JPS61230521A - Calibration method of integration and converter and integration ad converter using said calibration method - Google Patents

Abstract

PURPOSE:To eliminate an error generated due to response delay in a voltage comparator and to apply correct AD conversion at all times by providing a current correction means to the 2nd and 3rd reference current source, giving a correction data to them for calibration. CONSTITUTION:The 2nd reference current source 2 and the 3rd reference current source 3 are provided with DA converters 17, 18 constituting the current correction means, correction currents i4, i5 are given by giving a correction data respectively to the DA converters 17, 18 and the current i4, i5 are set so as to satisfy the relation of current ratio i1/(i1+i4)= specified value A and (i2+i4)/(i3+i5)= specified value B with respect to the actual current ratios i1/i2 and i2/i3. Then an error generated due to the response delay of a voltage comparator 11 is eliminated by applying the calibration in this way at the start of use and the correct calibration is applied automatically so as to apply correct AD conversion at all times.

Description

DETAILED DESCRIPTION OF THE INVENTION "Field of Industrial Application" The present invention relates to, for example, voltage and current measurements, etc.; improvements in integral type AD converters used;

"Prior art" An integral type AD converter first integrates the signal under test (voltage signal/current signal) for a certain period of time, and then supplies a reference current with the opposite polarity to the signal under test from a reference current source in the direction in which the integrated voltage returns to zero. Then, perform the second integral. This is an AD conversion method in which a clock pulse having the same known wave number is measured by a counter L during the second integration, and this measured value is obtained as a digital value of the signal under test.

C after the completion of the second integration for the purpose of improving the accuracy of AD conversion.
Multi-slope type A that performs the third and fourth integrals
D converters are in practical use.

Conventional multi-slope type AD using Figures 3 to 5
The structure and operation of the converter C: will be explained.

In this example, there are three reference currents [1, 2, and 3. The case of a multi-slope AD converter that performs up to 4 integrations is shown.

In the figure, 4 indicates an analog integrator. This analog integrator 4
An input side C switching circuit 5 is provided. The switching circuit 5 converts the voltage to be measured vx into a current signal ix through the resistor 13 and supplies it to the analog integrator 4c, a state where the reference current 11 of the first reference current source 1 is supplied to the analog integrator 4c, and a state where the current signal ix is supplied to the analog integrator 4c. The state of the reference current igyz' of the second reference current source 2 and the state given to the analog integrator 4, and the reference current js of the third reference current source 3
t' Analog integrator 4 (-giving state (; performs switching operation.

The specific structure of the switching circuit 5 is general (-analog switch elements sw, sW using field effect transistors).
2. sw3. sw4 is used. These switch elements SW1 to SW4 are turned on and off by a switch drive circuit 5AC. The switch drive circuit 5A receives commands from a controller 6 configured by, for example, a microcomputer (
: Controlled accordingly.

Reference numeral 7 indicates a ROMY that stores a program for operating the controller 6 in a predetermined order. Reference numeral 8 indicates a RAM that temporarily stores AD conversion values and the like each time. 9 indicates a clock pulse source.

A voltage comparator 11 is provided on the output side C2 of the analog integrator 4, and the voltage comparator 11 performs an operation of detecting the state in which the integrated voltage V of the analog integrator 4 passes through the zero point. Reference numeral 12 indicates a counter for counting the clock pulses Pc of the clock pulse source 9 and obtaining an AD conversion value.

(Operation) In this circuit structure, the switching circuit 5 first changes the voltage to be measured vx
Select analog integrator 4゛C through resistor 13:
A current IX is applied from the voltage source to be measured vx, and this current ix is integrated for a fixed time T1 as shown in FIG. 5 turns off the switch element SWl, controls the switch element SW2 to turn on, applies a voltage -vl of opposite polarity to the voltage to be measured vx through the resistor 14, and applies the first reference current 11 to the analog integrator 4; The second integration period M2 begins. When the second integration period M2 starts, the counter 12 receives the clock pulse PC from the clock pulse source 9.
Start counting. When the integrated voltage V of the analog integrator 4 passes through the zero point, the voltage comparator 11 detects this, and immediately after this detection, the first clock pulse Pc1 (FIG. 4) C:
The second integral operation is stopped in synchronization. Therefore, the integrated voltage V of the C analog integrator 4 exceeds the voltage C of opposite polarity by a small amount from the zero point, as shown in FIG.

In this state, the switching circuit 5 turns off the switch element SW2 and turns on the switch element SWs. By this switching operation, the analog integrator 4 is supplied with a reference current 12 having a polarity opposite to that of the first reference current 11 from the second reference current source 2, and a third integration period M81m is entered. In the third integration period M8, the integrated voltage of the analog integrator 4 changes again in the opposite direction C. When the integrated voltage V crosses the zero point, the third integration is completed in synchronization with the next clock pulse Pc2C. When the third integration is completed, the switching circuit 5 turns off the switch element SWs.
;Then, the switch element 5W4V is controlled to be on. This switching (-Therefore, the third reference current i3 is given to the analog integrator 4 from the third reference current source 3. By providing the third reference current i3, the integrated voltage of the analog integrator 4 changes toward the zero point. Then, the fourth integral ends at the zero cross point.

The ratio of the first reference current it, the second reference current 12, and the third reference current i3 is set as, for example, 1:1/10:1/1D, respectively, and in the second integration period M2, the upper bit of the counter 12 is Input terminal C: inputs the clock pulse PC to count, and in the third integration period M8, the input terminal of the middle digit of the counter (- gives the clock pulse Pc, and in the third integration period M4, the input terminal of the lower digit bit) Applying a clock pulse Pc to make it count; the voltage to be measured vx
It is possible to convert the value of <A/D with resolution down to the lower digits.

"Problems to be Solved by the Invention" Conventional multi-slope AD converter C shown in FIG.
In this case, each current value of the first, second, and third reference current sources 1°2.3 is accurate (for example, 1:1/10:1/1D).
; If the ratio is set, the digital values obtained in the first, second, and third integration periods M2 p MS + M4 have linearity C; it is possible to obtain an AD conversion value with excellent accuracy.

By the way, if the current ratio of the first reference current source 1, the second reference current source 2, and the third reference current source 3 deviates from the initial value, the second
Digital value obtained in integration period M2 and second integration period M
The digital value C2 obtained in step 8 and the digital value C2 obtained in the third integration period M4 contain mutual errors, resulting in a problem of poor accuracy.

If the values t of the currents il, i, .multidot.18 of the first reference current source 1, second reference current source 2, and third reference current source 3 can be directly measured, calculate the ratio and initial value C; Can be fixed. However, a normal user can directly measure each current value of the reference current sources 1, 2.3, and determine whether the ratio matches the initial setting value (d), and if they do not match, the reference current It is difficult to perform operations such as modifying each current value i1, i2, i3 of the source.

For this reason, the present applicant integrates the first reference current 11 of the first reference current source 1 for a certain period of time until the integrated voltage of the analog integrator 4 obtained as a result of this first integration returns to zero (; 2
A second integration is performed on the reference current 12 of the reference current source 2, and the ratio of the reference currents 11 and 12 is calculated from the ratio of the first integration time and the second integration time, and the AD conversion value is corrected using the value of the ratio. (−
We are proposing an integral type AD converter.

According to an integral type AD converter adopting this method, even if the current ratio of the reference current source deviates from its initial state, correction is automatically performed and correct AD conversion can always be performed.

However, this method has the following disadvantages. In other words, the reference current 11 of the first reference current source l is first integrated,
When the reference current j2Y of the second reference current source 2 is second integrated,
In order to find the current ratio by the ratio of the first integration time and the second integration time, the end of the second integration time must be the point CL when the integrated voltage V of the analog integrator 4 returns to zero (Y). In the calibration mode, as shown in FIG.
The ratio TV-(: Therefore, the first reference current 11
and the second reference current amount 2, the ratio i, /i2 is calculated. In other words, since the charge equilibrium condition is 1tTa i2'rb=o, ii/i 2 ''Ttly/1';1,
il/i2 can be measured from the time Ta and the count value C of the Tb\ counter 12; therefore, knowing C;.

However, since the voltage comparator 11 that detects that the integrated voltage of the analog integrator 4 has returned to zero C has a relatively slow response speed, there is a problem that the delay time is included as an error. In other words, the response delay time of the voltage comparator 11 is T. Then, the second integral time becomes CTb+T(2 lines C in FIG. 5), and the time Tc becomes an error.

An object of the present invention is to provide an integral type AD conversion method and an integral type AD converter that can eliminate errors caused by the delay time Tc of the voltage comparator 11 and also can automatically correct C2. It is something to do.

[Means for Solving the Problems] In the present invention, the current of the first reference current source is integrated at least twice at different times, and the current of the second reference current source (: therefore, the second Perform the integration twice, and calculate the difference between the obtained time values by performing the second integration twice (; Therefore, calculate the time value CL to obtain the time value with the delay time Tc of the voltage comparator removed, and calculate the time value of this difference. The present invention proposes a calibration method for determining the true ratio of the first reference current and the second reference current from the first integration time, and further provides an integral type AD converter using this calibration method.

According to this invention, it is possible to obtain an accurate current ratio that eliminates the response delay of the voltage comparator and the error that occurs, and highly accurate calibration can be performed.

"Example" Two calibration methods C, which are the first invention of this application, will be explained using FIG. The method for calibrating an integral type AD converter according to this invention is to calibrate the first reference current i1c of the first reference current source 1 twice at different times as shown in Figure 1.
Perform the integration. In the figure, v1 shows the integrated voltage waveform of the analog integrator 11 during the first α-Q integration operation, and v2 shows the integrated voltage waveform of the analog integrator 11 during the second time. In this example, Ta
The case where □=2 Ta□ is shown.

The second reference current source 2 C for the first integral C for the first calibration.
Therefore, a second integration for calibration is performed. -th first integration time T'at and second integration time Tbx+Tct.

Second first integral! On the other hand, the second reference current source 2: Therefore, the second integration is performed. This first integral time Tal! , the second integration time is Tbg +Tcg.

By performing two calibration AD conversion operations, the second integration time (Tbt ” Tct ) of the first time and the second integration time (Tct ” Tct ) of the second time are
Tbg + Tc2) is obtained. TCt”F Tc
If g, (Tbz +Tcs) (Tb1+Tc
t). Here, the relationship between the first integration time is 2 T31
= Tag, so the relationship between the second integration time is also 2 Tb
t=Tb2.

Therefore, (Tbg +'ro1) (Tbt +TC1)=
Tbt・・・・・・・・・(1) The time value Tbt obtained from the first equation is the voltage comparator 1
This value does not include the delay time Tc of 1.

From the relationship of charge balance, 'IT'ai + '2・TI)1...
・・・・・・・・・・・・(2) Therefore, from this second equation == Kabuto ・・・・・・・・・・・・(3) I
2T'at is obtained. In the third equation (-, the first integral time Ta□ is a predetermined constant time, so it is a wise value, and the two
AD conversion value (Tbi +
By determining Tb1 from the first equation using Tc1) and (Tbg + TC?), the current ratio 11/'2' can be determined.

The current ratio i, /i is set in advance to A (A is, for example, 10.1
(Select a value such as D ≦;).
The deviation from the specified value A can be determined. Therefore, it is possible to correct the current ratio i1/l2 to a predetermined value A; for example, by correcting the current 12 of the second reference current source 2, it is possible to perform a correct AD conversion operation. can.

Similarly, the relationship between the second reference current source 2 and the third reference current source 3 can be determined by calculating the current ratio i2/is. This current ratio i2/i3 is defined as B in advance. ; Therefore, the actual current ratio from two AD conversion operations is 127r, zr
: can be determined, and the deviation from the specified value B can be determined from the result.

Therefore, according to the calibration method C2 according to the present invention C, accurate current ratios i, /i2 and i2/i8 can be obtained without including errors due to response delay of the voltage comparator 11, and correct calibration can be performed. be able to.

``Embodiment of the second invention'' In the second invention of this application, as shown in FIG. 18 is provided, and correction data is provided to the DA converters 17 and 18C, respectively.
'4)! =A and (i11+”4)/(is+1s)
The three units are configured to set ≦:i4 and i so that =B is satisfied.

In addition, in the calibration mode, the current 11 of the first reference current source 1 is changed to C;
Integrating the current 12 of the second reference current source 2 twice at different predetermined times, and integrating the current 12 of the second reference current source 2 until each integrated voltage returns to zero,
A means for determining the difference between these two time values, and integrating the time value of this difference and the first reference current. A controller consisting of a microcomputer is used to determine whether the time ratio matches a predetermined value and to determine the amount of deviation from the predetermined value if the time ratio does not match the predetermined value. 6 (: Therefore, it is constituted.

In other words, the AD conversion operation for calibration is performed twice (by determining the actual current ratios i1/i2 and 12/'s, and calculating the deviation from the specified values A and B by calculating the actual current ratio ζ). A value for correcting the deviation is calculated by the controller 6 (; and the calculated value '&DA converters 17 and 18L;
The A conversion output is added to the currents 12 and 18 of the second reference current source 2 and the third reference current source 3 through resistors 1, 9 and 20 for calibration.

Like this (- for example by configuring C
; Operate in calibration mode, and calibrate by applying calibration values to DA converters 17 and 18C in this calibration mode (-).
Therefore, correct AD conversion can always be performed.

"Effects of the Invention" As described above, according to the first invention of this application (-), correct calibration can be performed by eliminating errors caused by the response delay of the voltage comparator 11.

Further, according to the second invention C2 of this application, it is possible to automatically perform correct calibration that eliminates the error caused by the response delay (-) of the voltage comparator 11. Therefore, always C: Correct AD
An AD converter that performs a conversion operation can be provided.

[Brief explanation of drawings]

FIG. 1 is a waveform diagram for explaining in detail the first invention of this application, FIG. 2 is a block diagram for explaining in detail the second invention of this application, and FIG. 3 is for explaining the prior art. FIGS. 4 and 5 are waveform diagrams for explaining the operation of a conventional integral type AD converter. 1: First reference current source, 2: Second reference current source, 3: Third reference current source, 4: Analog integrator, 5: Switching circuit, 5A:
Switch drive circuit, 6: Controller, 7: ROM, 8: RA
M, 9: Clock pulse source, 11: Voltage comparator, 12:
Kakunta, 17, 18: D constituting current correction means
A converter. Patent applicant: Takeda Riken Kogyo Co., Ltd. Representative: Takuil Kusano 5 times

Claims (2)

[Claims] (1) A, having at least a first reference current source and a second reference current source having a polarity different from that of the first reference current source and having a current value set to a value smaller than the current value of the first reference current source; The first and second reference currents output from these two reference current sources perform second and third integration on the integral value of the signal under test, and the second and third integral time values are used to In an integral type AD converter configured to convert the amount of the signal under measurement into a digital value, B. In the calibration mode, the current of the first reference current source is integrated twice at different predetermined times, and the integrated voltages are respectively Integrate the current of the second reference current source until it returns to zero, find the two times during which the reference current of this second reference current source is integrated, and calculate the difference between these two time values without including the error. Determine the true AD conversion value, calculate the ratio of the first reference current value and the second reference current value from the ratio of this true AD conversion value and the time during which the first reference current is integrated, and calculate this current value. A method for calibrating an integral type AD converter, in which correct AD conversion can be performed by correcting the values of the first reference current and the second reference current so that the ratio of the first reference current and the second reference current becomes a predetermined value. (2) A, having at least a first reference current source and a second reference current source having a polarity different from that of the first reference current source and having a current value set to a value smaller than the current value of the first reference current source; The first and second reference currents output from these two reference current sources perform second and third integration on the integral value of the signal under test, and the second and third integral time values are used to In an integral type AD converter configured to convert the amount of the signal under measurement into a digital value, B. In the calibration mode, the current of the first reference current source is integrated twice at different predetermined times, and the integrated voltages are respectively means for integrating the current of the second reference current source until it returns to zero, determining two times during which the reference current of the second reference current source is integrated, and determining a difference between these two time values; means for determining the ratio between the time value of this difference and the time during which the first reference current is integrated; D. determining whether the time ratio matches a predetermined value; E. current correction means for correcting the current value of the second reference current source based on the deviation amount and calibrating the time ratio so that it becomes a specified value. An integral type AD converter consisting of: