JPH10197288A - Insertion circuit for encoder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an insertion circuit for encoder in which dynamic accuracy is enhanced by eliminating unevenness of insertion pulse and over speed detection can be carried out without sacrifice of dynamic accuracy. SOLUTION: A phase angle conversion circuit 1 generates a phase angle data PH at the period of a clock CK1 and holds the generated data. The PH is updated through a data update circuit 2 by taking the difference from a current value data CNT and adding the difference to a CNT while setting an upper limit and the updated PH is held. A difference data DELTA1 is integrated through an integration circuit 3 at a high rate clock CK2 and a carrier signal is QUADEN is generated every time when the integrated value reaches the periodic ratio of the CK1 and CK2. A two-phase square wave generator 5 generates a digital twophase square wave data at the timing of QUADEN. An over speed detection circuit 6 monitors the difference data DX and outputs an over speed alarm signal under specified decision conditions.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention obtains high-resolution two-phase square wave data by digitally interpolating a two-phase sine wave signal of an encoder for detecting position, angle, velocity, angular velocity and the like. It relates to an interpolation circuit.

[0002]

2. Description of the Related Art There is a processing limit in the interval between gratings formed on an encoder scale. Therefore, in order to measure an interval finer than the scale grating, it is necessary to further subdivide and interpolate the spatial period of the phase change of the sinusoidal signal output from the encoder. For this reason, various interpolation circuits have conventionally been used. ing.

FIG. 11 shows an example of a conventional interpolation circuit using digital processing. When the encoder 40 outputs the two-phase sine wave signals INA and INB of the A and B phases whose phases are shifted from each other by 90 °, the two-phase sine wave signals INA and INB are respectively converted into first signals by the sample and hold circuits 41a and 41b. The digital data DA and DB are sampled by the clock CK1 and are converted by the A / D converters 42a and 42b.
Is converted to Based on the digital data DA and DB, phase angle data PH of each sampling point is obtained by using a memory 43 storing a look-up table. The look-up table is created based on the following equation 1 using an arctangent function (ATAN).

[0004]

## EQU1 ## PH = ATAN (DA / DB)

FIG. 12 shows the relationship between the two-phase sine wave signals INA and INB and the phase angle data PH. Memory 43
Is obtained from the clock CK1
Are sequentially taken into the register 44 and further transferred to the register 45. The difference between the phase angle data shifted by one cycle held in these registers 44 and 45 is obtained by a subtractor 46, and based on the difference data DX, a second higher-frequency second signal synchronized with the first clock CK1 is obtained. The two-phase square wave data is generated by the two-phase square wave generation circuit 47 driven by the clock CK2.

[0006] The two-phase square wave generating circuit 47 generates the difference data D
Two-phase square wave output OUTA, OU according to the polarity and absolute value of X
4 phases (0,0) of the combination of TB (A, B),
By selecting a state transition between (1, 0), (1, 1), and (0, 1), digital two-phase square wave data including information on the moving direction of the scale is generated. The state transition diagram is as shown in FIG. When the phase angle data PH increases (when the difference data is positive), (0,0) → (1,0) → (1,1) →
(0, 1) → (0, 1) → (0, 1) → in the direction indicated by DOWN when the difference data is negative.
By performing a state transition of (1,1) → (1,0), digital two-phase square wave data is generated.

FIG. 14 shows a digital 2 signal when the difference data DX in each cycle of K-1, K, K + 1, K + 2 of the first clock CK1 is 0, 1, 5, 3 according to the conventional method. This shows how square wave data can be obtained. All the rising and falling edges of the two-phase square wave data OUTA and OUTB correspond to so-called interpolation pulses, and the external counter shown in FIG. 14 indicates the value of the external counter that counts the interpolation pulses. Like this
The phase angle data PH sequentially obtained by the first clock CK1 is subjected to an increase / decrease process of the difference data DX between the phase angle data PH and the previous value by the high frequency second clock CK2 within the clock cycle, thereby obtaining the first data. , Two-phase square wave data OUTA, OUTB interpolated by the number of divisions determined by the period ratio of the second clocks CK1, CK2.

[0008]

However, the above-mentioned conventional interpolation method using digital operation has a problem that the density of interpolation pulses becomes large as is apparent from FIG. This is because the difference data for each cycle of each phase angle data is directly increased or decreased by the second clock CK2 having a higher frequency than the first clock. The density of the interpolation pulse causes a decrease in the dynamic accuracy of the encoder, that is, the accuracy of real-time position detection during encoder movement. In particular, in a numerical control device or the like, when speed information is obtained from a position difference at regular time intervals, the accuracy of the speed information deteriorates.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and aims at improving dynamic accuracy by eliminating the density of interpolation pulses, and performing overspeed detection without impairing the dynamic accuracy. It is an object of the present invention to provide an interpolation circuit of an encoder which enables the above.

[0010]

SUMMARY OF THE INVENTION An interpolation circuit of an encoder according to the present invention digitally samples a two-phase sine wave signal output from the encoder with a 90 ° phase shift by a first clock having a constant frequency. Phase angle data conversion means for converting data into data and obtaining phase angle data corresponding to each sampling phase from the correlation of the obtained digital data; and phase angle data sequentially obtained by the phase angle data conversion means and one cycle before Differential means for obtaining a difference from the phase angle data; and a first clock which synchronizes the differential data obtained by the differential means with a second clock having a higher frequency than a first clock synchronized with the first clock. Time interpolating means for converting the pulses into pulses which are uniformly discretized within the period of And 2-phase square wave generating means for generating a square wave data, is characterized in that by monitoring the difference data obtained by said difference means and a Overspeed sensing means for determining the overspeed of the encoder.

[0011] The interpolation circuit of the encoder according to the present invention also samples the two-phase sine wave signals output from the encoder by 90 ° out of phase with a first clock having a constant frequency, and converts them into digital data. Phase angle data conversion means for obtaining phase angle data corresponding to each sampling phase from the correlation of the obtained digital data,
The difference data obtained by taking the difference between the phase angle data sequentially obtained by the phase angle data conversion means and the phase angle data one cycle before is limited to the upper limit value and added to the phase angle data one cycle before. Data updating means for holding the phase angle data to be updated, and integrating the difference data with the upper limit limited by a second clock higher in frequency than the first clock synchronized with the first clock; Integrating means reset by the first clock;
Carry detecting means for detecting that the integrated value has reached or exceeded the cycle ratio of the first clock and the second clock and generating a carry detection signal; and the carry detecting means synchronizing with the second clock. Two-phase square wave generating means for generating digital two-phase square wave data at the timing of a rising detection signal; and overspeed detecting means for monitoring the difference data obtained in the data updating means to determine overspeed of the encoder It is characterized by having.

Preferably, in the present invention, the phase angle data conversion means includes A / D conversion means for sampling each of the two-phase sine wave signals with the first clock and converting them into digital data, A memory for storing a look-up table for obtaining the phase angle data from data, and first data holding means for sequentially holding the phase angle data read from the memory at the first clock; The updating means includes the first
Second data holding means for holding phase angle data updated by the phase angle data sequentially taken into the first data holding means and the phase angle data one cycle before in synchronization with the clock of Subtraction means for calculating a difference between the phase angle data taken in the first data holding means and the phase angle data held in the second data holding means; and an upper limit value of the difference data obtained by the subtraction means. A limiter that limits the period ratio between the first clock and the second clock to be equal to or less than the difference data whose upper limit is limited by the limiter and the phase angle data held by the second data holding unit; An adder for updating the phase angle data held in the second data holder is provided.

Preferably, in the present invention, the phase angle data conversion means includes A / D conversion means for sampling each of the two-phase sine wave signals with the first clock and converting the same into digital data, A memory for storing a look-up table for obtaining the phase angle data from data, and first data holding means for sequentially holding the phase angle data read from the memory at the first clock; The updating means includes the first
Second data holding means for holding phase angle data updated by the phase angle data sequentially taken into the first data holding means and the phase angle data one cycle before in synchronization with the clock of Subtraction means for taking a difference between the phase angle data taken in the first data holding means and the phase angle data held in the second data holding means; and taking an absolute value of the difference data obtained by the subtraction means An absolute value unit, a polarity detection unit for detecting the polarity of the difference data obtained by the subtraction unit, and an upper limit value of the absolute value of the difference data obtained by the absolute value unit, the first clock and the second clock. A limiter for limiting the period ratio of the clock to not more than
Polarity adding means for adding the polarity detected by the polarity detecting means to the difference data whose upper limit value has been limited by the limiter, and the difference data to which the polarity has been added by the polarity adding means and the second data holding means. An adder for adding the held phase angle data and updating the phase angle data held in the second data holding means is provided.

In the present invention, the overspeed detecting means is obtained by, for example, a comparing means for detecting that the absolute value of the difference data is equal to or larger than a judgment reference value and outputting an excess detection signal, and the comparing means. Determining means for fetching the excess detection signal in synchronization with the first clock and determining that overspeed has occurred when a predetermined number of excess detection signals are continuously obtained, and outputting an overspeed alarm signal; Is done. In the present invention,
The overspeed detection unit also includes a comparison unit that detects that the absolute value of the difference data is equal to or greater than a determination reference value and outputs an excess detection signal, and a polarity detection unit that detects the polarity of the difference data, The excess detection signal obtained by the comparing means is continuously detected in synchronization with the first clock, and during that time, it is determined that the polarity detecting means has detected the polarity inversion of the difference data, and the overspeed is detected. Determining means for outputting an alarm signal.

According to the present invention, the two-phase sine wave signal output from the encoder, the difference data for each cycle of the phase angle data converted by the first clock, and the second frequency higher than the first clock are used. The pulse is equally time-interpolated within the period of the first clock in synchronization with the clock of
It is converted to a two-phase square wave. Specifically, there is provided a data updating means for updating the phase angle data by sequentially adding the difference data of the phase angle data for each cycle to the phase angle data of one cycle before by limiting the upper limit value. Then, the differential data obtained by the data updating means and having the limited upper limit is integrated with the second clock while resetting at every first clock cycle, and the integrated value becomes equal to the cycle of the first and second clocks. Each time the ratio is reached, a carry detection signal is generated. This carry detection signal is equivalent to an interpolation pulse, and generates two-phase square wave data that changes state by switching the counting direction according to its polarity. By setting the upper limit value of the differential data of such a phase angle and integrating the differential data in which the carry range is set to the cycle ratio of the first and second clocks, the number corresponding to the differential data value is obtained. The carry detection signal is generated at an integer multiple of the second clock cycle at smaller intervals as the difference data value increases, and the first detection signal is generated at the first clock position.
Are distributed substantially evenly within the clock cycle of
Therefore, the dynamic accuracy of the encoder is improved by making the interpolation pulse distribution uniform.

In the present invention, a first clock for performing data sampling is replaced with a second clock for interpolation.
Is limited by the performance of the external counter, and it is preferable to increase the frequency within the limited range. At this time, since the number of interpolation pulses to be interpolated within the cycle of the first clock is the cycle ratio of the first and second clocks, the upper limit value of the difference data of the phase angle data is preferably set to the cycle ratio ( Or less) and an integration operation of the difference data in which the data range in which the carry signal is generated is also set to the same cycle ratio, makes the interpolation pulse based on the difference data uniform within the cycle of the first clock. Can be distributed to

On the other hand, even if the sliding speed of the encoder is restricted by the specification, the sliding speed may increase instantaneously due to handling, and an instantaneous fluctuation of the two-phase sine wave signal may occur due to an electrical factor. Therefore, if the upper limit value of the difference data is simply set, a large variation is discarded, which causes erroneous counting. In contrast, in the present invention, by introducing a feedback function to the data updating means,
Such erroneous counting can be prevented. That is, when the magnitude of the phase angle difference data instantaneously exceeds the upper limit value, the range updated in one cycle is limited to the upper limit value or less. Since the difference from the phase angle data is calculated and updated, the correct current value data is autonomously restored and complemented even if there is a large instantaneous change. Therefore, erroneous counting as an incremental counter is avoided.

Further, in the present invention, if an absolute value unit for converting the difference data sent to the integration means for pulse equalization into an absolute value is provided in the data updating means, the detection algorithm of the carry detection means makes the difference data Irrespective of the sign, the carry is easily detected. If such an absolute value device is not used, high-speed performance is sacrificed, but the circuit scale is correspondingly reduced.

Furthermore, in the interpolation circuit according to the present invention, there is an encoder feed speed that the circuit can tolerate. If the feed speed exceeds the allowable limit, the circuit malfunctions. Therefore, the present invention includes an overspeed detecting means for monitoring the phase angle difference data and performing an overspeed judgment so as to detect and report such a malfunction. In particular, the determination of overspeed by the overspeed detection means must be performed on condition that the difference data is equal to or greater than the determination reference value and that the excess of the determination reference value is repeated a predetermined number of times in synchronization with the first clock. Thus, it is possible to prevent a situation in which an overspeed alarm signal is frequently output due to mechanical vibration or the like, and to reliably perform the overspeed determination without impairing the high-speed response of the interpolation circuit.

[0020]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows an interpolation circuit according to one embodiment of the present invention. Although an encoder is not shown, the encoder outputs a two-phase sine wave signal and is, for example, a magnetic encoder using a magnetoresistive element and a magnetic scale. However, the present invention is not limited to this, and the present invention can be applied to photoelectric type and other encoders.

This interpolation circuit converts the two-phase sine wave signals INA and INB output from the encoder into a first clock CK.
1 (period PCK1) and digitally process to obtain phase angle data PH, and a first clock CK based on the obtained phase angle data PH.
And a data updating circuit 2 for sequentially updating and holding the current value data in one cycle. Further, as means for equalizing interpolation pulses,
The difference data DX for each cycle of the phase angle data obtained in the data update circuit 2 is fetched and the second clock CK2
(Period PCK2). The integrator 3 is set so that a carry is performed for each data range M = PCK1 / PCK2, and a carry detector 4 for detecting the carry is provided. A two-phase square wave generating circuit 5 for generating two-phase square wave data in the counting direction according to the polarity of the difference data is provided. Further, an overspeed detection circuit 6 is provided which monitors difference data DX of the phase angle data obtained in the data update circuit 2 for each cycle and determines overspeed of the encoder.

The specific circuit configuration and operation of each unit are as follows. The phase angle conversion circuit 1 includes two-phase sine wave signals INA and IN
B sampled and held by the first clock CK1, respectively, sample and hold circuits 11a and 11b, A / D converters 12a and 12b for converting a sampled value into digital data, and A / D converters 12a and 12b.
And a look-up table memory 13 for obtaining phase angle data based on the digital data obtained in step 12b. The obtained phase angle data PH is stored in the first clock CK
1 is held in the first register 14. The look-up table is created based on Equation 1 described above.

The data update circuit 2 receives the first clock CK
2 that stores the current value data CNT driven by
The present value data CNT held in the second register 27 is used as the phase angle data PH which is sequentially newly sent to the first register 14 in the phase angle conversion circuit 1 by the subtracter 21. Are obtained, and are sequentially updated by the obtained difference data DX. In the case of this embodiment, the difference data DX calculated by the subtractor 21
, The absolute value is obtained by the absolute value device 22 and the polarity value data POL is obtained by the polarity detection circuit 23, and the absolute value of the obtained difference data DX is set to the difference data DELTA1 whose upper limit is limited by the limiter 24. You. The upper limit LMT of the absolute value of the difference data DX limited by the limiter 24 is set to M = PCK1 / PCK2 or less.

Then, the polarity value data POL is again added to the difference data DELTA1 by the polarity adding circuit 25, and
DELTA2 is added to the current value data CNT by the adder 26, and the added value S1 is taken into the second register 27. That is, the first clock CK
In synchronization with 1, the current value data CNT is fed back to the subtractor 21 to obtain a difference from the phase angle data PH.
The current value data CNT is repeatedly updated based on the difference data DX.

The operation of the phase angle conversion and the phase angle data update up to this point will be described with reference to specific numerical examples. PCK1 =
200 ns (= 5 MHz), PCK2 = 25 ns (= 40
MHz), these period ratios are M = PCK1 / P
CK2 = 8, which is also the upper limit LMT in the limiter 24 in this embodiment. When the difference data obtained by the subtracter 21 is DX = -10, the difference data obtained by the absolute value device 22 and the limiter 24 is DELTA1 = 8.
The polarity data POL is negative, and the polarity addition circuit 25
The difference data to which the polarity is added again by DELTA2 =
−8. That is, the value to be updated is suppressed below the upper limit LMT limited by the limiter 24, and the current value data CNT
Is updated repeatedly.

More specifically, a case where the interpolation number INP = 400 will be exemplified. At this time, the input sine wave signals INA, IN
The relationship between the period λ of B and the resolution R of the two-phase square wave outputs OUTA and OUTB is given by the following equation (2).

[0027]

## EQU2 ## R = λ / INP

When λ = 4 μm and INP = 400, R =
4 μm / 400 = 0.01 μm Assuming that the resolution of the A / D converters 12a and 12b is 10 bits, the address of the look-up table memory 13 requires 20 bits. The data range of the phase angle data PH converted by the memory 13 is the interpolation number INP, and is a numerical value between 0 and 399. Minimum scale of phase angle data PH (1L
SB) corresponds to 360 ° / INP = 0.9 °. At this time, the algorithm from the phase angle data PH to the present value data CNT in the periods 1 to 4 of the first clock CK1 is shown as a specific numerical example in Table 1 below.

[0029]

[Table 1]

Table 1 shows a case where the initial value of the current value data CNT is 200 and the phase angle data PH = 200 is obtained in the cycle 1. In this cycle 1, CNT does not change. In period 2, the encoder moves and PH = 19
8 is obtained. At this time, the difference data DX = −
2 and the difference data D of the absolute value passing through the limiter 24
ELTA1 = 2, and the added value S1 to the current value data CNT is −2 + 200 = 198. This becomes the current value data CNT of the next cycle 3, where the phase angle data P
When H = 188 is obtained, since the difference data DX = −10, the absolute value DELTA1 is limited by the upper limit LMT = 8. In the period 4, the case where the phase angle data PH = 192 larger than the current value CNT is obtained is shown. At this time, the added value is S1 = 192, which is equal to PH. That is, the update value is temporarily insufficient in the previous cycle 3, but is complemented in the cycle 4 in which the update amount is reduced.

If the amount to be updated is equal to or less than the upper limit, the first
Can be completely updated within one cycle of the clock CK1. The sliding speed V of the encoder that satisfies such a condition
Is represented by Equation 3 below.

[0032]

V ≦ LMT × R / PCK1 = 8 × 0.01 [μm] / 200 [ns] = 400 [mm / s]

In the case of this embodiment, V> 400 m instantaneously
Even at m / s, the current value data CNT is autonomously restored by the feedback control described above. V> 4
When the state of 00 mm / s continues and the difference data DX increases, a situation occurs in which the current value data delayed by one cycle of the sine wave is fed back, and at this time, a malfunction occurs. In the above example, the case where the absolute value of the difference data exceeds 200 corresponds to this. Therefore, the maximum feed rate at which the circuit can continuously respond is defined by Equation 3.

Next, the difference data DELTA1 with the upper limit obtained by the data updating circuit 2 and the polarity data PO
The specific configuration and operation of the portion of the integration circuit 3 that performs the pulse equalization processing will be described using L. The integration circuit 3 has a third register 31 for storing the difference data DELTA1 with the first clock CK1, and stores the difference data DELTA1 held in the register 31 in the second high-frequency signal synchronized with the first clock CK1. An adder 32 accumulates and holds the clock CK2 and a fourth register 33. The fourth register 33 is reset by the first clock CK1. That is, the integrated value INTPCNT is determined by the first clock C
The difference data DELTA1 obtained in a certain cycle of K1 is sequentially doubled every cycle of the second clock CK2 in the next cycle,
Three times, ...

The carry detection circuit 4 calculates the integral value INTPCN
Each time T reaches the period ratio M, the first clock C
The carry detection signals QUADEN of a number corresponding to the size of the difference data DELTA1 are generated within the cycle of K1. This carry detection signal QUADEN becomes an interpolation timing signal. Specifically, in the case of M = LMT = 8, the change of the integral value obtained according to the magnitude of the difference data DELTA1 and the carry detection caused by this change within the cycle of the first clock CK1. The state of the signal QUADEN is shown in FIG. If the difference data DELTA1 = 8, the carry detection signal Q at all timings of the second clock CK2
It can be seen that even if UADEN is generated and the difference data DELTA1 becomes smaller, the carry detection signal QUADEN can be obtained in a form dispersed within the cycle of the first clock CK1. This is a result of the setting of the upper limit LMT = M in the data update circuit 2 and the integration operation of the difference data DELTA1 with the data range set to M.

The two-phase square wave generating circuit 5 generates the two-phase square wave data by the state transition shown in FIG. 13 at the timing of the carry detection signal QUADEN and in the counting direction based on the polarity value data POL of the difference data. OUTA, OUT
B is output. Table 2 below shows that the difference data DELTA in the periods K, K + 1, K + 2 of the first clock CK1.
Taking the case where 1 = 1, 5, 3 as an example, the integral I
9 shows changes in NTPCNT, a carry detection signal QUADEN, and an external counter.

[0037]

[Table 2]

That is, at the cycle K of the first clock CK1,
Assuming that the differential data DELTA1 is 1, INTPCNT is first initialized to INTPCNT = 0 at the first clock CK1, and the integrated value INTPCNT is obtained at the period 1 to 8 of the second clock CK2.
And the difference data DELTA1 are added. In Table 2, upper digits exceeding the data range M = 8 are also shown for easy understanding, but these upper digits are unnecessary in data processing and can be omitted in an actual processing circuit. INTPCNT = 8 in the period 8 of the second clock CK2,
When the carry detection signal QUADEN is generated, two-phase square wave data is output at that timing. Hereinafter, the cycle K +
Similarly, when the integral value INTPCNT exceeds an integral multiple of 8, a carry signal QUADEN is generated at 1, K + 2.

The above interpolation operation is shown in FIG. 3 corresponding to the conventional FIG. As is apparent from comparison with FIG. 14, the interpolation is equalized within the period of the first clock CK1. As described above, according to this embodiment, the phase angle data is configured to be updated by feedback using the difference data, the upper limit value of the difference data is set to the cycle ratio M of the first and second clocks, and By setting the data range of the integrating circuit for generating the carry detection signal for interpolation by integrating the difference data to the same cycle ratio M, the interpolation pulses are evenly distributed within the cycle of the first clock. It becomes possible. Moreover, even when the magnitude of the difference data of the phase angle data instantaneously exceeds the upper limit, the correct current value data is restored and complemented autonomously by the feedback mechanism, and erroneous counting as an incremental counter is performed. It is to be avoided.

In this embodiment, the difference data D
By using X as the absolute value data DELTA1,
The carry detection algorithm in the integration circuit is simplified. That is, when expressed in 2's complement binary notation, for example, +7
0111 and -7 become 1001 and there is no symmetry, so that the carry detection algorithm differs between positive and negative until now. When this is converted into an absolute value, +7 becomes the polarity bit 0+
0111 and -7 are polarity bits 1 + 0111, and carry detection can be performed by the same algorithm. This carry detection requires high speed for the above-described interpolation processing. In the above example, it is necessary that addition of 4 bits can be completed within the cycle of the second clock CK2. It is effective to simplify the data structure by converting to an absolute value as shown in FIG.

However, in the present invention, the absolute value conversion of the difference data is not always necessary. For example, the configuration shown in FIG. 4 in which the absolute value conversion circuit in FIG. 1 is omitted may be adopted. Although it is difficult to perform high-speed processing as compared with FIG. 1, the circuit scale is reduced, and especially when the number of interpolations is large, the effect of the reduction is increased.

For a specific method of realizing each circuit section, for example, as a look-up table memory, R
A nonvolatile memory such as an OM, an EPROM, an EEPROM, and a flash memory is used. The arithmetic circuit includes a digital logic circuit (general-purpose gate, gate array, F
PGA, etc.), a programmable operation circuit (DSP, embedded microcontroller, etc.), software and the like are used. For the number of interpolations, the output of the encoder is 2
Since the number of interpolations is already 4 by using a phase sine wave signal, the present invention can be applied to an interpolation circuit having an arbitrary natural number of 5 or more. In particular, the use of 2 ⁿ (n is a natural number) is advantageous because an arithmetic circuit such as an adder is simplified. Further, the period ratio M of the first and second clocks CK1 and CK2 may use an arbitrary natural number. In particular, if M = 2 ^m (m is a natural number), the equalizing pulse generating circuit section is simplified, and as a result, the delay time is reduced, and high-speed performance is easily realized.

Next, a specific configuration example of the overspeed detection circuit 6 for monitoring the difference data DX and detecting the overspeed will be described. FIG. 5 is a configuration example of the overspeed detection circuit 6. Since the difference data DX is a signal indicating a delay in following the position, in this embodiment, the absolute value | DX | of the difference data DX is determined by a certain determination reference value DXALM.
Is different from the difference data | DX |
A magnitude comparator 61 is used to detect that it has become M or more. The magnitude comparator 61 calculates the difference data | DX |
Upon detecting the above, an excess detection signal DXOVR is output. This is the first condition for overspeed detection. For example, the periods of the first and second clocks CK1 and CK2 are CK1 = 200 ns and CK2 = 25 ns, respectively.
, These ratios are the maximum number of pulses that can be time-interpolated, and in this example are eight. On the other hand, difference data | DX |
Can take a value of 8 or more, so the determination reference value DXALM may be set to a large value of 8 (limit value) or more. However, if the difference data | DX | exceeds half the number of interpolations,
Since the autonomous restoration function of the circuit is impaired, the criterion value D
XALM is selected within the range of 1/2 or less of the interpolation number.

On the other hand, if the sensitivity of the overspeed detection is set too high, an alarm is issued frequently, and the frequency of emergency stop increases. This can be caused by insufficient signal adjustment or deterioration of signal quality due to contamination of the encoder sensor, mechanical vibration, electromagnetic noise, or the like. That is, when the sine wave signals INA and INB deviate from the ideal state, a situation in which the difference data DX instantaneously exceeds the determination reference value DXALM in a part of 360 ° may occur before reaching the allowable feed speed. In order to prevent such a situation, the second condition of the overspeed detection may be that the difference data | DX | continuously exceeds the determination reference value DXALM by a predetermined number N consecutive times.

Therefore, in FIG. 5, the excess detection signal DXOVR =
A counter 62 for counting 1; a judgment circuit 63 for judging that the count value of the counter 62 has become N; and an RS flip-flop for holding the judgment result of the judgment circuit 63 and outputting an overspeed alarm signal OSALM 64. The counter 62 counts the excess detection signal DXOVR = 1 in synchronization with the first clock CK1, but is reset by DXOVR = 0. Therefore, during the N consecutive first clocks CK1, the overspeed alarm signal OSALM = 1 is obtained only when the excess detection signal DXOVR = 1.

FIG. 6 shows another example of the overspeed detection circuit 6 for detecting overspeed under the same conditions as in FIG. In this embodiment, the magnitude comparator 6
The N-stage shift registers 651 to 65 operated by the first clock CK1 are supplied with the excess detection signal DXOVR obtained from
N, and sequentially transfers the data. The AND gate 66 detects that the output of each stage has become all 1, and the detected output is held by the RS flip-flop 64.
The RS flip-flop 64 is reset by an external alarm reset signal.

When the overspeed detection circuit 6 of FIGS. 5 and 6 is realized by software, the flow is as shown in FIG. That is, the overspeed alarm signal OSALM and the count value K are initialized as 0 (S
1), it is determined whether or not the difference data | DX | is greater than or equal to the determination reference value DXALM (S2). If YES, the count value K is counted up (S3).
The count value K is initialized (S4). Count value K is N
Is determined (S5). If NO, the same operation is repeated, and if YES, an overspeed alarm signal OSALM = 1 is output.

Next, the operation of this embodiment will be clarified based on specific experimental data. The experimental conditions are as follows. In FIG. 1, the A / D converters 12a and 12b have a 10-bit A / D converter.
A D converter, the number of interpolations is 400, and the input range of the A / D converters 12a and 12b is 2.6V ± 1.0V.
The A and B phase sinusoidal input signals INA and INB are shown in Table 3 below.
Conditions 1-4. In Table 3, the sinusoidal signal is represented by a ± b using the center voltage a [V] and the amplitude b [V]. The parentheses of INB indicate the phase difference from INA.

[0049]

[Table 3]

Various selections of the judgment reference value DXALM and the judgment count value N as the overspeed judgment conditions are shown in Table 3 below.
FIG. 8 shows the results of an experiment of overspeed detection performed on the signal input conditions 1 to 4 described above. From these results, the criteria for overspeed determination were DXALM = 64, N = 1,
If the values are set to 2, 4, and 8, overspeed detection can be reliably performed at a circuit response limit of 100 kHz in any of the input signal conditions 1 to 4. If other criteria are set, overspeed is detected at less than 100 kHz, and accurate overspeed detection cannot be performed.

FIG. 9 shows another example of the configuration of the overspeed detection circuit 6. This overspeed detection circuit 6
The principle is to detect the moment when the polarity of the difference data DX is inverted when the circuit becomes overspeed that cannot be followed. As described above, in the interpolation circuit of this embodiment, the interpolation number 400 is a difference between the phase angle data PH and the data (current value data) CNT one cycle before even if the speed is instantaneously overspeeded. Data DX is -199 ≦ D
X ≦ 199, that is, the period of the input sine wave signal is 180
° In the range of delay, malfunction does not occur due to the feedback function of the circuit. When the overspeed state continues and the difference data DX goes out of the above-mentioned range, the circuit acts to lock with a tracking value delayed by one cycle, and at this time, the polarity of the difference data DX is inverted. The overspeed detection circuit 6 in FIG. 9 uses this polarity inversion as an overspeed determination condition.

That is, as in FIGS. 5 and 6, the magnitude comparator 61 is provided for detecting that the difference data | DX | has become equal to or greater than the determination reference value DXALM. A flip-flop 83 is provided to take in the excess detection signal DXOVR by the magnitude comparator 61 in synchronization with the first clock CK1. AND gate 85
Is provided. On the other hand, a polarity detection circuit 81 that detects the polarity of the difference data DX and outputs the polarity data DXPOL such that DXPOL = 1 when DX ≧ 0 and DXPOL = 0 when DX <0, and outputs the polarity data DXPOL to the first Flip-flop 8 which takes in in synchronization with clock CK1
2 and an EXOR gate 84 for detecting a mismatch between the current polarity data DXPOL1 and the polarity data DXPOL2 one cycle before, that is, a polarity inversion.

For example, as shown in FIG. 10, the difference data DX increases in the positive
If it exceeds 0, the polarity is inverted. Before and after |
DX | are both large and continuously exceed the criterion value DXALM, so the excess detection signals DXOVR1 and DXOVR2
Are both 1 and this is detected by the AND gate 85. The EXOR gate 84 detects the polarity inversion of the difference data DX. Therefore, by taking the product of the outputs of these gates 84 and 85 by the AND gate 86, the excess detection signal DXOVR = 1 is continuously obtained in synchronization with the first clock CK1, and the difference data DX is generated during that time. The polarity reversal, that is, the tracking of one cycle delay due to the overspeed is detected. This A
The output of the ND gate 86 is taken into the RS flip-flop 74, and the overspeed alarm signal OSALM =
You can get 1.

As described above, according to this embodiment, it is possible to monitor the difference data DX of the phase angle data in the interpolation circuit and to reliably detect the overspeed even if the quality of the sinusoidal signal is reduced to some extent. it can. In particular, if the overspeed detection circuit shown in FIGS. 5 and 6 is used, overspeed detection will not be performed until the limit frequency almost equal to the maximum feed speed of the encoder is reached. It is possible to prevent a situation in which an overspeed alarm is issued more frequently, and to sufficiently draw out the response performance of the circuit. Therefore, the reliability of the high-speed interpolation circuit is high.

[0055]

As described above, according to the present invention, the two-phase sine wave signal output from the encoder is used to convert the difference data for each cycle of the phase angle data converted by the first clock into one cycle. Phase angle data updating means for limiting the upper limit value and sequentially adding the current value data to update the current value data is provided, and the difference data with the upper limit value obtained in the data updating means is limited to the first clock. Higher frequency second
In this case, the integration is performed while resetting at every first clock cycle, and a carry detection signal is generated every time the integrated value reaches the cycle ratio of the first and second clocks. In addition, two-phase square wave data that changes state by switching the counting direction according to the polarity is generated. As a result, the carry detection signals of the number corresponding to the difference data value obtained within the first clock cycle, that is, the interpolation pulses, are evenly distributed within the first clock cycle, and the dynamic accuracy of the encoder is improved. Further, according to the present invention, an overspeed detecting means for monitoring difference data of the phase angle and performing an overspeed determination is provided, so that the overspeed determination can be reliably performed without impairing the dynamic accuracy improvement of the interpolation circuit. Can be.

[Brief description of the drawings]

FIG. 1 shows a configuration of an interpolation circuit according to an embodiment of the present invention.

FIG. 2 shows how a carry detection signal is generated by the interpolation circuit of the embodiment.

FIG. 3 shows an operation of a two-phase square wave output by the interpolation circuit of the embodiment.

FIG. 4 shows another configuration example of the interpolation circuit.

FIG. 5 shows a configuration example of an overspeed detection circuit of the embodiment.

FIG. 6 shows a configuration example of an overspeed detection circuit of the embodiment.

FIG. 7 shows an operation flow of the overspeed detection circuit of FIGS. 5 and 6;

FIG. 8 shows a performance experiment result of the overspeed detection circuit.

FIG. 9 shows a configuration example of another overspeed detection circuit.

FIG. 10 is a diagram for explaining an operation of the overspeed detection circuit of FIG. 9;

FIG. 11 shows a configuration of a conventional interpolation circuit.

FIG. 12 shows a relationship between a two-phase sine wave signal and phase angle data.

FIG. 13 shows a state transition of the two-phase square wave generation circuit.

FIG. 14 shows the operation of a two-phase square wave output by a conventional interpolation circuit.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Phase angle conversion circuit, 11a, 11b ... Sample hold circuit, 12a, 12b ... A / D converter, 13 ... Memory, 14 ... First register, 2 ... Data update circuit, 2
DESCRIPTION OF SYMBOLS 1 ... 2nd register, 22 ... absolute value device, 23 ... polarity detection circuit, 24 ... limiter, 25 ... polarity addition circuit, 26 ... adder, 27 ... 2nd register, 3 ... integration circuit, 31 ... 3rd , 32 ... adder, 33 ... fourth register,
4 ... Carry detection circuit, 5 ... Two-phase square wave generation circuit, 6 ... Overspeed detection circuit, CK1 ... First clock, C
K2: second clock.

Claims (6)

[Claims] 1. A two-phase sine wave signal output from an encoder and shifted by 90 ° is sampled by a first clock having a constant frequency and converted into digital data.
Phase angle data conversion means for obtaining phase angle data corresponding to each sampling phase from the correlation between the obtained digital data, and phase angle data obtained one cycle before by the phase angle data sequentially obtained by the phase angle data conversion means. A difference means for obtaining a difference, wherein the difference data obtained by the difference means is synchronized with a second clock having a higher frequency than the first clock synchronized with the first clock and is equalized within a period of the first clock. Time interpolating means for converting into discrete pulses, two-phase square wave generating means for generating digital two-phase square wave data in synchronization with the pulse obtained by the time interpolating means, Interpolating the encoder, comprising: overspeed detecting means for monitoring difference data to determine overspeed of the encoder. circuit. 2. A two-phase sine wave signal having a 90 ° phase shift output from an encoder is sampled by a first clock having a constant frequency and converted into digital data.
Phase angle data conversion means for obtaining phase angle data corresponding to each sampling phase from the correlation between the obtained digital data, and phase angle data obtained one cycle before by the phase angle data sequentially obtained by the phase angle data conversion means. Data updating means for limiting the upper limit value of the difference data obtained from the difference, adding the difference data to the phase angle data of one cycle before and retaining the updated phase angle data, and the difference data having the limited upper limit value Integrating means, which is integrated with a second clock higher in frequency than the first clock synchronized with the first clock and holds the integrated value, reset by the first clock; Carry detection means for detecting that the period ratio of the first clock and the second clock has been reached or more and generating a carry detection signal; Two-phase square wave generating means for generating digital two-phase square wave data at the timing of a rising detection signal; and overspeed detecting means for monitoring the difference data obtained in the data updating means to determine the overspeed of the encoder And an interpolation circuit for an encoder. 3. The phase angle data conversion means: A / D conversion means for sampling each of the two-phase sine wave signals with the first clock and converting the same into digital data; A memory storing a look-up table for obtaining angle data; and first data holding means for sequentially holding the phase angle data read from the memory at the first clock, wherein the data updating means comprises: Second data holding means for holding phase angle data updated by the phase angle data sequentially taken into the first data holding means in synchronization with the first clock and the phase angle data one cycle before. Subtracting means for taking a difference between the phase angle data taken in the first data holding means and the phase angle data held in the second data holding means. A limiter for limiting an upper limit value of the difference data obtained by the subtraction means to a cycle ratio of the first clock and the second clock, and a difference data having an upper limit value limited by the limiter; 3. An adder for adding the phase angle data held by said data holding means and updating the phase angle data held by said second data holding means. The interpolation circuit of the encoder. 4. The phase angle data conversion means: A / D conversion means for sampling each of the two-phase sine wave signals with the first clock and converting the same into digital data; A memory storing a look-up table for obtaining angle data; and first data holding means for sequentially holding the phase angle data read from the memory at the first clock, wherein the data updating means comprises: Second data holding means for holding phase angle data updated by the phase angle data sequentially taken into the first data holding means in synchronization with the first clock and the phase angle data one cycle before. Subtracting means for taking a difference between the phase angle data taken in the first data holding means and the phase angle data held in the second data holding means. An absolute value device that takes the absolute value of the difference data obtained by the subtraction device; a polarity detection device that detects the polarity of the difference data obtained by the subtraction device; and the difference obtained by the absolute value device A limiter for limiting the upper limit value of the absolute value of the data to a cycle ratio of the first clock and the second clock, and a polarity detected by the polarity detection means to the difference data whose upper limit value is limited by the limiter. A polarity adding means for adding, and the difference data to which the polarity is added by the polarity adding means and the phase angle data held in the second data holding means are added and held by the second data holding means. 3. The encoder interpolation circuit according to claim 2, further comprising an adder for updating the phase angle data. 5. An overspeed detecting means for detecting that the absolute value of the difference data is equal to or greater than a determination reference value and outputting an excess detection signal, and the overdetection obtained by the comparing means. Determining means for fetching a signal in synchronization with the first clock and continuously determining the overspeed when a predetermined number of excess detection signals are obtained and outputting an overspeed alarm signal; An interpolation circuit for an encoder according to claim 1. 6. An overspeed detecting means, comprising: a comparing means for detecting that an absolute value of the differential data is equal to or greater than a determination reference value and outputting an excess detection signal; and a polarity for detecting a polarity of the differential data. Detecting means for continuously detecting the excess detection signal obtained by the comparing means in synchronization with the first clock, and determining that the polarity detecting means has detected polarity inversion of the differential data during the detection; 3. An encoder interpolation circuit according to claim 1, further comprising: a judging means for outputting an overspeed alarm signal.