JPH0524445B2 - - Google Patents

Description

[Detailed description of the invention]

(Industrial Application Field) The present invention relates to a method and apparatus for interpolating encoder read signals. In particular, the invention relates to an interpolation method and device that allows high-resolution measurements on the basis of at least two periodic waveform signals obtained from an encoder. (Prior art) An encoder for automatic reading in a length or angle measurement instrument has a phase difference from the detection section.
Generates two signals close to sine waves that differ by 90 degrees.
Then, in the arithmetic processing section of the encoder, these two signals are combined to form signals of various phases, and the amount of displacement from the reference position is obtained by counting the points at which these signals become zero. In this method, measurement accuracy, or resolution, can be improved by increasing the number of signals, but in reality, it is not possible to greatly increase the resolution due to offset voltage or drift of the zero value detection circuit. . (Object of the Invention) An object of the present invention is to provide a method and apparatus for interpolating an encoder read signal, which is free from the limitations of conventional methods and can dramatically improve resolution. (Structure of the Invention) The present invention provides an encoder read signal interpolation method for detecting a moving position using a plurality of periodic waveform signals having mutually different phases obtained from an encoder detection unit in response to a change in position. one period of the periodic waveform signal is selected as the reference signal, one period of the reference signal is divided into a plurality of interpolation intervals by the points where each value of the periodic waveform signal becomes zero, and each interpolation interval The number of interpolation intervals from the start point of one cycle of the reference signal to the interpolation interval including the measurement point is determined by pairing two signals that should be in the same direction at and detecting the positive/negative states of the two paired signals. N _P
The values X _P and Y _P of the two signals changing in the same direction at the measurement point, the amount α indicated by the amount of change in the position of the encoder, and the starting point of one cycle of the reference signal are determined. An encoder read signal interpolation method characterized in that the position of the measurement point is calculated from the equation B = [{X _P / (X _P − Y _P )} + N _P ]・α. be. The present invention also provides an encoder read signal interpolation device for detecting a movement position based on two periodically varying encoder read signals whose phases are different from each other by 90°. 180° with respect to the reference signal.
a signal forming means for obtaining signals with different phases; and one of the plurality of interpolation intervals including the measurement point, which is obtained by dividing one period of the signal by points at which each of the signals becomes zero, one of which is positive. subtracting means for generating a subtraction signal by subtracting the value of one of two signals, one of which is negative and changing in substantially the same direction, from the value of the other signal; An interpolation device for an encoder read signal, comprising a division means for generating a division signal divided by a subtraction signal, and a multiplication means for multiplying the division signal by a predetermined coefficient to obtain an interpolation result. (Effect of the invention) In the present invention, in the interpolation interval divided by the zero points of a plurality of signals, the position of the measurement point is
Since it is calculated using a proportional calculation formula based on the values of two signals selected in the interpolation interval, the resolution can be greatly improved compared to the conventional method that counts the zero points of the signal. can. (Description of Examples) Examples of the present invention will be described below with reference to the drawings. <Principle of interpolation> Upper-level measurement The detection section of the encoder generates two signals with a waveform similar to a sine wave with a phase difference of 90 degrees, as in conventional devices. For one of the signal waveforms, upper level measurement is performed by counting the number of waveforms up to the measurement point. Intermediate measurement Divide one cycle that includes a measurement point into multiple interpolation intervals, and detect the interpolation interval that includes the measurement point. For this purpose, two signal waveforms generated by the detection section of the encoder and an inverted signal of one of these signal waveforms are used. FIG. 1 shows the signal waveform of the encoder detection section and its inverted signal approximated to a triangular wave. In the figure, the output signals of the encoder detection section are indicated by a and b, and the inverted signal of signal a is indicated by . Zero crossing points of signals a and b a ₀ , b ₀ , a ₁ , b ₁ ,
A ₂ divides one cycle including the measurement points into four interpolation intervals 0, 1, 2, and 3. As described later,
If more virtual signals are formed using the signals from the encoder detection section, the interpolation interval can be divided into smaller sections. Median measurement is performed by detecting in which interpolation interval the measurement point p is included among the interpolation intervals divided in this manner. For this purpose, we sequentially select the two signals that determine each interpolation interval as a pair (hereinafter referred to as scanning), obtain the number of scans n until one becomes positive and the other becomes negative, and measure the measurement point p.
The number N _P of the interpolation interval including the measurement point p (meaning the number of interpolation intervals from the zero-crossing point a ₀ of the signal a to the interpolation interval including the measurement point p) is determined. For example, the first
In the example shown, the signal combinations and number of scans are obtained as shown in the table below. X signal Y signal Number of scans a b 0 b 1 If there are three signal combinations, set the number of scans for the third signal combination to be 2, and if there are four signal combinations, set the number of scans to the third signal combination. The number of scans n of the fourth signal combination is assumed to be 2 and 3, respectively. The number of scans n obtained in this way
From this, an interpolation interval N _P including the measurement point p is calculated.
In other words, among the pair of signals that determine the interpolation interval,
The signal with a lead in phase is the X signal, and the signal with a delay is the Y signal.If the X signal is positive and the Y signal is negative, N _P =n, and if the X signal is negative, Y
If the signal is positive, N _P = n + N _M /2 (however,
(N _M is the number of interpolation intervals within one period), the interpolation interval number N _P can be obtained. For example, in the case of measurement point p in FIG. 1, when the X signal is selected as a and the Y signal is selected as b, that is, when the number of scans n=0, a positive X signal and a negative Y signal are obtained. Therefore, N _P
=n=0. Further, when the measurement point is at position p', when the X signal is selected as b and the Y signal is selected, that is, when the number of scans is n=1, a negative signal is obtained for the X signal and a positive signal is obtained for the Y signal. Therefore, the interpolation interval number N _P
becomes N _P =n+4/2=3. However, here N _M = 4. Lower measurement Among the interpolation intervals divided in the intermediate measurement,
By knowing the position of a measurement point in an interpolation interval in which the measurement point is included, the position of the measurement point within one period of interpolation is detected with high resolution. If the value of the size of one interpolation interval expressed by the rotation angle or movement amount of the encoder is α, then the starting point of the interpolation interval including the measurement point is from the starting point a ₀ of one period in which interpolation is performed to αN _P It is located far away. Then, in the interpolation interval including the measurement points,
The two signals, namely the X signal and the Y signal, selected as described in the section on intermediate measurement, are almost parallel.
Therefore, the rotation angle or movement amount l from the starting point of the interpolation section to the measurement point can be expressed by the following equation. l=X _P /X _P −Y _P α (1) Here, X _P and Y _P are the values at the measurement point p of the X signal and the Y signal. Therefore, the rotation angle or the amount of movement β from the starting point a ₀ to the measurement point p in one period in which interpolation is performed is expressed by the following equation. β=(X _P /X _P −Y _P +N _P ) α ...(2) Virtual signal In an actual encoder, an ideal triangular wave cannot be obtained due to light diffraction phenomena and the influence of light other than parallel light. The greater these influences, the closer the waveform becomes to a sine wave, and as a result, the linearity breaks down except near the zero-crossing point.
It is desirable to combine the two waveforms to create a virtual signal, and use only those signals near the zero-crossing point of these signals. Generally, two signals that differ by 90 degrees are output from the encoder detection unit, and are expressed as a=A·sin θ+O _a b=B·sin (θ−π/2)+O _b (3). Here A>O, B>O, and O _a
and O _b indicate the zero level offset values of signals a and b, and here it is assumed that O _a =O _b =O.
The signal i obtained by combining the two signals is: i=aK _a +bK _b =K _a A・sin θ+K _b B・sin (θ−π/2
)……(4). Here, K _a and K _b are two signals a,
This is a coefficient that determines the ratio when adding b. Equation (4) is expressed as follows. By appropriately setting K _a and K _b from equation (5),
It is understood that it is possible to obtain any sine wave. Let A=B, and K _a and K _b

When set to [Formula], from formula (5) i=Asin(θ−π/4)……(6) is obtained, and

【formula】

[Formula] i'=Asin(θ-3/4π)...(7) is obtained. These waveforms are shown in FIG. By using such a virtual signal, one period of the signal can be divided into even smaller interpolation intervals, and the accuracy of interpolation can be improved by using only the signal portion with high linear accuracy. <Correction of amplitude and zero level> As already mentioned, interpolation is performed by detecting two signal levels, and it is important that the zero level of the two signals and the relationship between their amplitudes are always constant. be. However, in the two signals obtained from the actual encoder detection section, there are changes in the zero level due to changes in light intensity of the light emitting diode and changes in temperature, that is, drift, and changes in amplitude due to changes in the distance between the movable plate and the stator plate. It is usually accompanied by zero level changes and amplitude changes due to pattern non-uniformity. Therefore, if linear interpolation reading is performed using the signal obtained from the encoder detection section, there is a possibility that highly accurate results cannot be obtained due to the effects of these zero level changes and amplitude changes. Therefore, in order to obtain highly accurate measurement results, it is necessary to correct the signal amplitude and zero level. The two signals obtained from the encoder detection section are
Both have waveforms close to sine waves, and the phases are mutually
Although they differ by 90 degrees, if the two signals have this relationship, when one signal has a zero value, the other signal will take the maximum or minimum value, or extreme value, and each signal will have a zero value. The maximum rate of change is shown near the value, and the minimum rate of change is shown near the extreme value. If we pay attention to such points, it becomes possible to obtain the extreme value of each signal by using each other's signals. To explain in more detail, for example in Fig. 2,
Signals a and b both have sinusoidal waveforms, and signal b is delayed in phase by 90° with respect to signal a. and,
At the zero-crossing point _a0 of the signal a, the signal b has a minimum value, and at the zero-crossing point _a1 , the signal b has a maximum value. Similarly, at the zero crossing points b ₀ and b ₁ of signal b, signal a takes the maximum value and minimum value, respectively. Therefore, the zero-crossing point of one signal may be detected, and the value of the other signal at that time may be considered as the extreme value of that signal. In this case, the influence of the zero level fluctuation of the signal is involved, but the rate of change of the signal is at its minimum near the extreme value, so the error is extremely small. The extreme values of signals a and b shown in equation (3) are
Let a _nax , a _nio , b _nax , b _nio be the amplitude A of signal a
and zero level O _a are expressed as A=a _nax −a _nio /2 ...(8) O _a = a _nax + a _nio /2 ...(9) and amplitude B of signal b and zero level O _b
is expressed as: B = b _nax - b _nio /2 ... (10) O _b = b _nax + b _nio /2 ... (11). Therefore, the zero level correction is
This can be done by subtracting O _a from signal a and O _b from signal b. To correct the amplitude balance of the two signals, one of the two signals is corrected to zero level from the ratio of A and B. That is, the correction is performed by multiplying the signal (b-O _b ) by A/B or by multiplying the signal (a-O _a ) by B/A. The equation of this correction relationship is Or becomes. The zero level of the two signals a′ and b′ corrected in this way becomes true zero, and the amplitudes of both A and B
It will be one of the following. Therefore, two signals with zero level changes and amplitude balance changes are
After correction, there is no change in the zero level, and the relationship between the amplitudes of both signals is always kept constant. <Specific Configuration> Overall Configuration FIG. 3 shows an example of a control circuit embodying the present invention. The illustrated control circuit includes the measuring section 1
0, a correction section 20, a virtual signal generation section 30, and an interpolation section 40. Measuring unit 10 The measuring unit 10 is a comparator CMP that receives signal a as input.
1 and a comparator CMP2 that inputs signal b.
As shown in FIG. 5, these comparators CMP1 and CMP2 convert their respective input signals into rectangular waves having a rising or falling edge at the zero crossing point of the signal. The outputs of the comparators CMP1 and CMP2 are given to the direction discriminator RL. The direction discriminator RL discriminates whether the moving direction or rotation direction of the encoder is rightward or leftward from the phase relationship of the rectangular wave signals from comparators CMP1 and CMP2. For example, if it is rightward, If the direction is to the left, an output pulse is generated to the output terminal DN. The output of the direction discriminator RL is the comparator
It is in the form of a pulse corresponding to one of the rectangular waves that are the outputs of CMP1 and CMP2, for example, the rectangular wave whose phase is advanced, and this output is the up-down counter.
The pulses are input to the UDC and the number of pulses is counted. This count allows the amount of movement of the encoder to be measured. The output of the up-down counter UDC is given to the control unit CTL, and the control unit CTL converts the output of the counter UDC into a rotation angle or a direct movement amount, and combines this with the calculation result by interpolation reading described later.
Display on DSP. Correction Unit 20 The correction unit 20 includes sample-and-hold circuits SH1 and SH2 that receive the signal a, and sample-and-hold circuits SH3 and SH4 that receive the signal b. Furthermore, a pulse generator POS1 that receives the output of the comparator CMP1 of the measuring section 10 and a POS2 that receives the output of the comparator CMP2 are provided. This pulse generator has an exclusive OR circuit 21 as shown in FIG. A comparator output is connected to via a resistor 22, and the connection between this resistor 22 and the input terminal of the exclusive OR circuit 21 is grounded via a capacitor 23. These pulse generators POS1, POS2 are connected to comparators CMP1,
Generates a pulse output that rises at the rising and falling points of the CMP2 output. That is, the pulse generator POS1 generates a pulse at the zero-crossing point of the signal a, and the pulse generator POS2 generates a pulse at the zero-crossing point of the signal b. The output of the pulse generator POS2 is connected to one input terminal of each of two AND gates A1 and A2. The other input terminal of AND gate A1 is a comparator
The other input terminal of AND gate A2 is connected to the output of CMP1, and the other input terminal of AND gate A2 is connected to the output of CMP1.
Connected to the output of CMP1. Therefore, as shown in FIG. 5, the AND gate A1 outputs an output when the signal b is at the zero cross point, that is, when the signal a has an extreme value, and when the signal a takes a positive value. occurs. In other words, AND gate A1 produces an output when signal a takes its maximum value. Similarly, AND gate A2 produces an output when signal b takes its minimum value (FIG. 5).
The outputs of these AND gates A1 and A2 are input to sample and hold circuits SH1 and SH2, respectively. Similarly, the output of pulse generator POS1 is
It is connected to one input terminal of each of two AND gates A3 and A4, and the other input terminal of AND gate A3 is connected directly to the output terminal of AND gate A4 via inverter I2. are connected to each. And gate A3
and gate A4 generates an output when signal b is at its minimum value (5th
figure). And the outputs of AND gates A3 and A4 are
The signals are input to sample and hold circuits SH3 and SH4, respectively (FIG. 5). The sample and hold circuit SH1 samples the signal a when receiving a pulse from the AND gate A1, and holds the sampled signal as it is during a period when no pulse is input. Therefore, the sample and hold circuit SH1 outputs approximately the maximum value of the signal a, that is, the signal a _nax . Similarly, the sample-and-hold circuit SH2 outputs a signal that is approximately the minimum value of the signal a.
a _nio , sample and hold circuits SH3 and SH4 detect b _nax , which is approximately the maximum value and approximately the minimum value of signal b,
b Output each _nio . The maximum and minimum values of a signal sampled in this way differ from the true maximum and minimum values, as they are affected by zero-level fluctuations in the signal, but as mentioned earlier, near the extreme values Since the rate of change in signal level is very small, the error is minute. Therefore, in the following discussion, this sampled signal will be treated as a maximum or minimum value. The outputs of the sample and hold circuits SH1 and SH2 are respectively connected to two input terminals of an adder AD1, and the output of the adder AD1, which is the sum of both, is given to the negative input terminal of a subtracter SB2. subtractor
A signal a is input to the positive input terminal of SB2, and the subtracter SB2 generates an output value obtained by subtracting the output of the adder AD1 from the signal a. Furthermore, the outputs of the sample and hold circuits SH1 and SH2 are connected to the positive and negative sides and input terminals of the subtracter SB1. Similarly, the outputs of sample and hold circuits SH3 and SH4 are output to the adder.
Adder that is input to AD2 and is the added value of both
The output of AD2 is given to the negative input terminal of subtracter SB4. The signal b is input to the positive input terminal of the subtracter SB4, and the output of the value obtained by subtracting the output of the adder AD4 from the signal b is obtained by the subtracter SB4. Further, the outputs of the sample and hold circuits SH3 and SH4 are respectively connected to the positive side and negative side input terminals of the subtracter SB3. Adders AD1, AD2 and subtracters SB1, SB3
is composed of operational amplifiers, and if all their gains are set to 1/2, the output of each is as follows. Output of adder AD1 = (a _nax + a _nio )/2 Output of subtracter SB1 = (a _nax − _{a nio} )/2 Output of adder AD2 = (b _nax + b _nio )/2 Output of subtracter SB3 = ( b _nax −b _nio )/2 That is, the outputs of adders AD1 and AD2 represent the zero levels O _a and O _{b of signals a and b} , respectively,
The outputs of subtracters SB1 and SB3 are signals a and
It represents the amplitudes A and B of b. The subtracter SB2 outputs a signal obtained by subtracting the zero level O _a from the signal a, that is, a signal subjected to zero level correction. Similarly, the subtracter SB4 outputs a signal obtained by performing zero level correction on the signal b. subtractor
The outputs of SB1 and SB3 are input to the divider DV1,
As its output, a signal with the ratio A/B of amplitudes A and B is obtained. The output of the subtracter SB4 is given to the multiplier MP1, and the output of the divider DV1 is also given to the multiplier MP1. Generates a signal corresponding to the product of the output signals. i.e. the multiplier
The output of MP1 is a signal obtained by adding zero level correction and amplitude correction to signal b. Therefore, the following signals a' and b' are output from the subtracter SB2 and the multiplier MP1, respectively. a'=Asin θ b'=Asin (θ-π/2) Signals a and b from the encoder detection section and signals a' and b' after correction are shown in FIG. 5, respectively. Virtual signal generation section 30 The virtual signal generation section 30 is a subtracter of the correction section 20.
It has an adder AD3 whose inputs are the output of SB2 and the output of multiplier MP1, and a subtracter SB5 whose positive input terminal is supplied with the output of multiplier MP1, and whose negative input terminal is supplied with the output of subtractor SB2. .
The adder AD3 and the subtracter SB5 have a gain set to 1/√2, so their outputs are as follows. Adder AB3 c' = Asin (θ-π/4) Subtractor SB5 d' = Asin (θ-3π/4) Interpolation section 40 The interpolation section 40 has a signal switch MPX consisting of an analog switch. , the signal a' from the subtracter SB2 and the signal from the multiplier MP1 are sent to this signal switch MPX.
b', signal c' from adder AD3 and subtracter SB5
The signals d' from the respective terminals are inputted. Furthermore, the interpolation unit 40 is provided with a subtracter SB6 that inverts the signal from the subtracter SB2 to generate an inverted signal', and the output signal' of this subtracter SB6 is also connected to a signal switch.
Input to MPX. Figure 6 shows signal switching
Input signals a′, b′, c′, d′, and ′ to MPX are shown. A counter CNT is provided to control the operation of the signal switch MPX.
selects two signals from the input signals a', b', c', d', and outputs them as signals X and Y according to the output signal of the counter CNT. Counter CNT is a 2-bit binary counter, and the control unit
Reset by reset signal R from CTL,
It is also connected to count the output signal of the oscillator OSC. When performing interpolation reading, a reset signal R is first generated from the control unit CTL, and then this is released. Counter CNT is the control unit
It starts operating when the reset signal R is released from CTL, and counts using the signal from the oscillator OSC as the clock. The resulting count value corresponds to the number of scans described above. In the case of the signals shown in FIG. 6, the signal switch MPX is configured so that the relationship between the number of scans n and the signals X and Y is as shown in the following table. n X Y 0 a'c' 1 c'b' 2 b'd' 3 d'' When counter CNT starts counting, as the counting progresses, the contents of signals It changes sequentially. Signal Y goes to comparator CMP3, signal X goes to comparator
are input to CMP4, comparators CMP3,
The output of CMP4 is given to two input terminals of exclusive OR circuit EO. comparator
CMP3 and CMP4 generate a high level output when the input is positive and a low level output when the input is negative, and the exclusive OR circuit EO generates a high level output when one of the outputs of the comparators CMP3 and CMP4 is high and the other is low. Occur. Exclusive OR circuit EO
The high level signal is inputted as the selection scan end signal E to the counter CNT and the control unit CTL. That is, the counter CNT starts counting when the reset signal R of the control unit CTL is released, and as the count progresses, the signal switch MPX sequentially changes the combination of signals X and Y as shown in the table above. Switch and output. Then, when one of the signals X and Y becomes positive and the other becomes negative, the exclusive OR circuit EO
A selected scan end signal E is generated, and the counter CNT stops counting. The count value of the counter CNT in this state is given to the control unit CTL. Further, the output of the comparator CMP3 is also input to the control unit CTL. The counter CNT counts one by one for each interpolation interval, and returns to 0 when the count reaches 3. In this example, five signals are used, and one period is divided into eight interpolation intervals.
The count of the counter CNT is configured to be repeated every half period of the signal a'. This relationship is shown in FIG. 6 as the number of scans n. Next, signal Y
Looking at the relationship between the positive and negative values, the signal Y always takes a negative value during the half cycle in which the signal a' takes a positive value, and when the signal a' takes a negative value, the signal Y takes a positive value. Therefore, the count value of counter CNT and comparator CMP3
The interpolation interval number N can be known from the output of .
For example, in the example shown in Fig. 6, the output of the counter CNT is "1" at the measurement point p, and the output of the comparator CMP is "1".
Since the output of counter CNT is low, it can be determined that the interpolation interval number N is "1", and at the measurement point p', the output of the counter CNT is also "1", but the output of the comparator
Since the output of CMP3 becomes high, it can be determined that the interpolation interval N is "5". Signals X and Y are applied to the positive and negative input terminals of subtracter SB7, respectively, and the difference between the two signals is applied to the subtracter SB7.
It is output from SB7 and given to divider DV2.
The signal X is also input to the divider DV2, which divides the signal X by the output of the subtracter SB7 and outputs a signal X/(X-Y). divider
The output of DV2 is input to the AD converter ADC, converted into a digital value, and given to the control unit CTL. Function of the control unit CTL The control unit CTL is an exclusive OR circuit EO
When a high level signal is received from the counter CNT and the output of the comparator CMP3 as described above, the number N _P of the interpolation interval including the measurement point pn is calculated, and then the A/D converter ADC Signal from X _P /
By adding (X _P −Y _P ) and multiplying that value by a coefficient α, the rotation angle or movement amount L from the starting point a ₀ of the signal a′ to the measuring point p is determined. That is, the calculation result of β={X _P /(X _P −Y _P )+ _NP }·α shown in equation (2) is obtained. Next, the control unit CTL adds the count value from the up-down counter UDC to this calculation result and displays the result on the display DSP. <Other Embodiments> FIG. 7 shows another embodiment of the present invention, and the same components as those in FIG. 3 are given the same reference numerals and their explanations will be omitted. In this example, the configuration for interpolation consists of a correction timing generating section 50, a data reading section 60, and a control section CTL'. Correction tanning generation section 50 Pulse generators POS1, POS2 and AND gate A arranged in the same manner as in the correction section 20 of the previous example
1, A2, A3, and A4. The output signals of the AND gates A1, A2, A3, and A4 are input to the OR gate OR, and are also applied to the latch circuit LT. As described above, each AND gate generates a pulse at approximately the maximum or minimum position of the signals a and b. Therefore, pulses are generated at the output terminal of the OR gate OR at almost all the maximum and minimum positions of the signals a and b. The output signal of this OR gate OR is supplied to the latch circuit LT and the control section CTL'. In the latch circuit LT, depending on the output signal of the OR gate,
It is stored which AND gate among AND gates A1, A2, A3, and A4 generates a pulse. The control unit CTL' uses the signal from the OR gate OR as the interrupt signal INT. Data reading section 60 Signal selector MPX' which receives signals a and b as input
Consists of an AD converter ADC and a storage RAM.
This storage RAM may be shared with the random access memory included in the control unit CTL'. The signal selector MPX' supplies the signal a or the signal b to the A/D converter ADC according to the signal MS from the control unit CTL'. A/D converter ADC is the control section
When the signal S from CTL' goes high, A/
Signal to start D conversion operation and give to control unit CTL'
Set AE to low level. When the A/D conversion operation is completed, the A/D converter ADC sends a signal to the control unit CTL'.
Set AE to high and send. The storage RAM is the control unit
The data from the A/D conversion ADC is written to a predetermined location using the address signal ADS from the CTL and the write signal W. Furthermore, in response to the read signal RD from the control section, the contents stored at a predetermined location are sent to the control section CTL' via the bus signal BUS. Control Unit CTL' The processing of the control unit CTL' can be roughly divided into interrupt processing and arithmetic processing. Interrupt processing When the interrupt signal INT is obtained from the OR gate, perform the following operations. (i) Temporary storage The contents currently selected by the signal selector MPX' are temporarily stored internally. (ii) Deciphering the interrupt request Depending on the contents of the latch circuit LT, whether the contents of the interrupt request are the maximum or minimum of the signal a, the maximum or the minimum of the signal b (a _nax , a _nio , b _nax , b _nio ). (iii) Selection of signals a and b If the result of decoding the interrupt request is a _nax a _nio , the signal selector MPX' is set to select signal a, and conversely, if b _nax or b _nio , the signal is selected. MPX' is set to select signal b. (iv) Writing to the storage unit First set the signal S to low, then set it to high.
Then wait for signal AE to go high. Also,
According to the deciphering of the interrupt request, a _nax , a _nio ,
Storage RAM corresponding to either b _nax or b _nio
The address is sent to the storage RAM through the address signal ADS. Then, when the signal AE becomes high, a high pulse is sent to the storage RAM using the write signal W, and the output data of the A/D converter is written into the storage RAM. Then, the signal s is set low to reset the A/D converter. (v) Return the signal selector MPX' to its original state according to the temporarily stored contents when entering the interrupt process. Furthermore, the signal s is made high to operate the A/D converter. That is, assuming that the A/D converter was operating before proceeding to the interrupt processing, this operation is performed to return it to its original state. The interrupt signal INT is generated sequentially in the order of _bnio , _anax , _bnax , and _anio . Therefore, by the above interrupt processing, a _nax , a _nio , b _nax , and
b _nio data is written. Arithmetic Processing During arithmetic processing, the following operations are repeated. However, if the interrupt signal INT occurs during calculation processing,
The process moves to interrupt processing, and returns to the beginning when the interrupt processing is completed. (i) Measurement of signal a Set the signal selector MPX' to select signal a using signal MS. Next, the signal s is changed from low to high, the A/D converter is operated, and the signal AE is waited for to go high.
When the signal AE goes high, the signals ADS and W are used to write the output data of the A/D converter (data of the signal a) to a predetermined location in the storage RAM. (ii) Measurement of signal b Set the signal selector MPX' to select signal b using signal MS. In the same manner as in "Measurement of signal a", data of signal b is written to a predetermined location in the storage section. (iii) Data transfer Send the address of a _nax to the storage RAM through the signal ADS, and set the signal RD high to send the address of a _nax to the storage RAM.
The data is transferred to the RAM of CTL′ in the control section. Similarly, the data of _anio , _bnax , _bnio , a, and b are transferred to the RAM in CTL' in the control section. (iv) Correction calculation Zero level O a of signal _a , amplitude A, and signal b
The zero level O _b and amplitude B of are expressed as (8), (9), (10),
Calculate according to (11). Next, the corrected signal data a' and b' are obtained using the following equation. a'=(a-O _a ) b'=A/B(b-O _b ) Here, O _a , A, O _b , B, a', b' are stored in the RAM in the control unit CTL' as necessary. shall
Not specified. The same applies to other data below. (v) Virtual signal creation Virtual signals c′ and d′ are obtained using the following equations. (vi) Find the inverted signal ' of the calculation signal a' in the interpolation interval N _P including the measurement point p from '=O-a'. signal
Data as X _P and Y _P are sequentially obtained from data a', b', c', d', and ' according to the following combinations, and it is checked whether one is positive and the other is negative. At the same time, the corresponding value is entered into the number of scans n. X _P Y _P n a'c' 0 c'b' 1 b'd' 2 d'' 3 When one becomes positive and the other becomes negative, this operation ends. Then, determine whether the signal Y _P is positive or negative,
N _P is determined by setting N _P =n if negative, and N _P =n+N _M /2 if positive. However, in this example, N _M =8. (vii) Interpolation calculation Measurement position p with respect to reference point a ₀ according to equation (2)
Find the rotation angle or movement amount β. (viii) The rotation angle or movement amount obtained from the up-down counter UDC and β are combined and displayed.

[Brief explanation of drawings]

Fig. 1 is a waveform diagram for explaining the principle of interpolation of the present invention, Fig. 2 is a waveform diagram showing an example of creating a virtual signal, and Fig. 3 is a processing circuit for encoder read signals to which the present invention is applied. 4 is a circuit diagram showing an example of the pulse generator used in the circuit of FIG. 3, FIG. 5 is a waveform diagram showing the amplitude and zero level correction of the encoder read signal, and FIG. 6 is a diagram showing the virtual signal. The waveform diagram shown together with the encoder read signal and the inverted signal, and FIG. 7 is a block diagram of a processing circuit according to another embodiment. 10...Measurement unit, 20...Correction unit, 30...Virtual signal generation unit, 40...Interpolation unit, CTL...Control unit, DSP...Display unit, CMP1, CMP2, CMP
3, CMP4...Comparator, POS1, POS2...Pulse generator, A1, A2, A3, A4...And gate, SB1, SB2, SB3, SB4, SB5,
SB6, SB7...Subtractor, CNT...Counter.

Claims (1)

[Scope of Claims] 1. An encoder read signal interpolation method for detecting a moving position using a plurality of periodic waveform signals having mutually different phases obtained from an encoder detection unit in response to a change in position, comprising the steps of: One of the signals is selected as a reference signal, one period of the reference signal is divided into a plurality of interpolation intervals according to the points at which each value of the periodic waveform signal becomes zero, and the same period is divided into a plurality of interpolation intervals in each interpolation interval. The number of interpolation intervals N P from the start point of one cycle of the reference signal to the interpolation interval including the measurement _point is determined by pairing two signals that should be in the same direction and detecting the positive and negative states of the two signals in the pair. The values of the two signals changing in the same direction at the measurement point X _P , Y _P
and the position of the measurement point from the start point of one cycle of the reference signal to the amount α indicated by the amount of change in the position of the encoder in the interpolation interval, Formula B = [{X _P / (X _P − Y _P )} An interpolation method for an encoder read signal, characterized in that the interpolation method is obtained by performing calculations based on +N _P ]・α. 2. In the interpolation method of item 1, an inverted signal is formed from at least one of the periodic waveform signals, and the position of the measurement point is determined by the periodic waveform signal and the inverted signal. Interpolation of the encoder read signal Method. 3. In the interpolation method of item 1 above, a virtual signal is formed by combining at least two periodic signals having different phases, and the position of the measurement point is determined using the periodic waveform signal and the virtual signal. Interpolation method. 4. In an encoder read signal interpolation device that detects a movement position based on two periodically fluctuating encoder read signals whose phases differ by 90 degrees from each other, at least one of the two signals is used as a reference signal. and a signal forming means for obtaining a signal having a phase difference of 180° with respect to the reference signal, and measuring points of a plurality of interpolation intervals in which one period of the signal is divided by points at which each of the signals becomes zero. subtracting means for generating a subtraction signal by subtracting the value of one signal from the value of the other signal among two signals, one of which is positive and one of which is negative and which changes in substantially the same direction in the interpolation interval including; An encoder reader comprising: a dividing means for dividing the value of one of the signals by the subtraction signal to generate a division signal; and a multiplication means for multiplying the divided signal by a predetermined coefficient to obtain an interpolation result. Signal interpolator. 5. The interpolation device according to item 4, further comprising a virtual signal forming means for forming a virtual signal by combining the two periodically fluctuating signals, and combining the periodically fluctuating signal and the virtual signal. An interpolation device for encoder read signals, characterized in that the position of a measurement point is determined by: