JPH0126003B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a multi-division, high-speed reading digital counting device for length measuring instruments, particularly digital length measuring instruments. Length measuring devices that electrically detect length using a linear scale or rotary encoder and display it digitally are well known, but in order to increase resolution, the number of divisions between scale pitches may be increased. This has a mutually contradictory relationship with increasing the reading speed, and it has been extremely difficult to construct a practical circuit using conventional methods.

The present invention solves the above problems and provides a digital counting device with a simple circuit configuration, high resolution, high speed reading, and no malfunction.

Digital scales generate output signals at fixed pitches and measure the length by counting the output signals, but the pitches are limited by the mechanical structure. In order to obtain the required minimum reading value, it is common practice to divide these pitches using electrical means. Methods for this division include the zero-crossing method, which divides a two-phase signal into four, and the resistance division method, which calculates the two-phase signal using a resistor, electrically converts it into a multiphase signal, and generates a pulse at each zero-crossing point. , a phase modulation division method is known in which a carrier wave is phase modulated with a scale signal and the phase difference is counted using a clock pulse. If the number of divisions is relatively small, for example 10
~20, it is possible to count and display all the pulses obtained using a conventional division method using a counter, but if the number of divisions is large, the processing time for division will increase. There is a limit to the response speed of the counter, making it impossible to perform high-speed measurements without counting errors.

The purpose of the present invention is to enable high-speed measurement with a large number of divisions. Specifically, the scale pitch P = 10 μm, the number of divisions between 1 pitch N = 100,
The embodiment of the present invention will be described assuming that the minimum reading value P/N=0.1 μm, but the invention is not limited to this.

FIG. 1 shows the basic concept of measurement, which is the premise of the present invention. When measuring the length X using a digital scale, the measurement starting point Ps is located at an arbitrary position on the scale, and in FIG. 1 is located at _M1 from the scale signal zero crossing point. This value of M ₁ can be read by the phase modulation division method since the motor is in a stopped or slow state at the starting point. Store this M ₁ at the beginning of the measurement. The scale signal is then detected by displacement along the scale, passing through a zero crossing point every pitch of the scale and generating a pulse. At high speeds, this number of pulses n is counted and stored. At the end point Pe of the measurement, it is stopped or at a low speed like at the starting point, so the value M ₂ from the last passed zero crossing point is read by the phase modulation division method and the measurement is completed. From the value obtained in this measurement, X=np+M ₂ −M ₁
The amount of displacement X can be found by calculating .

If such a measurement method is used, the dividing circuit only needs to operate normally when the starting point and ending point of the measurement are stopped or moving at low speed; no dividing operation is required when moving in between, and only the scale pitch can be counted. Because of this, it is possible to move at a sufficiently high speed.

However, when this display device is actually manufactured and operated, the following problems occur.

Even if the same scale signal obtained by reading the same scale is used, when two pulse generation circuits are used, the pulse generation positions cannot be exactly matched. In this example, a pulse signal every 10 μm is generated at the zero cross point of each pitch of the scale signal, and a pulse signal every 0.1 μm is generated by dividing the pitch between each pitch by 100 using the phase modulation division method.
The 100th number does not necessarily match both dimensionally and temporally. Generally, it is not possible to expect the accuracy of a pulse train generated every 10 μm to be greater than 0.1 μm.

A concrete example of this is shown in FIGS. 2 and 3. In FIG. 2, A is a scale signal, and B is a correct waveform that emits a pulse at its zero-crossing point. Then, find the length from the zero cross point A on the left side to the end point Pe. Here, between AB and BC, P = 10μ
It is m. C is an abbreviation of 100 divisions between BC, and if 99 pulses are counted between B and Pe, one pulse is 0.1 μm, so B,
The distance between Pe is 0.1μm x 99 = 9.9μm, and by counting the zero cross signal at point B once between A and Pe, it becomes 10μm x 1, and by adding 9.9μm to this, the correct distance is 19.9μm. . However, the pulse that should be generated at C now is as shown in D.
Suppose that it occurs before Pe. In other words, considering the case where the zero-cross signal has a (-) direction error of 0.2 μm or more from the correct generation position, the zero-cross signal reading is counted twice between A and Pe.
10 μm x 2 = 20 μm, while C has 99 counts, so 10 μm x 2 + 0.1 μm x 99 = 29.9 μm. That is, an error of 10 μm occurs with respect to the correct value.

Conversely, if the zero-cross signal is delayed by 0.2 μm or more, the waveform shown in FIG. 3E will be obtained. Furthermore, the end point
Pe is a point 0.1 μm (+) from point C. In this case, the correct value is 10μ since there is 2P between ACs.
m x 2 = 20μm, and the distance between C and Pe is 0.1μm, so the total
It is 20.1 μm. However, since the zero cross signal at point C is delayed, only one count is made between A and Pe, resulting in 10 μm×1=10 μm. And in division counting, it goes from 99 to 0 for C, and for Pe it goes from 0.1μm×
1 indicates 0.1 μm. (However, divisional counting is not an accumulation method, but a method in which 0 is counted after 99.) Therefore, A·Pe shows 10 μm + 0.1 μm = 10.1 μm.

Although Figures 2 and 3 only show the end point, the same phenomenon is likely to occur at the starting point, and in the worst case, there is a possibility that an error equivalent to two scale pitches will occur, making it difficult to put into practical use. It won't happen.

The present invention is designed to accurately capture n counting pulses at X=np+M ₂ -M ₁ to prevent such errors from occurring.

In Fig. 4, two-phase scale signals sin2π/p x, cos2π/px (x is the amount of displacement) are used, and these are applied to a phase modulation circuit 1 divided by 100 using a known method, and the signals are input at pitch intervals of 10 μm. is converted into 100 pulses every 0.1 μm and counted by division value counter 2. This system is called the sperm system. However, the counting method is not an accumulation method but a refresh method every 99 pulses.
Only the number of m) is displayed. and the starting point of measurement
At Ps and the end point Pe, the values M ₁ and M ₂ indicated thereby are entered into the memory circuit 3 and sent to the calculation display circuit 4. The value of this counter is determined by three (precise) digital comparison circuits 5 (comparators).
It is divided into areas. In other words, the first (fine) area 0.0~aa the first (fine) area a.a+1~bb-1 the first (fine) area bb~9.9 As a condition for the above classification, the first (fine) area is
0.0, and the maximum count value of 9.9 is included in the first (fine) area, so that the aa and bb values can be any number between them. For example, as shown in Figure 5, aa=
Divide into thirds as 3.3 and bb=6.6. Digital comparison circuit 5
always outputs in which area above the counter contents are located as a counter area signal.

Further, the two-phase scale signal is input to the scale signal zero-cross signal circuit 6 and is then input to the area segmentation circuit 7.
This is connected to the coarse system. Then, using the counted pulse points every 10 μm as one boundary, the area is divided into three (coarse) regions as shown in FIG. 6, as an example.

6th (coarse) region An area that includes the part immediately after the counting pulse generation point 2nd (coarse) area An intermediate area between the 1st and 2nd areas 3rd (coarse) area An area that includes the part immediately before the counting pulse generation point FIG. The points A ₁ and A ₂ are the counting pulse generation points, the area including A ₁ and A ₂ is the first (coarse) area, and the area immediately before that is the second (coarse) area. The width of the second area can be determined arbitrarily, and it is easy to define it by the zero-crossing point of the cos signal as shown in FIG. 6, but the method of division is not limited to this. The above two systems, fine and coarse, always output a signal indicating which region of the scale it is in, and input it to the region signal switching circuit 8.

The two-phase scale signal is also sent to the speed detection circuit 9 shown in FIG. 4, where the moving speed with respect to the scale is detected from the repetition frequency of the scale signal, and a preset phase modulation 100 division circuit operates accurately. A signal indicating whether the vehicle is currently moving in a low speed range or a high speed range is sent to the area signal switching circuit as a speed state signal. The area signal switching circuit receives the high and low speed state signals and selects (precision) in the low speed state.
(fine) (fine) signal, (coarse) signal at high speed
The (coarse) (coarse) signal is sent to the next 10 μm pulse generation circuit (scale pitch pulse generation circuit) 10.

Here, the 10 μm pulse generation circuit performs the following operation at low speed. That is, a 1UP pulse is generated when changing from a region ((Sei), but (Sei) is omitted hereafter) to a region. A 1DOWN pulse is generated when there is a change from 0 to 1, and no pulse is generated for changes in other areas (←→, ←→). In other words, when the content of the division value counter changes from 9.9 to 0.0 by the 1UP pulse, a 1UP signal is sent to the upper digit (10μm digit) and counted.
When the value changes from 0.0 to 9.9 by the 1DOWN pulse, the upper digit is decreased by 1.

On the other hand, at high speed, the scale domain signal (coarse),
(Rough) (Rough) (hereinafter, (Rough) will be omitted) is 10μm
When it enters the pulse generation circuit and becomes empty,
Generate a 1UP pulse as in the case of (precision) and +10μ
Let it be m. Conversely, 1DOWN for changes from
Generates a pulse and reduces the displayed value by 10 μm.

The operation of this 10 μm pulse generation circuit is to generate pulses of 1 UP and 1 DOWN in response to changes in the area, and is characterized in that it is performed regardless of whether it is fine or coarse. In other words (sei) →
Even when changing from (coarse) to (coarse), the same 1DOWN pulse is generated, and (fine) → (coarse),
Emit the same 1UP pulse from (coarse) to (fine). Here, 10μm performs this kind of operation.
The pulse generating circuit 10 can be easily constructed by a combination of a known flip-flop circuit and gate. Then, the signal from the 10μm pulse generation circuit 10 enters the counter 11, and the calculation display circuit 4
sent to.

In the present invention described above, FIGS.
A description will be given of how the occurrence of the counting error shown in the figure can be prevented. In Figure 2, if you simply switch from coarse to fine, when a 10 μm pulse is generated before point C, the true value is 19.9 μm.
It was previously explained that counting as 29.9 μm occurs. Therefore, in Figure 7, it is now 19.9μm.
If you suddenly stop from high speed at the position, the end point Pe
can be switched from high speed to low speed. However, due to the error in the coarse system, the generation point of the 10 μm pulse is passed, so the coarse area classification is (coarse) →
(coarse) changes result in 1UP. However, since it stops at Pe and switches to low speed, a precision signal is input as the area signal. In the seminal system, the count value is 9.9, which is the domain (sei). By switching to low speed, the area signal changes from (coarse) to (fine), and the 10μm pulse generation circuit generates a 1DOWN pulse, so the incorrect display of 29.9μm at high speed is correct when switching to low speed (or when stopped). The value will be 19.9μm.

In addition, in the case shown in Figure 3, (coarse) is (fine)
changes to 1UP pulse is output and 10μm is added.
The correct value is 20.1μm. In other words, regardless of whether it is fine or coarse, by increasing → by 1 and → by decreasing by 1, the pitch pulse counter of np will be operated by the signal generated by the 10 μm pulse generation circuit. do,
This is sent to the calculation display circuit.

As is clear from the above explanation, according to the present invention, if accuracy is obtained at the low speed or stopped state where the division counter can operate at the starting point and end point of measurement, there will be an error in the generation of zero-crossing pulses in between. Not only is the final display always corrected so that it is always an accurate value even if it is included, but also high-speed measurement is possible because there is no need for multi-division counting between pitches. This also makes it possible to widen the tolerance range for the scale pitch pulse generation error, which is determined by the setting ranges of the area (coarse) and (coarse) at the 10 μm pitch described above. In other words, if the area is divided into three equal parts, the tolerance is 1/3 of 1 pitch = 3.3 μm, and if determined using the method shown in Figure 6, 1/4 = 2.5 μm, making it possible to manufacture the zero-cross counting pulse generation circuit at low cost. becomes easier,
It also has the characteristics that high accuracy can be expected.

[Brief explanation of drawings]

FIG. 1 is an explanatory diagram showing an overview of the measurement of the present invention, FIGS. 2 and 3 are explanatory diagrams regarding the occurrence of reading errors, FIG. 4 is a block diagram showing the configuration of the present invention, and FIG. FIG. 6 is an explanatory diagram of an example of area setting, FIG. 6 is an explanatory diagram of an example of area setting in coarse system, and FIG. 7 is an explanatory diagram of reading order according to the present invention. 1... Phase modulation division circuit, 5... Digital comparison circuit, 7... Scale signal area division circuit, 8...
Area signal switching circuit, 10...n-digit pulse generation circuit.

Claims (1)

[Claims] 1. For digital scales that output periodic signals in response to displacement, there is a system that counts the number of scale pitches (referred to as a coarse system), and a system that divides and interpolates within one pitch of the scale. 2 of the system that counts the value (selected as the seminal system)
In a counter with a digital display that uses a displacement speed detection signal to switch between using the coarse system at high speeds and the precision system at low speeds, the interval between one pitch of the coarse system scale signal is used as the generation point of the pitch pulse. A scale signal region dividing circuit divides the signal into a region including a pitch pulse, a region immediately before the pitch pulse generation point, and a region in between, and outputs a region signal of . A digital comparator circuit divides the circuit into an area containing the number 0, an area containing the maximum count value, and an area in between, and outputs accurate and coarse area signals. A digital counting device comprising a pulse generating circuit that transmits an up/down signal that increments a scale pitch counter by 1 when moving from one region to another, and decreases the scale pitch counter by 1 when moving from one region to another.