JPS62297715A - High-resolution absolute value pulse coder - Google Patents

Abstract

PURPOSE:To improve the resolution by generating continuously a signal which indicates an increase or decrease having a period 1/N (N>1) that of a signal indicating the increase or decrease in value to be measured according to the increase or decrease in the value to be measured. CONSTITUTION:An increase/decrease detection part outputs the signal (a) which indicates the increase or decrease in the value to be measured every time the value to be measured reaches a specific discrete value. An absolute value counting part inputs and counts the signal (a) and stores its counted value. An interpolation signal generation part when inputting the signal (a) outputs continuously the signal (c) indicating the increase or decrease with the period 1/N (N>1) the period of the signal indicating the increase or decrease. A high- resolution absolute value counting part reads the stored value (b) out of the absolute value counting part and then inputs and counts the signal (c) in units 1/N units of the stored value (b), thereby outputting the sum of the both. Consequently, resolution which is almost N times is obtained.

Description

Detailed Description of the Invention 3. Detailed Description of the Invention [Field of Industrial Application] The present invention relates to an absolute value pulse coder, and particularly to a high resolution absolute value pulse coder.

[Prior art and problems to be solved by the invention]

Pulse coders are mainly used for precise position detection of moving objects, such as in numerical control (NC). Pulse coders are generally
There are incremental type and absolute type pulse coders, but an absolute value pulse coder has been proposed that has a simple incremental type mechanical part and can essentially function as an absolute type pulse coder (for example, a special Patent application No. 1983-073868, patent application No. 1983
-073871, Japanese Patent Application No. 59-073872).

FIG. 2 schematically shows the configuration of a conventional absolute value pulse coder.

As shown in FIG. 3, the detection unit 1 includes a rotary code plate 111 that is fixed to a rotating shaft that rotates in response to the movement of the object to be measured, and rotates with the rotation of the rotating shaft; It has light emitting elements 112a, 112b and light receiving elements 113a, 113b which are provided oppositely. The rotation code plate 111 is provided with a light-transmitting part and a light-blocking part along the circumference, and as the rotation code plate 111 rotates, the light emitting element 1
Among the emitted light from 12a and 1I2b, the rotation code plate 11
The light that has passed through the transparent part 1 is transmitted to the light receiving element 113a, ll3.
At b, the human phase and B phase signals are generated with a 90° phase shift, and are output through amplifiers 114a and 114b as sine waves DA and D with a 90° phase shift. (Fig. 4 (1), (2)) Here, the phases of DA and DI are:
There is a one-to-one correspondence with the position of the object to be measured. Rotation code plate 1] If 1 is forward rotation, the human phase signal is 90% higher than the B phase signal.
In the case of reverse rotation, the A-phase signal lags the B-phase signal by 90''. (FIG. 4 (1) (2)) The rectangular pulse generation circuit 12 amplifies the above-mentioned DA and Dg and outputs them as rectangular pulses pA and p. (FIG. 4 (1) The absolute value counting unit 13 detects the rotational direction of the rotary code plate 111, that is, the moving direction of the measurement object, based on the sign of the phase difference between the above-mentioned PA and P■, At each rise and fall of pH, if the rotation code plate 111 is rotating in the forward direction, the count value is increased, and in the case of reverse rotation, the count value is decreased. (
FIG. 4(1) Since the absolute value pulse coder normally cooperates with the numerical control device 2 during operation, it receives power from the power source within the numerical control device 2. However, when the numerical control device 2 is stopped, the power is also cut off, and the pulse coder cannot operate depending on the power source.

However, in order to guarantee output signals in absolute format, the above-mentioned pulse coder not only has to maintain the previous state in the absolute value counter 13 even if the power is turned off, but also requires that a moving object be detected while the numerical control device 2 is stopped. When the rotary code plate 111 moves and rotates, the absolute value counter 13 must update the count value with a value corresponding to the movement. For this reason, a backup battery 14 is connected in parallel with the power supply.
is provided so that one of the power sources is always applied, and it can operate as an absolute value pulse coder.

The resolution of the conventional absolute value pulse coder mentioned above is up to 1μ.
m, and measurement with higher precision than this was impossible.

SUMMARY OF THE INVENTION An object of the present invention is to provide an absolute value pulse coder having a resolution that is one step higher than that of conventional absolute value pulse coders.

[Means for solving problems]

In order to achieve the above object, the present invention outputs a signal indicating an increase or decrease in the measurement target value every time the measurement target value reaches a predetermined discrete value in accordance with the increase or decrease in the measurement target value. an increase/decrease detector that counts and stores signals indicating the increase/decrease from the increase/decrease detector; an absolute value counter that counts and stores signals indicating the increase/decrease from the increase/decrease detector; , an interpolation signal generation unit that continuously generates a signal indicating increase/decrease having a period of 1/N of the signal indicating increase/decrease, reads out the memory contents of the absolute value counting unit, and starts the interpolation immediately after the reading is finished. A high-resolution absolute value pulse coder is provided, which includes a high-resolution absolute value counting section that counts the output signal of a signal generation section in units of 1/N of the counting unit of the stored contents and outputs the sum of both.

[For production]

FIG. 1 shows the main components of a high-resolution absolute value pulse coder according to the present invention and the relationships among them.

The increase/decrease detection section outputs a signal a indicating an increase/decrease in the measurement target value every time the measurement target value reaches a predetermined discrete value.

The absolute value counting section inputs the signal a indicating the increase/decrease from the increase/decrease detecting section, counts and stores it.

When the interpolation signal generating section receives the signal indicating the increase/decrease from the increase/decrease detecting section, the interpolation signal generating section generates a signal indicating the increase/decrease having a period of 1/N of the period of the signal indicating the increase/decrease, where N is a number larger than 1. Output C continuously.

The high-resolution absolute value counting section reads out the stored value of the absolute value counting section, then inputs the output signal C of the interpolation signal generating section, and calculates the value in units of 1/N of the units of the stored values. Count and output the sum of both.

〔Example〕

The configuration of one embodiment of the present invention is shown in FIG. The portion within the broken line indicated by 1' is basically the same as the conventional absolute value pulse coder indicated by 1 in FIG.

The ABS LS 113' in 1' corresponds to the absolute value counting section 13 in FIG.
The rise and fall (signal indicating increase/decrease) of PA and Ps (same as I) are manually controlled, and the rise and fall of both PA and
It is counted and stored in a counter (not shown) provided in the LSI 13'. When the numerical control device 2' (hereinafter abbreviated as NC) issues a read command to the ABS LS 113', the ABS LS 113'
In this embodiment, the above stored count data is divided into 4 bytes and stored as PAτ as shown in FIG. 6(1).

Ps? Using the + line, connect the LSI (#2
) Transfer to 4.

However, generally, even during the transfer of the 4-byte count data, the object to be measured moves, and a signal indicating an increase or decrease is input to the ABSLS 113' in the form of a rising or falling edge of PA or PR. At this time, the ABS LS113
' is provided internally in a memory (not shown) separate from the part that stores the above-mentioned count values, and stores information (rising, rising, rising, and the context of their phases).

When the count data transfer divided into 4 bytes is completed,
The ABS LS 113' then converts the above information stored in the memory into the rising edge and falling edge (signal indicating increase/decrease) of two rectangular pulses with a phase shift of 90''.
In the form of ptto, pHo lines, the LSI
(#2) Transfer to 4. (Figure 6 (2)) During this transfer, the measurement target generally moves and the PA
, pH17) The rise and fall are ABS LSI13
However, since the above transfer speed is sufficiently faster than the input speed of P A and P a, eventually all the information stored in the memory will be transferred. Ru. At this point, data transfer from the ABSLS I 13' ends.

From now on, the ABS LS 113' continues to count the inputs of PA and pH using the counter described above until it receives the next read command, while from the outputs of PAO and Pm'i5,
The inputs of Pa and P are output as they are. (
FIG. 6(3)) Also, here, A shown in FIG.
Other outputs TRDT and CUNT from the BS LS113' to the LSI (#2) 4 are control signals for the above-mentioned 4-byte counting data transfer and accumulation signal transfer in the memory, as shown in FIG. , TRDT is "1" during the above-mentioned 4-byte counting data transfer and is "0" at other times, while CUNT is "1" during the above-mentioned 4-byte counting data transfer, and CUNT is "1" during the above-mentioned 4-byte counting data transfer. It is "1" until the transfer of the product signal is completed, and is "0" at other times.

Next, the function of the interpolation circuit 3 is shown in FIG. As shown in FIGS. 7(1) and (2), the interpolation circuit 3 used in this embodiment uses two sine waves DA with a 90° phase shift.
, DB (the output of the detection unit 11' shown in Fig. 3, which is the same as that shown for the conventional absolute value pulse coder in Fig. 4), the phase at that point, and the measurement target. The direction of increase/decrease in value, that is, in this embodiment, the direction of movement of the measurement target is detected, and the period of the period of DA, Dg is determined in response as shown in FIG. 7 (1) and (3). When DA, Dg advances by a phase equal to -), (that is, corresponding to the increase in phase, Da, I)s
90 as shown in Figure 7 (3).
Outputs two phase-shifted rectangular pulses, one rising edge or one falling edge of FPa, FpH. (Here, 90° is F]) a, when the period of FPu is 360°. ) Here, the phase relationship between the FPA and FPI is:
The phase relationship between P and Pg should be the same.

Therefore, between a certain rising edge or falling edge of PA and Ptt corresponding to D, DB and the next falling edge or rising edge,
The rising and falling edges of N FPA and FPI are always output. That is, FPA and FPI have a period of 1/N of Ptt and Ps.

As mentioned earlier in the explanation of the conventional technology, the phases of DA and DI are 1:1 relative to the position of the measurement object (generally the measurement object value).
Therefore, from the above explanation of the function of the interpolation circuit 3', it can be seen that the rise and fall of FPA and FPI are between the measurement target values that conventionally corresponded to the rise and fall of P, , Pl. It can be seen that the output corresponds to the value obtained by dividing the value into N equal parts.

In addition, the values obtained by dividing these measurement target values into N equal parts are determined by the above method of determining the phase based on the peak values of the input sine waves DA and DI, and the changes in the periods of Da and Ds, that is, the measurement target values. is not affected by the rate of change of

By the way, in the interpolation circuit 3 in this embodiment, N=10
It is.

LSI (#2) 4 is the aforementioned control signal, T”RD
T=1. When CUNT=1, ABS LS113
The above-mentioned 4-byte count data output as PAO and PIIO from PAX and PBX are output as they are from PAX and PBX, and sent to the numerical control device 2' through driver 5.

Next, TRDT=0. LS when CUNT=1
I (#2) 4 input PAτ, P80 and output PAX
+P 1llll relationship is shown in FIG.

P'Ao + P m? When a signal indicating an increase/decrease in i' (rising and falling of two rectangular waves whose phases are shifted by 90 degrees) is input, LSI (#2) 4 outputs 1 as shown in FIG.
Rectangular pulses with a phase shift of 90° are generated corresponding to N counts (10 counts in this example as described above) in the numerical control device (NC) 2', which will be described later, in response to the input of signals indicating the increase and decrease. Send rising and falling signals.

Next, when TRDT=0 and CUNT=0,
LSI (#2) 4 is rXJ in FIG. 6 (3)
and in Figure 9 (1) and (2), rX
At the moment when the first rise or fall of the square wave of pao or Pa1r, indicated by J, rYJ, is input, pAo and P
It is detected which one of the mo pulses is leading in phase, and when the PAi phase is ahead of the PIG, as shown in FIG. (The rising and falling edges of two rectangular pulses in which the phase of A phase is 90° ahead of that of B phase) is P AX +
output as P IK, then the P AO* P 8
0 first rise or fall of the square wave (“Y” in Figure 9)
”), the output FP of the interpolation circuit 3 after the output corresponding to
.

Output FPI as is as PAIl and P□.

Conversely, input the rising or falling edge of the first rectangular wave of PAo or Pmci and calculate Pm? from PAir. When the phase of i is leading (r in Fig. 6 (3) and Fig. 9 (1)
As shown in FIG. 9 (1), the rXJO
The output FPA of the interpolation circuit 3 after the output corresponding to the point,
Output FpH as is as PAX, P□.

The numerical control device f(NC) 2' receives the output of the above-mentioned LSI (#2) 4 via the driver 5.

NC2' first reads the 4-hyde count data sent in units of 1 μm.

When NC2' finishes reading the 4-byte counting data, it then sends the above A as FAX and PBX.
Two rectangular waves with a phase shift of 90 degrees from each other, shown in FIG. 8, corresponding to the memory accumulation signal of BS LS113' are input manually, and NC2' outputs these two pulses as shown in FIG. Each time the rising or falling of is detected, if the A phase is ahead of the B phase, and if the B phase is ahead of the +IA phase, it is counted as -1. Note that at this time, the NC2' reads in units of 0.1 μm.゛When the transfer of the signal corresponding to the memory accumulation signal of the above ABS LSI is completed, in LSI (#2) 4, T
In response to RDT=O, C0NT=0, the ninth
Although PAX and PIIX shown in the figure are transmitted, the NC 2' continues to count these pAx and Pax in units of 0.1 μm as in the case of the signals shown in FIG. 8 above.

Then, in NC2', these count values in units of 0.1 μm are added to the 4-byte count data read in units of 1 μm earlier to obtain the desired high-resolution (0.1 μm) absolute value.

The high-resolution pulse coder in this embodiment has a bathoteribus assop 1 in its absolute value pulse coder 1' portion, which is similar to the conventional absolute value pulse coder (FIG. 2 1) described above.
4' (Fig. 5 1'), even if the NC power is turned off, the count data in units of 1 μm is always held in the counter in the ABS LS 113' of the absolute value pulse coder 1'. Even when the NC power is turned off and then turned on again to measure the position, there is no need for troublesome initial position adjustment unlike in conventional incremental pulse coders. Furthermore, for the reason stated in the explanation of the interpolation circuit 3, it also has the advantage of not being affected by the rate of change of the value to be measured.

〔effect〕

According to the absolute value pulse coder according to the present invention, when N is a number larger than 1, it is possible to obtain a resolution N times higher than that of the above-mentioned conventional absolute value pulse coder.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the main components of a high-resolution absolute value pulse coder according to the present invention and the relationship between them. FIG. 2 is a diagram showing an outline of the configuration of a conventional absolute value pulse coder. The figure shows the configuration of the detection units 11 and 11' of the absolute value pulse coders l and 1' in FIGS. 2 and 5, and FIG. 4 shows the configuration of the absolute value pulse coders l and 1 in FIGS. This is a diagram showing the signal waveform and counting timing at ′.
1) is a diagram showing the case when the rotary code plate rotates in the positive direction, and (2) is a diagram showing the case when the rotary code plate rotates in the negative direction. FIG. FIG. 6 is a diagram showing the time course of signals related to the transfer of data and signals from the ABSLS 113' in the configuration of FIG. 5, and FIG. FIG. 8 is a diagram showing the function of the interpolation circuit when the phase difference between DA and D is positive.
(#2) 4, every time the memory accumulation signal transferred from the ABS LS113' rises or falls, the next N
An example of how a signal corresponding to 10 counts in C is output is shown in (1) to (4) when the A phase is ahead of the B phase, and in (5) to (8) when the A phase is ahead of the A phase. A diagram showing the case where the B phase is advanced, and FIG. 9 shows the LSI (#2) in the embodiment of FIG.
Output FAX and P□ in response to the first PA'ii or Pm'ii input after 4-byte counting data transfer and memory accumulation signal transfer are completed. In (1), when phase B is ahead of phase A. (2) is a diagram showing a case where the phase of the A phase is ahead of the phase of the B phase. (Explanation of symbols) l and 1'... Absolute value pulse coder (resolution 1 μm
) 2 and 2'... Numerical control device (NC) 3... Interpolation circuit 4... LSI (#2) 5... Drivers 11 and 11'... Detection units 12 and 12'. ...Rectangular pulse output circuit 13...
Absolute value counting section 13'...ABS LSL14 and 14'...Basoteribasokuasop 111.
...Rotational code plates 112a and 112b...Light emitting elements 113a and 113b...Light receiving elements 114a and 114b...
Amplifier 115: Power supply for light emitting element O-, signal b-m indicating increase/decrease, absolute value count data signal C-, output in response to signal indicating increase/decrease a. Signal sequence showing an increase/decrease with a period of 1 of the period of the signal a (1) Bo:Ot"-m-(2) Tsubaki1(5)FkC
, “0- (6) Fko: O-1f-10 counts
-10 counts (7) P, (11
,1~,. -1-Douichiichi Figure 8 Procedural Amendment (Voluntary) September 2', 1985: / Japan Patent Office Commissioner Kuro [JJ Akio 1, Indication of Case 1986 Patent Application No. 139187 2, Name of the invention: High-resolution absolute value pulse coder 3, Relationship with the person making the amendment Patent applicant name: FANUC Co., Ltd. 4 Address of agent: 8-10-5, Toramon-chome, Minato-ku, Tokyo 105
Drawing to be amended (Figure 8) 6. Contents of amendment The drawing (Figure 8) will be amended as shown in the attached sheet. 7. Attached document inventory correction drawing (Figure 8) 1 copy (1) Culture o
->7 (2) Export 1 (5) PAQ=o (6) FkQ:O-1r
(7),,,1(8),1. ot-11 Figure 8

Claims (1)

[Claims] 1. An increase/decrease detection unit that outputs a signal indicating an increase/decrease in the measurement target value every time the measurement target value reaches a predetermined discrete value in accordance with the increase/decrease in the measurement target value; an absolute value counting unit that counts and stores the signal indicating the increase/decrease from the increase/decrease detector, where N is a number larger than 1, and indicates the increase/decrease in correspondence with the signal indicating the increase/decrease from the increase/decrease detector; An interpolation signal generation section that continuously generates a signal indicating an increase/decrease having a period of 1/N of the signal, reads out the memory contents of the absolute value counting section, and outputs the output of the interpolation signal generation section immediately after the reading is finished. A high-resolution absolute value pulse coder comprising a high-resolution absolute value counting section that counts signals in units of 1/N of the counting unit of the stored content and outputs the sum of both. 2. The high-resolution absolute value pulse coder according to claim 1, wherein the increase/decrease detection section includes a portion that outputs a sine wave having a phase corresponding one-to-one to the measurement target value. 3. The interpolation signal generation section detects the phase of the sine wave output in the increase/decrease detection section, and the phase is 1 of the period of the sine wave output.
Every time /4N advances, the period is 1/N. A high-resolution absolute value pulse coder according to claim 2, which outputs a signal indicating an increase or decrease. 4. Claim 1, wherein the signal indicating the increase/decrease comprises two parallel signals with a phase shift, and the positive/negative of the phase difference between the two signals corresponds to the increase/decrease. The high-resolution absolute value pulse coder according to item 2 or 3.