampling data over several cycles of the signal under test.

Therefore, the present invention does not particularly increase the measurement time.

In this way, low-frequency signals under test can be measured without increasing memory capacity, making it possible to make products smaller and lower in price.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram showing an example of the configuration of a sampling type measuring instrument according to the present invention, and FIGS. 2 and 3 are diagrams explaining the operation of the exponential averaging method. 1.2...S/8 circuit, 3.4...80C16,7
...Memory, 8,9...Address counter, 11.
...Arithmetic unit, 12...CPU, 13...Selector, 15-17...Clock generator.

Claims (1)

[Claims] A device for sampling and measuring a signal under test includes means (12, 13) capable of selecting and outputting clock signals of different periods, and a signal under test in synchronization with the clock signal applied by this means. AD conversion means (1,
3), a sampling type measurement comprising a memory for storing output data of the AD conversion means, and an arithmetic unit for reading out the data stored in the memory and adding an exponential averaging operation to calculate the value of the signal under measurement. vessel.