JPH01216717A - Nc synchronous control system - Google Patents

Abstract

PURPOSE:To cope with the multiple-kind small-quantity production of gears by time-differentiating encoder outputs of a master shaft, a slave shaft and a work shaft respectively, calculating positional errors between individual shafts, adding or subtracting them, and guiding the results to the second motor. CONSTITUTION:In a gear grinding machine, output signals of encoders 16, 33 and 60 fitted to a master shaft 14 fitted with a tool 12 driven by the first motor 10, a slave shaft 42 and a work shaft 56 are counted by counters 20, 34 and 64 of feed forward, semi-closed loop and full-closed loop control panels 18, 28 and 62 respectively and time-differentiated for each clock signal Ts into speed data. The feed forward command signal Sff from an arithmetic unit 22 and positional errors between the shafts calculated by arithmetic units 30 and 66 are added or subtracted by an adder 26 and outputted to the second motor 32 to control the rotation of the slave shaft 42. The multiple-kind small- quantity production can be flexibly coped with, gears can be processed with high precision.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to NC synchronous control applied to machines that perform synchronous machining by numerical control (hereinafter referred to as NC synchronous machines), such as gear grinders and hobbing machines. Regarding the system - layer details:
The synchronization relationship between the grinding tool attached to the master axis, the workpiece attached to the workpiece axis, and the servo motor attached to the slave axis that drives the workpiece is monitored on a time-sharing basis, and the optimum synchronization relationship is always maintained. The present invention relates to an NC synchronous control system that enables control.

[Background of the Invention] Recently, it has become common practice to mesh a grinding tool, for example, a grinding wheel with spiral protrusions on its circumferential surface, with a gear to be ground, which is a workpiece, and to rotate the grinding wheel shaft and the gear shaft to be ground synchronously. This has led to the adoption of synchronous processing type gear grinding machines that finish the gear to be ground into gears of predetermined dimensions. In this case, in order to finish the gear with the specified dimensions, the synchronization between the grinding wheel shaft and the gear shaft to be ground must be maintained.
That is, it is necessary to process the grinding wheel while keeping the alignment between the protrusions of the grindstone and the valleys of the tooth row of the gear to be ground within a predetermined range. If the gear is machined without being properly maintained in synchronization, various machining errors will occur, such as crooked teeth or a change in tooth pitch.

An example of this type of NC synchronous control device is the device disclosed in Japanese Patent Publication No. 59-35729. That is, the apparatus is provided with a detection means for detecting the rotational position of a hob shaft for cutting gears or a hob shaft drive motor, and based on a position signal generated from the detection means, the workpiece drive motor is It is configured to rotate in synchronization with the hob shaft drive motor.

[Purpose of the Invention] The present invention was made in connection with such technology, and is used in gear grinding machines, hobbing machines, etc., and is based on a master-slave system (rotary position information is acquired by a rotary encoder attached to a master shaft). This is an NC synchronous control system that uses the synchronous rotation ratio and parameter gains (speed error gain, position error gain, etc.). Because it is structured so that all numerical data can be entered, the workpiece specifications (in the case of gears, the number of workpiece teeth, modinir, helix angle, etc.) or the machine specifications (number of tool threads, gear train, decelerator, etc.) The purpose of the present invention is to provide a highly flexible machine tool that can instantaneously switch between the two types and can handle high-mix, low-volume production.

Furthermore, as parameters gain, mechanical system constants (motor specifications, load inertia, viscous resistance, spring constants, etc.) and rotation speed change, it is possible to memorize the optimum state values for each parameter gain, mechanical system constants (motor specifications, load inertia, viscous resistance, spring constants, etc.) It is an object of the present invention to provide an NC synchronous control system that can instantaneously set a gain according to a state.

It is also an object of the present invention to provide an NC synchronous control system that can perform processing of helical gears, etc., since it is also capable of performing differential calculations that define the traverse feed of the traverse shaft.

[Means for Achieving the Object] In order to achieve the above object, the present invention provides a first motor for rotationally driving a tool mounted on a master shaft, a second motor for rotationally driving a slave shaft, and a second motor for rotationally driving a slave shaft. The NC synchronous control system includes a work shaft connected to the shaft by a rotation transmission means and a workpiece pivoted to the workpiece shaft, and the workpiece is ground by the tool, the master shaft, the slave shaft, and the workpiece shaft first to third encoders mounted on shafts; a calculation means for calculating at least a position error between the respective axes after time-differentiating the output signals of the respective encoders; and addition and subtraction of the output signals of the calculation means. and a means for adding and subtracting, and the output signal of the adding/subtracting means is introduced into the second motor.

The present invention also provides a first motor for rotationally driving a tool mounted on a master shaft, a second motor for rotationally driving a slave shaft, a work shaft connected to the slave shaft by a rotation transmission means, and a first motor for rotationally driving a tool mounted on a master shaft. An NC synchronous control system in which the workpiece is ground by the tool, including a workpiece that is attached to a workpiece axis, and a third motor that is attached to a traverse axis that integrally traverses the slave axis and the workpiece axis. The first to fourth encoders attached to the master axis, the slave axis, the work axis, and the traverse axis, and after calculating the time differentiation of the output signals of the respective encoders, at least detecting the position error between the respective axes. The present invention is characterized in that it comprises an arithmetic means for calculating and a means for adding and subtracting the output signal of the arithmetic means, and is configured to introduce the output signal of the adding and subtracting means to the second motor.

Further, in the present invention, when the calculation means calculates the position error calculation between the master axis and other axes, the output signal of the low-resolution encoder among the encoders attached to each axis is multiplied and then the output signal is high-resolution. It is characterized in that it compares with the output signal of the encoder and calculates the difference.

[Embodiments] Next, preferred embodiments of the NC synchronous control system according to the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a schematic configuration diagram of a gear grinding machine to which the NC synchronous control system according to the present invention is applied. In FIG. A tool motor is shown as a first motor that rotates a motor. The tool motor 10 is connected to a grinding wheel 12 via a master shaft 14 which is a tool shaft, and a pulse generator 16, which is a first rotary encoder, is coaxially connected to the master shaft 14. has been done.

In this embodiment, the rotation speed of the motor 10, that is,
Master shaft 140 rotations N. is 3000 rpm, and the resolution of the pulse generator 16, that is, the master axis encoder resolution R, is 60000 p/r (PULSES
/REVOLUTION) There is. The output signal P G r of the Halssian generator 16 is introduced into the feedforward calculator 22 via the 4-week multiplier counter 20 in the feedforward control panel 18 .

The calculation result of the feedforward calculator 22 is sent to the feedforward command signal Sff via the D/A converter 24.
is introduced into the first input terminal of the adder 26 as a signal.

On the other hand, master axis speed data SM, which is the calculation data of the feedforward calculator 22, is introduced into the semi-closed loop calculator 30 in the semi-closed loop control panel 28. In this case, the servo motor 3 as the second motor is connected to the other input terminal of the semi-closed loop computing unit 30.
An output signal PG2 is introduced from a pulse generator 33, which is a second rotary encoder pivotally attached to the rotary encoder 2, via a quadrupling counter 34. Based on this feedback output signal PG, the semi-closed loop arithmetic unit 30
The semi-closed loop command signal Sf2 is introduced into the second input terminal of the adder 26 via the /A converter 38.

The output signal Ss of the adder 26 controls the rotation speed of the servo motor 32 for driving the workpiece via the servo amplifier 40.

In this case, the slave shaft 42 which is attached to the servo motor 32
An inertia damper 44 that absorbs the inertia force of the system and a first gear 46 are coaxially mounted on the other end side via a coupling 43. A second gear 48 meshes with the first gear 46, and the second gear 48 is connected via a shaft 50 to a third gear 52 whose axial direction can be changed, and the third gear 52 is connected to a fourth gear 52. meshes with the gear 54 of. In this case, the fourth gear 54 is fixed coaxially to the work shaft 56, and the work 31 as a gear to be ground is fixedly disposed at one end of the work shaft 56. On the other hand, a pulse generator 60, which is a third rotary encoder, is attached to the other end of the work shaft 56 via a coupling 57, and the output signal Pd3 of the pulse generator 60 is transmitted to the other end of the work shaft 56. The signal is introduced into a full closed loop computing unit 66 via a quadrupling counter 64.

A hysteresis brake 74 is installed on the work shaft 56 via a first pulley 68, a belt 70, and a second pulley 72, and the output signal of the adjusted potentiometer 76 is connected to an amplifier 78.
The braking force is configured to be varied by the voltage signal amplified by.

On the other hand, the quadruple signal of the output signal PG3 of the pulse generator 60 introduced into the full closed loop calculator 66 is compared with the master shaft speed data S, l from the feedforward calculator 22, and then the comparison result is is introduced to the third input terminal of the adder 26 via the D/A converter 79 as a full closed loop command signal S0.

Further, the servo motor 32, the workpiece 31, etc. are placed on a traverse table 80, and the traverse table 80 can be freely moved forward and backward in the six directions of arrows by a driver smoke 82, which is a third motor, via a pole screw (not shown). be. A pulse generator 84, which is a fourth rotary encoder, is pivotally attached to this driver smoke 82, and the output signal PG of this pulse generator 84 is as follows.
On the one hand, it is introduced into the full closed loop calculator 66 via the quadrupling counter 86 in the full closed loop control panel 62, and on the other hand, the quadrupling counter 8 in the semi-closed loop control panel 28.
8 to the semi-closed loop arithmetic unit 30. Therefore, the amount of traverse movement in the six directions of the arrows due to the rotation of the driver smoke 82 is determined by a predetermined differential calculation as will be described later by the full-closed loop calculation unit 66 and the semi-closed loop calculation unit 30, and the calculation result is converted to the D/A. The semi-closed loop command signal SF2 and the full closed loop command signal Sf are transmitted through converters 38 and 79.
1 is added to the adder 26.

Note that a sampling clock T obtained by dividing the oscillation frequency of a crystal oscillator (not shown) is introduced into the clock input terminal CK of the feedforward control panel 18, the full-closed-loop control panel 62, and the semi-closed-loop control panel 28. Ru. Further, in this embodiment, the sampling time 1 of the sampling clock Ts is 300 μs.

Next, FIG. 2 shows the five quadrupling counters 20 shown in FIG.
．． 34.64.86 right and 88 are detailed block circuit diagrams, the quadrupling counters 20.34.64.86 and 88 are respectively pulse generators 16.33.6.
It consists of counters 100a and 100c that count the rising edges of the A-phase and B-phase pulses output from 0 and 84, and counters 100b and 100d that invert the A-phase and B-phase pulses and count the falling edges, respectively. The output data of the counters 100a to 100d is time-differentiated every sampling clock T and converted to speed data, and then added by an adder 108 and converted to quadruple output data.

The gear grinding device to which the NC synchronous control system according to the present embodiment is applied is basically configured as described above, and its operation and effects will be explained next.

First, Table 1 shows the specifications of the workpiece 31 and values such as the resolution of the pulse generators 16, 33, 60, 84, etc. according to this embodiment. These values are manually entered into memories B (not shown) of the feedforward calculator 22, full closed loop calculator 66, and semi-closed loop calculator 30 from input means (not shown).

Table 1 First, we will explain the calculation contents in the feedforward system and position loop system, that is, the fully closed loop control system and the semi-closed loop control system, assuming that the workpiece is a spur gear. In the case of a helical gear, the calculation contents of each control system will be explained in consideration of the amount of movement in the traverse direction.

First, the case of grinding a spur gear will be explained. In this case, the grindstone 12 is rotated by the tool motor 10 at a rotation speed Nx =
When rotated at 300 rpm, the output signal PG as a continuous pulse signal from the pulse generator 16,
is generated, and the output signal PG is introduced into the feedforward calculator 22 via the counter 20 in the feedforward control panel 18. Therefore, when the grindstone 12 as a tool rotates at N)I = 300Orpm, the pulses generated from the pulse generator 16 every sampling time is = 300μs, in other words, the master axis speed data S)I is the following (1) 90 as shown in the formula
0p/sample.

Nil 5M = -X RX X ts (p/sample
e) = -X 60000x 300xlO-' = 9
QQ (p/sample) −
(1) Next, grindstone 1 with rotation speed NK = 300Orpm
2, the workpiece rotation speed Nw when the number of teeth Z of the workpiece 31 is Z=60 is derived from the following equation (2),
Its value will be 50 rpm. Note that in equation (2), reference symbol P is the number of threads of the tool, which in this case corresponds to the number of threads of the grindstone 12, and this value is assumed to be 1 as described above.

= 50 (rpm)...
(2) Next, considering that the reduction ratio Q of the gear train 45, which is the rotation transmission means interposed between the work 31 driven by the servo motor 32, is 24:1 (Q=24). The rotation speed N of the servo motor 32 is the following (
3) It is sufficient to rotate at 120 rpm as shown in the formula.

Ns = Nw x Q (rpm) = 50 x 24 = 1200
(rpm) (3) Therefore, the voltage value to be applied to the servo motor 32 is determined from the rated specification values of the servo motor 32 and the servo amplifier 40.
For example, the rated rotation speed N5. If the rated input voltage V of the servo motor 32 at =300Orpm is V, =6V, it is understood that the number of rotations per IV is 500rpm/V, and in the end, in order to rotate at 120Orpm, It is sufficient to supply 2.4V to the servo motor 32.

This relationship is shown in equation (4).

= −= 500 (rpm/V) ... (4)
In this case, the number of bits of the D/A converter 24 is 12 bits, and the output voltage corresponding to 12 bits is ±IOV, and the voltage 2 for rotating at Ns = 120 Orpm is set.
．． In order to obtain 4V, the value V (D/A24) =2539 may be supplied to the D/A converter 24, as shown in equation (5).

Therefore, the value of the feedforward command signal Stf is given as a value obtained by converting this value V(D/A24) into an analog signal. Note that in this embodiment, the servo amplifier 40 operates as a voltage follower. Therefore, in this state, the grinding wheel 12 has N14 = 300Orpm.
As long as the workpiece 31 is rotating at Nil = 5Orp
Rotates synchronously at m.

However, since errors such as analog offset or drift occur in the D/A converter 24 or the servo amplifier 40, a position loop control system is required. Next, the lid lube control system will be explained.

The position loop control system is a semi-closed loop control system consisting of a semi-closed loop control panel 28 that obtains a converted feedback output signal PG2 from the slave axis 42 to the work axis 56, and a feedback output signal PG directly from the work axis 56. It is composed of a full closed loop control system whose main component is a full closed loop control panel 62.

In this case, a semi-closed loop control system that obtains a feedback output signal PG2 directly from a relatively stable servo motor 32 without backlash errors caused by the gear train 45 is controlled with a high gain, and in order to correct errors in the gear train 45, etc. The full closed loop control system is controlled with low gain. In other words, the full closed loop control system tends to become an unstable loop due to the gear train 450 backlash or the rigidity of the gear train 45 itself.
Since it is difficult to control with high gain, it is used as an auxiliary loop.

Next, a semi-closed loop control system, which is a main loop with high loop gain control, will be described. In this case, in order to obtain highly accurate synchronous rotation, it is necessary to perform accurate position control.

In this synchronous calculation, first, the position output signal PG generated from the pulse generator 16 obtained by the feedforward control system described above is converted into master shaft speed data SM,
Next, the position output signal PG generated from the pulse generator 33 obtained by the semi-closed loop control system is converted into slave axis speed data Ss. Next, the multiplication value (3) IXR5) of the master axis speed data S)I and the resolution Rs of the pulse generator 33, and the slave axis speed data S of the pulse generator 33.

The resolution R)l of the pulse generator 16 and the workpiece 31
The multiplication value (Ss XRMXZ) with the number of teeth Z may be controlled to be the same value. In this case, position error E
p is calculated as shown in equation (6), and the master axis speed data S, l and slave axis encoder resolution ratio Ro = R
If the multiplication value of s/RX is equal to the value obtained by multiplying the slave axis speed data Ss by the master axis encoder resolution ratio R, = 1 and the number of workpiece teeth Z, the master axis 14 and the workpiece axis 56 are completely connected. It is determined that it is synchronized with. On the other hand, if there is a difference between the multiplication values, a value obtained by multiplying the position error Ep by the position loop gain (not shown) is output to the D/A converter 38, so the servo The motor 32 rotates to correct the position error E.

Ep” (Master axis speed data x Slave axis encoder resolution upper limit) - (Slave 1 axis speed data x Master axis encoder resolution ratio x Number of workpiece teeth) = (SxxRx/Rs) - (ssx I XZ) = 90
0X4-(60XIX60)=0-(6) where,
The slave axis speed data S is as follows = 59 (p/s
ample) - (7) Said No. (
6) The calculation result using the formula is the master axis 14 and slave axis 42.
This shows that the two are rotating in perfect synchronization.

Next, let's consider a case where synchronization is out of order and correction must be made. As shown in the above equation (7), the slave shaft speed data Ss is synchronized because Ss = 60, but the rotation of the slave shaft 420 is slightly delayed, and for example, the value of the slave shaft speed data Ss is Ss =5
9 (servo motor 320 rotation speed N, converted to 11
Assuming that the output signal PG, which is only 80 (rpm)), is introduced into the semi-closed loop control panel 28, the position error EpO value will be 60 as shown in the following equation (8).

Ep = (900x 4) - (59x 1 x60
)=60 (p)...(8) Therefore, when this value is multiplied by the position loop gain Kp in the semi-closed loop calculator 30 and introduced into the D/A converter 38, the servo motor 32 speeds up the rotation and reduces the delay. trying to make up for it. On the contrary, if the slave axis 42 is slightly ahead and the position error E becomes a negative value (Ep<0), the output value of the D/A converter 38 also becomes a negative value, and the servo motor 32 operates in the direction of slowing down to normal rotation. Note that the unit of the position error E is [p] = [:ρulse] because of the master axis speed data S. This is because there is virtually no difference even if the slave axis speed data Ss is considered to be a value integrated over the sampling time ts.

Next, position loop control of a fully closed loop control system having low loop gain control will be explained.

First, master axis speed data S that constitutes the full closed loop control system. As mentioned above, S)I = 9
00 CP/sample). On the other hand, the workpiece axis speed data Sw is given by the following equation (9).

= -X324000 x300 x lo-'= 9
7.2 (p/sample) −
t9) Therefore, the master axis speed data SKI and the workpiece axis speed data Sv are compared in the full closed loop calculator 66 and then integrated. And position error E P
2 is calculated and the position error EP2 is multiplied by the position loop gain KP2, and then a full closed loop command signal Sfl is sent to the adder 26 via the D/A converter 79 to perform feedback control. In this case, in order to compare the master axis speed data S)I and the workpiece axis speed data Sw at the same level, normalization calculations are performed as shown in the following equations 01 and 00, and the master glaze normalized speed data S, lS and work axis normalized speed data Sws are calculated.

Therefore, the position error EP2 is given by the following (support) equation.

EP2= f (Sws 5w5) d (samp
le) = Q...@Here, the fact that the value of the position error E p 2 is zero means that the master shaft 14 and the work shaft 56 are rotating synchronously.

As described above, according to the present invention, when calculating the position error EP2 between the master axis 14 and the work axis 56, the output signal PG+ of the pulse generator 16, which is a master axis encoder with a low encoder resolution, is multiplied by a predetermined value to obtain the encoder resolution. Pulse generator 60, which is a high workpiece axis encoder
Arithmetic processing is performed to compare the output signal PG3 with the output signal PG3.

Therefore, highly accurate calculation is possible without reducing the accuracy of encoder resolution. Note that, in the present invention, this principle is also applied to position error calculations between other axes.

In this way, in the case of spur gears, it is possible to accurately rotate them synchronously using the above-described feedforward control system and the position loop control system. By the way, if synchronous rotation is insufficient with only the feedforward control system and position loop control system, the velocity loop control system differentiates the position error Ep and controls it with the velocity error Ev, and the velocity error Ev differentiates and controls the acceleration error. Acceleration loop control system controlled by E, integrates the position error Ep and calculates the position integral error E
It is also possible to add various PID control systems as necessary, such as a position integral error control system controlled by PI.

The above explanation is an explanation of the operation of spur gear machining as the first application example.

Next, as a second application example, an application example of machining a helical gear will be described. In this case, the traverse direction (
It is necessary to detect the amount of movement (in the axial direction of the gear) and apply differential correction to the synchronization relationship. This differential correction may be performed by detecting the amount of traverse movement by a pulse generator 84 mounted on the traverse shaft 83 and performing differential correction on the synchronous rotation using a value obtained by calculation including the workpiece torsion angle β.

Therefore, the diameter d on the pitch circumference of the helical gear is the following (
2) As shown in the formula, it is 165.5 mm (see Table 1 for the meaning of each symbol).

In this case, the deviation angle Y on the pitch circumference due to the workpiece width W = 20 mm is approximately 6.457 as shown in the following α-th equation.
It becomes °.

That is, when traversing the workpiece width W=20 mm, it is necessary to perform a synchronization correction of 6.457° on the pitch circumference of the workpiece. Converting this into the number of feedback pulses per one rotation of the work shaft 560 and calculating the required amount of feedback pulses Pd corresponding to the output signal PC of the pulse generator 84, the following (4305 pulses as shown in equation 10) Similarly, if the required amount of feedback pulses P, + is calculated from the deviation angle Y on the pitch circumference, the same result can be obtained (Equation 00 % Equation % (
) Therefore, the feedback pulse Pd for about 4305 pulses traversing the work width W=20++un is corrected. Actually, the sampling time ts
For example, since the traverse speed is set to 1 mm/sec in this embodiment, the pulse generated per sampling, that is, the traverse axis speed data St, is calculated from the following α-th force equation 3 (p
/sample).

At this time, the differential speed command data Sd is expressed as shown in equation 00.

Si = 1 x Rt X tl = 1 x 10000 x 300 xlO-' =
3 (p/sample)...α, therefore, 0,06457 per sampling time t
Pulse correction will be applied. Since this value has only a decimal part, integer operations cannot be performed on it, so it is normalized by multiplying it by a predetermined value before calculation is performed. In other words, the differential speed command data S,
= 0.06457 is an irrational number, and errors will accumulate, but the predetermined multiple α for normalization should be set to a large value, in this case, so that the differential speed command data S becomes an integer. It has been confirmed that as long as the selection is made, the machining error can be suppressed to a range that does not pose a practical problem when the workpiece width W is within several tens of dimensions. Note that in this embodiment, the predetermined multiple α
There is no problem if the actual value of is about 100,000 or more. The above is an explanation of the operation of helical gear machining as the second application example.

Next, regarding the NC synchronous control system shown in FIG. 1 that operates as described above, the feedforward control panel 18,
The interrelationship between the full closed loop control panel 62 and the semi-closed loop control panel 28 will be described in more detail with reference to the flowcharts of FIGS. 3A-C, 4A-C, and 5. In this flowchart, the letters a, b, and c appended after the reference symbol STP indicating a step correspond to control by the feedforward control panel 18, semi-closed loop control panel 28, and fully closed loop control panel 62, respectively. It is.

Therefore, first, in steps 1a to step IC, the initial data, that is, the data shown in Table 1, is input to the feedforward calculator 22, semi-closed loop calculator 30, and fully closed loop calculator 66, respectively, and predetermined calculations are performed. (STPla to 5TPlc). In this case, the predetermined calculation is performed on items that do not need to be calculated within the sampling time t and can be calculated in advance, such as the number of revolutions per IV of the servo motor 32, resolution ratio R+, Ra, etc.

Next, a position output signal PG is sent from the pulse generator 16 attached to the master shaft 14 to the feedforward calculator 22.
, is the counter 2 in the feedforward control panel 18
0 and multiplied by 4 (STP 2a 5STP
3a), the data output from the counter 20 is time-differentiated by the feed-forward calculator 22 and sent to the grinding wheel 120.
Master axis speed data S corresponding to rotation. (Section (1) above)
(see formula) is calculated (STPda).

Next, a synchronous calculation is executed in the feedforward calculator 22 based on this master axis speed data Sol, and slave axis speed data S is calculated based on the above equation (7) (STP5a).

Master axis speed data S, I obtained in step 4a
is transferred in parallel to the semi-closed loop computing unit 30 (STP (3a)). Next, a synchronous operation based on the above-mentioned (9) is performed to calculate workpiece axis speed data S (STPTa).

Next, the master axis speed data S. is transferred in parallel to the fully closed loop computing unit 66 (STP8
a). Next, the feedforward command signal srr to obtain the predetermined rotation speed N s =1200 of the slave shaft 42 is multiplied by the feedforward loop gain Kt based on the slave shaft speed data Ss calculated in step 5a, and the D/A is introduced into the converter 24 (S
TP 9a, 5TPIOb), D/A converter 24
The feedforward command signal srr, which is the output signal of STP20, is introduced into the first input terminal of the adder 26 (STP20
).

On the other hand, the semi-closed loop control panel 28 firstly
The position output signal PG2 is introduced from the pulse generator 33 attached to the slave shaft 42 and multiplied by 4 by the counter 34 (STP2b 5STP3b).

This counter value is differentiated and converted into slave shaft speed data Ssl corresponding to the semi-closed loop system in the same manner as in equation (7) above (STP4b).

This slave axis speed data SS+ is compared with the slave axis speed data S, which is the synchronous calculation result corresponding to the feedforward system based on the above equation (7) (STP5b
).

On the other hand, the traverse shaft position output signal PG4 from the pulse generator 84 mounted on the traverse shaft 83 is sent to the counter 8.
The semi-closed loop arithmetic unit 30 is multiplied by 4 through 6
The traverse axis speed data 5tl (see equation 071) corresponding to the semi-closed loop is calculated (STP6b 5STP7b). Next, differential calculation shown in the above-mentioned Tomen formula is performed (STP8h).

On the other hand, in step 9b, the feedforward calculator 22
Receive master axis speed data Sll from (STP9
b). Next, in steps 10b to 12b, the position error EP corresponding to the semi-closed loop system is calculated.
Calculate I (STPlob to 5TP12b). That is, the master axis speed data SI in step 10b
4 and the integral value of the traverse shaft differential speed data Stl in step llb are added, and the slave shaft speed data S in step 12b is calculated from the sum of the added values.
The integral value of Sl is subtracted.

Next, in step 13b, the position error EPI is differentiated with respect to time to calculate the velocity error EVI, and in step 14
In b, the position error EPI is integrated to calculate the position integral error EPIt. Next, in step 15b, the position error EPI calculated in steps 10a to 12b is multiplied by the position gain KPI (ST
P13b). Next, in step 16b, the speed gain KV is added to the speed error EVI calculated in step 13b.
I is multiplied by step 14b in step 17b.
The integral gain K is added to the position integral error EPII calculated by
it is multiplied (STP15b to 5TP17b).

The gain multiplication results calculated in steps 15b to 17b are added for each term (STP18b), and this output signal is sent to the second terminal of the adder 26 as a semi-closed loop command signal Sf2 via the D/A converter 38. (STP19b).

Next, the full closed loop control panel 62 receives an output signal PG corresponding to the position from the pulse generator 60 mounted on the workpiece shaft 56 (5TP2c), multiplied by 4 (5TP3C), differentiates it with respect to time, and obtains the workpiece shaft speed data S. , is calculated (STP4c). This work shaft speed data Sw is compared with the synchronization calculation result based on the equation (9) (STP5c). Subsequently, in Step 6C and Step 7C, the full closed loop computing unit 66 operates the pulse generator 84 attached to the traverse shaft 83 in the same way as the semi-closed loop computing unit 30.
After multiplying the output signal PG by 4, the traverse axis speed data St2 corresponding to the full closed loop system is obtained by differentiating the output signal PG.
Calculate (STP7C). Next, the differential calculation shown in equation 01 is performed (STP8C).

On the other hand, in step 9C, the master axis speed data 8.4
is transmitted in parallel from the feedforward calculator 22, and the full closed loop calculator 66 receives the data. Next, steps 10c to 12
Calculate the position error E at 'c (STPloc
to 5TP12C). That is, the integral value of master axis speed data S, I in step 10C and step IIC
The integral value of the traverse shaft differential speed data St2 in step 12C is added, and the integral value of the slave shaft speed data SS2 in step 12C is subtracted from the sum of the added values.

Next, in step 13c, the position error EP2 is differentiated with respect to time to calculate the velocity error Ev□, and in step 14
At C, the position error EP2 is integrated to calculate the position integral error EPI2. Next, each gain constant is multiplied in steps 15G to 17C (S
TP15C to 5TP17C). That is, step 1
Position error E calculated in step 12C at 5C
P2 is multiplied by the position loop gain KPI, and step 1
Speed error E calculated in step 13C at 6C
V2 is multiplied by speed gain Kd2, and in step 17C, the position integral error Ep calculated in step 14C is calculated.
I* is multiplied by integral gain Ki2. The gain multiplication results calculated in steps 15C to 17c are added for each term (STP18C). After that, the added data is sent to the third third of the adder 26 via the D/A converter 79 as the full closed loop tW command signal St+.
is introduced into the input terminal of

Therefore, the adder 26 adds the feedforward command signal Sff, the semi-closed loop command signal Sf2, and the full closed loop command signal S0, and the added data is sent to the slave axis servo motor 32 via the servo amplifier 40.
(STP20 to STP22).

As described above, the feedforward control panel 18, the full-closed loop control panel 62, and the semi-closed loop control panel 28 are organically coupled to perform synchronous calculations.

[Effects of the Invention] As described above, according to the present invention, rotational position information is detected by a rotary encoder attached to the master shaft of a gear grinder or hobbing machine, etc., and the position information is processed to control the slave shaft. In an NC synchronous control system using a method of controlling and obtaining synchronous rotation, the so-called master-slave method, the synchronous rotation ratio, each parameter gain, etc. are all configured to be manually controlled using numerical data. For this reason, the work specifications (
In the case of gears, it is possible to instantly change the number of workpiece teeth, module, helix angle, etc.) or machine specifications (number of tool threads, gear train reduction ratio), etc., and it is possible to configure a flexible machine that can handle high-mix, low-volume production. I can do it.

Furthermore, each parameter gain (speed error gain, position error gain, etc.) memorizes the optimum state value as the mechanical constant (motor, load inertia, viscous resistance, spring constant, etc.) or rotation speed changes. If you keep it open, you can instantly set the gain that suits the machining conditions.

Furthermore, since the structure is such that synchronous calculations can be performed for traverse feed using a traverse table, it is possible to perform helical gear machining and the like with extremely high accuracy.

Furthermore, when calculating the position error between the master axis and workpiece axis or between the master axis and slave axis, the output signal of the axis equipped with an encoder with low precision resolution is multiplied to produce an encoder with high precision resolution. Since the system compares the output signal of the attached axis with the output signal of the attached axis, it is possible to process with high precision without reducing the resolution of the encoder.

Although the present invention has been described above with reference to preferred embodiments, the present invention is not limited to these embodiments.
Of course, various improvements and changes in design are possible without departing from the gist of the present invention.

[Brief explanation of the drawing]

1 is a schematic configuration diagram of a gear grinding device to which the NC synchronous control system according to the present invention is applied; FIG. 2 is a detailed configuration diagram of a quadruple counter in the gear grinding device shown in FIG. 1; 4A to 4C and FIG. 5 are flowcharts for explaining the operation of the gear grinding device shown in FIG. 1. 10... Tool motor 12... Grinding wheel 14...
Master ldl 16...Pulse generator 18...Feedforward control panel 22...Feedforward calculator 24...D/A converter 26...Adder 28...Semi-closed loop control panel 30...・
Semi-closed loop computing unit 31...Work
32... Servo motor 33... Pulse generator 42... Slave axis 45... Gear train 60
...Pulse generator 62...Full closed loop control panel 66...
Full closed loop calculator 74...Hysteresis brake 82...Driver smoke 84...Pulse generator Sx...Master axis speed data S,...Slave axis speed data Sv...Workpiece axis speed data Sl ...Traverse axis speed data

Claims (3)

[Claims] (1) A first motor that rotationally drives a tool that is pivoted on a master shaft, a second motor that rotationally drives a slave shaft, a work shaft that is connected to the slave shaft by a rotation transmission means, and a work shaft that is pivoted on the work shaft. an NC synchronous control system, the workpiece being ground by the tool;
first to third encoders pivotally attached to the master axis, the slave axis, and the workpiece axis; a calculation means for calculating a position error between at least the respective axes after time-differentiating the output signals of the respective encoders; An NC synchronous control system comprising means for adding and subtracting an output signal of the arithmetic means, and configured to introduce the output signal of the adding and subtracting means to the second motor. (2) A first motor that rotationally drives a tool that is pivoted on the master shaft, a second motor that rotationally drives a slave shaft, a work shaft that is connected to the slave shaft by a rotation transmission means, and a work shaft that is pivoted on the work shaft. The NC synchronous control system includes a third motor pivoted on a traverse shaft that integrally traverses the workpiece, the slave shaft, and the workpiece shaft, the workpiece is ground by the tool, and the workpiece is ground by the tool. first to fourth encoders pivotally attached to the master axis, slave axis, work axis, and traverse axis; calculation means for calculating at least a position error between the respective axes after time-differentiating the output signals of the respective encoders; , and means for adding and subtracting the output signal of the calculating means, the NC synchronous control system being configured to introduce the output signal of the adding and subtracting means to the second motor. (3) In the system according to claim 1 or 2, when calculating the position error calculation between the master axis and the other axes, the calculation means outputs an output signal of an encoder having a low resolution among the encoders attached to each axis. An NC synchronous control system is characterized in that after multiplying the signal, the output signal of a high-resolution encoder is compared with the output signal of the encoder, and the difference is calculated.