JPS6125501B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a numerically controlled machine tool for machining a workpiece into a non-perfect circular shape whose external shape changes depending on the axial position, such as a piston in an internal heat engine. The purpose of this invention is to improve the machining accuracy of a workpiece by correcting the positional deviation of a tool machining surface due to thermal displacement of a mechanism or the like during machining. Generally, when machining a non-circular workpiece whose external shape changes depending on the axial position, such as a piston of an internal heat engine, with a numerically controlled machine tool, the first
As shown in the figure, the workpiece is machined by moving the tool spirally along the outer circumferential surface of the workpiece. However, for machining that requires high profile accuracy, such as the master piston used to copy and machine a piston. When machining an object, the feed amount of the tool in the workpiece axis direction relative to the rotation of the workpiece is set low, and the tool moving speed is also set low so that the pitch P of the spiral trajectory drawn by the tool is small. is normal. Therefore, when such a workpiece is machined using numerical control, it takes a long time to complete the process. For example, when machining a 100 mm long workpiece with a helical path pitch of 0.5 mm and a workpiece rotation speed of 5 rpm, it takes approximately
It takes 40 minutes, and it takes several hours or more to complete machining of the workpiece. By the way, in general numerically controlled machine tools, heat is generated from bearings etc. during machining of a workpiece, and the feed mechanism, tool stand, etc. are gradually thermally deformed, so machining of the workpiece may take several hours. In this case, the position of the tool machined surface is changed by thermal displacement, and the final shape of the workpiece is determined by numerical control data (hereinafter referred to as
There is a risk that the shape will be far different from the shape specified in the NC data (called NC data). In particular, the precision of the cross-sectional shape in the axial direction (barrel precision) is a problem for workpieces such as master pistons, etc., but it takes several tens of minutes to an hour to traverse the tool once. Since machining is performed over a long period of time, the position of the tool machining surface changes due to thermal displacement during one traverse of the tool, making it impossible to obtain high barrel accuracy. The present invention has been made in view of this point, and measures the dimensions of a machined surface machined by a tool at every moment at an axial position close to the machining point, and
Feed control that calculates the theoretical dimensions of the machined surface at the measurement position by accumulating the amount of tool movement programmed as numerical control data, and controls the tool feed if a deviation occurs between the two. The present invention is characterized in that a correction pulse is distributed to the means. Embodiments of the present invention will be described below using a numerically controlled grinder as an example. In FIG. 2, 10 is a slide table placed on the bed 11 of the grinding machine, and this slide table 10 can be moved in the left-right direction (Z-axis direction) by a guide surface 11a formed on the bed 11. It is moved in the z-axis direction by a servo motor 12 fixed to the side surface of the bed 11. This slide table 10
A headstock 15 that supports a main spindle 14 that is rotationally driven by a servo motor 13, and a tailstock 17 that supports a tailstock center 16 are placed on the top. A main shaft sensor 18 and a positioning pin 19 are fixed to the portion. With this main shaft sensor 18 and tailstock center 16,
A workpiece W is held, and the positioning pin 19 determines the phase between the workpiece W and the main shaft 14. Further, on a guide surface 11b formed at the rear of the bed 11 and extending in a direction perpendicular to the workpiece axis direction (X-axis direction), a grinding wheel rotated by a motor 20 is mounted as a tool 21. tool stand 22
The tool stand 22 is connected to a servo motor 23 fixed to the rear end of the bed 11 via a feed screw (not shown). On the other hand, a dimension measuring head 24 for measuring the dimensions of the machined surface of the workpiece W is installed on the floor at a position facing the tool 21 while holding the workpiece W therebetween. It is in contact with the outer circumferential surface of the workpiece. The position of the measurement point PC that this contactor 24a contacts is as shown in FIGS. 3a and 3b.
It is located 180 degrees away from the machining point PG in the rotational direction of the workpiece W (C-axis direction), and the machining start point is 1/2 of the helical pitch from the machining point PG in the workpiece axial direction. It is adjusted so that it is located at a position shifted in the direction of . Further, the tip of the contact 24a is formed to have the same shape as the tip of the tool 21, which prevents correction errors from occurring due to mismatch between the shape of the machined surface of the tool 21 and the shape of the contact 24a. ing. The outer diameter dimension of the workpiece W measured by the dimension measuring head 24 is supplied to an AD converter 25 and converted into a digital signal. 30 controls the rotation of the workpiece W and the movement of the tool 21 by distributing pulses to the drive units 26, 27, and 28 that control the rotation of the servo motors 12, 13, and 23, thereby controlling the workpiece W in the NC state. This numerical control device 30 is a numerical control device that processes into a shape specified by data, and this numerical control device 30 includes an arithmetic processing device 31 constituted by a small computer such as a minicomputer, and a numerical control operation to be performed on this arithmetic processing device 31. A storage device 32 stores a control program to be executed and NC data read through a tape reader TR, and a processing unit 31 distributes command pulses to drive units 26, 27, and 28 according to commands. It is configured by a pulse distribution circuit 33. The pulse distribution circuit 33 latches the pulse generator OSC, which generates pulses at time intervals according to the periodic data output from the arithmetic processing unit 31, and the 32-bit pulse pattern data output from the arithmetic processing unit 31. This latched pulse pattern data is sequentially output in synchronization with the pulses output from the pulse generator OSC, thereby
Shift registers SRX, SRC, and SRZ that simultaneously distribute pulses to three axes, C and Z, and a pulse generator OSC
The pulses output from the shift register SRX,
AND gate AG that is not given to SRC, SRZ and shift register from AND gate AG
A detection counter DC detects that 32 pulses are applied to SRX, SRC, and SRZ, and detects that transmission of the pulse pattern set in shift registers SRX, SRC, and SRZ is completed, and an arithmetic processing unit 31 When the pulse distribution start signal PGS is given from the detection counter DC, the AND gate AG is opened, and when the pulse distribution completion signal PGE is given from the detection counter DC, the reset state is entered and the AND gate AG is opened.
It is composed of a flip-flop FF and a closing flip-flop. When the arithmetic processing unit 31 calculates the number of pulses that must be distributed to each axis while the workpiece rotates by a unit angle from the NC data, it creates a pulse pattern corresponding to the number of pulses that must be distributed to each axis. The pattern table shown in Table 1 is searched for each axis and set in shift registers SRX, SRC, and SRZ, and then a pulse distribution start signal PGS is output.
As a result, the set data are simultaneously output from the shift registers SRX, SRC, and SRZ, and three-dimensional interpolation is performed.

[Table] In addition, in this example, the NC data is generated by repeating the profile data in which the final shape of the workpiece is programmed by the amount of movement of the tool 21, the cutting command of the tool 21, and the machining cycle of the workpiece. The operation command for the program and the programmed sequence data are read into the arithmetic processing unit 31 separately. In this embodiment, profile data is input through tape T1, and sequence data is input through tape T2. Arithmetic control device 3
1 reads these data, stores them in the storage device 32, and performs numerical control based on the NC data stored in the storage device 32. The profile data provided by tape T1 is as shown in FIG.
Data for moving in the axial direction is stored for each block. This X-axis data is a program in which the amount by which the tool 21 is moved while the workpiece W rotates by a unit angle is programmed for each rotational angular position of the workpiece, and the number of data corresponding to the size of the unit angle is is programmed for each rotation area. For example, if the unit rotation angle of the workpiece W is 2.5 degrees, 144 pieces of X-axis data will be programmed for each rotation area. In addition, data is programmed at the delimiter of each rotation area to identify the data delimiter and to identify the rotation number of the following data. On the other hand, the sequence data read by tape T2 is as shown in FIG. 5, which shows an example in which rough grinding is performed twice and then fine grinding is performed once to complete the machining. The programs for each grinding process are almost the same; first, the cutting of the tool 21 is performed according to the programmed movement command in one direction of the Grinding of W is automatically performed so that the grinding trajectory by the tool 21 draws a spiral trajectory. A subsequent movement command in the + direction of the X-axis causes the tool 21 to move backward by a predetermined amount, and a subsequent M code 59 returns the tool 21 to its original position in the Z-axis direction. FIG. 6 shows the movement of the tool 21 during this grinding cycle. In this example, the setting unit in the X-axis and Z-axis directions is 0.5 μm per pulse, and the parallel value of the left end surface with the right end surface of the workpiece W as a reference is
Since it is for 200 pulses, the 1st coarse grinding and the 2nd
In rough grinding, a cut of 50 μm is made, and in fine grinding, a cut of 25 μm is made. In addition, the G codes G01 and G02 in the program specify the grinding mode, and when G01 is specified, the pulse distribution speed is accelerated and processing is performed at the rough grinding speed.
When 02 is designated, the pulse distribution speed is slowed down so that machining is performed at a fine grinding speed. In this embodiment, the workpiece W is
The pulse distribution speed is set so that the workpiece W is rotated at a speed of 10 rpm to perform machining, and during precision grinding, the workpiece W is rotated at a speed of 5 rpm to perform machining. These NC data are read by the tape reader TR and stored in the storage device 32, and then when the start switch provided on the operation panel (not shown) is pressed, the arithmetic and control unit 31 is activated as shown in FIG. Run the program shown below. The first two steps 10 and 11 of this program are steps to initialize the internal registers, and in step 10, the angular position register AR that stores the current position of the machining point in each axis direction,
Of the rotation speed register RR, Z-axis position register ZR, and X-axis position register point
Set the distance xi from the PS workpiece W center OW. Also, in step 11, the dimension measuring head 2
Theoretical dimension register that stores the theoretical workpiece dimensions of the measuring point PC whose dimensions are measured by 4.
The TDR is reset to zero, and the measurement point angle position register MAR and measurement point rotational speed register MRR, which store the position of the measurement point PC, are preset to -72 and zero, respectively. Thereafter, in step 12, the sequence data stored in a predetermined storage area of the storage device 32 is read out, and in subsequent steps 13 to 15, the programmed code is determined. If the read code is an X code, the process proceeds to step 16, where the programmed movement amount following the X code is read, and at step 17, the pulses are distributed to the X-axis as many times as required according to the commanded movement amount. This pulse distribution to the X axis is performed by a pulse generator.
This is done by presetting predetermined cycle data in the OSC and repeatedly setting appropriate pulse pattern data in the shift register SRX. As a result, the tool 21 is accurately moved in the X-axis direction by the commanded movement amount. After this, the process moves to step 18, and steps 10 and 11 are performed.
The amount of movement x is added to the X-axis position step XR initialized in step XR and the theoretical dimension register TDR, and the process returns to step 12. Also, if the read code is G
If it is a code, it is determined in steps 19 and 21 whether the code is G01 or G02, thereby determining whether the machining to be performed from now on is rough grinding or fine grinding. Then, in step 20 or step 22, data on the pulse oscillation cycle depending on the type of grinding is output to the pulse generator OSC. On the other hand, if the read code is an M code, it is determined in steps 23 to 25 what the M code is, and various auxiliary operations are performed. Related to this example are MO2, M59, and M
60, other M codes are omitted. If the code read out now is MO2, this indicates the end of the program, so reading out the sequence data is stopped and the numerical control operation is stopped. If it is M59, the process moves to step 26 and the tool 21 is stopped. In order to return the motor to its original position in the Z-axis direction, a predetermined number of pulses are distributed to the drive unit 28 based on the contents of the current position counter, etc. If the read M code is M60, the program jumps to the grinding subroutine CGR shown in FIG. 8, and grinds the workpiece W. First step 30 of grinding subroutine CGR
is a step for determining whether the workpiece W has made one rotation based on whether the stored value of the angular position step AR is 144. If the stored value is 144, steps 31 and 32 are executed. The process then proceeds to step 35, and if the stored value is 144, the process proceeds to step 33. When the process moves to step 31, the memorized value of the angular position step AR is set to zero again, and the rotational speed register is set to zero.
1 is added to RR, and it is stored that the workpiece W has rotated once and reached the angular position of the machining start point. Step 32 is a step for correcting the positional deviation of the tool machining surface caused by thermal displacement of the feed mechanism. In this step, as will be described later, the dimensions of the workpiece measured by the dimension measuring head 24 and this The deviation between the theoretical value of the measurement point whose dimension is being measured by the dimension measurement head 24 is calculated, and if the deviation is 0.5 μm or more, the drive unit 26 is corrected in the positive or negative direction. Send one pulse to correct the tool position. Step 33 is a measurement point angular position register.
In this step, it is determined whether the memorized value of MAR has reached 144. If the memorized value has become 144, the measuring point angle position register MAR is set to zero in step 34, and the measuring point rotation speed register is set to zero.
Add 1 to MRR and proceed to step 35. If the count value has not reached 144, proceed to step 3.
Go directly to step 5. Steps 35 to 37 are steps for performing pulse distribution, in which the X-axis data corresponding to the current position of the machining point is transferred from the profile data stored in the storage device 32 in the first step 35 to the angle position register AR and the rotation speed register RR. Read by referring to the value. For example, if the rotation of the workpiece W is the first rotation and the angular position is at a position rotated by 7.5 degrees from the machining start point, data of X4 in the data area of the first rotation is read out. Step 36 is a routine that searches for a pulse pattern corresponding to the number of pulses distributed to each axis and sets it in shift registers SRX, SRC, and SRZ.The register SRX contains the number of pulses distributed to the X axis read out in step 36. A pulse pattern according to x is set, and the shift register SRC and
Each SRZ has a pulse pattern and tool 21 according to the number of pulses to rotate the workpiece W by 2.5 degrees.
In order to move 3 μm in the Z-axis direction, pulse pattern data corresponding to the number of pulses 25 and 3 is set. Thereafter, in step 37, the pulse distribution circuit 33 sends a pulse distribution start signal PGS to the flip-flop FF to instruct it to start pulse distribution. Incidentally, in step 36, a signal instructing the drive units 26, 27, and 28 in the direction of movement is also output. Pulse distribution start signal to flip-flop FF
When PGS is applied, flip-flop FF is set and AND gate AG is opened, so that the pulses output from pulse generator OSC are applied to shift registers SRX, SRC, and SRZ. As a result, pulses corresponding to the pulse patterns set in the shift registers SRX, SRC, and SRZ are simultaneously output from the shift registers SRX, SRC, and SRZ, and are applied to the drive units 26, 27, and 28, respectively. As a result, the machining point PG of tool 21
is moved spirally on the surface of the workpiece W, and the workpiece W is processed. On the other hand, the arithmetic processing unit 31 executes the program from step 38 to step 41 while such pulse distribution is being performed. That is,
When the operation of step 37 is completed, step 38
, and set the number corresponding to the number of pulses distributed to each axis in the X-axis position register XR, Z-axis position register ZR,
Add it to the angle position register AR and correct the current position memory of the processing point PG. In this embodiment, x is added to the X-axis position register XR, 3 is added to the Z-axis position register ZR, and 1 is added to the angular position register AR. Steps 39 and 40 are steps for calculating theoretical dimension values at the measurement point PC for measuring the dimensions of the workpiece W. In step 39, the measurement point angle is calculated from the profile data stored in the storage device 32. X specified by the stored value of the position register MAR and the stored value of the measurement point rotation speed register MRR.
The axis data is read, and in step 40 the read X-axis data is added to the theoretical dimension register TDR. Then, in step 41, 1 is added to the measurement point angle position register MAR. The measurement point position register MAR contains -72 at the start of numerical control machining.
is preset as the initial value and is incremented by 1 every time the workpiece W rotates 2.5 degrees, so the memory of the measurement point angular position register MAR is stored in the angular position register AR that stores the angular position of the machining point PG. The value will always be 72 lower than the current value. Therefore,
The X-axis data read out in step 39 is stored in a storage area 72 times earlier than the X-axis data read out in step 35 for pulse distribution. Added to dimension register TDR. As a result, the theoretical value of the workpiece dimensions at the measurement point PC located 180 degrees away from the processing point PG in the rotational direction is corrected every time the workpiece W rotates by a unit angle. In step 42, shift register SRX,
In the step of determining whether all the pulse patterns set in SRC and SRZ have been detected, this is determined based on whether the distribution completion signal PGE is sent from the detection counter DC of the pulse distribution circuit 33. When it is determined that the pulse distribution is completed, the process moves to step 44, and it is determined whether the end point has been reached based on the stored value of the Z-axis position register ZR, etc., and if it is not the end point, the process returns to step 30. , if it is the end point, step 1 in Figure 7
Returning to step 2, numerical control is performed based on the next sequence data. Returning to step 30, the angular position register AR
It is determined whether the workpiece W has completed one rotation and reached the angular position to start machining based on whether the count value has reached 144, and it is determined that the workpiece W has not reached the angular position to start machining. In this case, the process moves to step 33 and pulse distribution is performed in the routine from step 35, and when it is determined that the machining start angular position has been reached, the positional deviation of the tool machining surface is corrected in step 32. After that, the process moves to step 35 and pulse distribution is performed. Therefore, every time the workpiece W rotates once, the positional deviation of the tool machined surface is corrected in step 32. In a normal grinding machine, the position of the tool machining surface does not shift beyond the allowable amount during one rotation of the workpiece, so it is sufficient to correct the tool position every time the workpiece W rotates once. can be done. Figure 9 shows a specific example of a routine for correcting the machining surface position of the tool 21, and shows the ideal dimension value at the measurement point PC.
When the deviation Δx between xt and the measured value xm exceeds 0.5 μm, one correction pulse is distributed to the X axis. That is, in step 50, the measured dimension xm of the workpiece W at the measuring point PC is read from the AD converter 25, and the theoretical dimension value xt at the measuring point PC is read from the theoretical dimension register TDR.
The theoretical dimension xt stored in the theoretical dimension register TDR is subtracted from the measured dimension xm read in step 1 to calculate the deviation △x, and in step 52 the absolute value of the calculated deviation △x is Determine whether it is larger than 0.5 μm. And the absolute value of deviation |
When it is determined that △x| is larger than 0.5 μm, a pulse pattern for distributing one pulse to the X axis is set in the shift register SRX in step 53, and whether the deviation △x is positive or negative is determined in step 54. In step 55, a signal instructing the drive unit 26 in the direction of movement is output, and in step 56, a pulse distribution start signal PGS is output. As a result, one correction pulse is output to the drive unit 26, and the position of the tool machined surface is corrected. When it is determined in step 57 that the pulse distribution is complete, the process moves to step 35 of the grinding subroutine CGR shown in FIG. 8, and the position of the tool machined surface is controlled based on the corrected position of the tool machined surface. High-precision machining is performed. In the above embodiment, the tip shape of the tool 21 is formed to have the same shape as the tip shape of the contact 24a of the dimension measuring head 24, and the theoretical and measured dimension values obtained by accumulating NC data are calculated. The amount of positional deviation was easily determined by allowing the deviation to become the positional deviation of the tool machining surface, but if the amount of positional deviation was determined by performing some correction calculations, the tip of the tool 21 could be easily determined. The shape and the tip shape of the contactor 24a do not have to be the same. In the above embodiment, the position is 180 degrees away from the machining point of the workpiece W in the rotational direction of the workpiece W, and is shifted by 1/2 of the helical pitch toward the machining start position in the workpiece axis direction. The dimensions of the workpiece W were measured based on the position, but the theoretical dimensions at the measurement point were calculated by changing the NC data reading position according to the measurement point position. , it can be set anywhere between the machining point and the machining start point. However, if the measurement point is located too far away from the machining point, there will be a time lag between machining with the tool and measurement, making it impossible to accurately compensate for positional deviations on the tool machining surface. It is better to set it near the machining point. Furthermore, in the above embodiment, the dimensions of the workpiece are measured every time the workpiece rotates once, and this is compared with the theoretical dimension value of the measurement point, and only when the deviation between the two exceeds the allowable value The tool position was corrected by one pulse, but if the number of measurements of the workpiece dimensions is increased, for example, the dimensions are measured and the tool position is corrected every time the workpiece rotates by a unit angle. , the accuracy of the cross-sectional profile in the direction perpendicular to the axis of the workpiece can also be improved. Alternatively, the tool position may be corrected by detecting the deviation between the measured dimension and the logical dimension and sending a correction pulse corresponding to this deviation to the drive unit. The tool position may be corrected by calculating the next correction amount from the stored value. Furthermore, in the above embodiment, the theoretical dimension value at the dimension measurement point was calculated by accumulating the NC data, but it is preferable to store the theoretical dimension value at the point where the dimension measurement is performed in the storage device. If you do that,
There is no need to accumulate NC data to obtain theoretical dimension values. As described above, in the numerically controlled machine tool of the present invention, the dimensions of the machined surface machined by the tool are measured moment by moment near the machining point, and
By accumulating the amount of tool movement programmed as numerical control data, the theoretical dimension value of the machined surface at the measurement position is calculated, and if a deviation occurs between the two, the tool is moved in the direction to eliminate the deviation. Since the workpiece is machined by correcting feed, even if the position of the tool machining surface shifts due to thermal displacement of the feed mechanism, this positional shift of the tool machining surface will not occur during machining of the workpiece. This is successively corrected, and the position of the tool machined surface is prevented from shifting beyond the allowable range. Therefore, even when machining a workpiece that requires high precision, such as a piston master, over a long period of time, the tool machined surface will not shift beyond the allowable value due to thermal displacement, and such workpieces will This has the advantage that the accuracy of the barrel can be prevented from deteriorating due to the influence of thermal displacement, and workpieces whose cross-sectional shape changes depending on the position in the axial direction, such as piston masks, can be machined with high precision. In addition, in the present invention, the theoretical dimension value of the machined surface at the measurement position is calculated by accumulating the amount of tool movement programmed as numerical control data. There is no need to set changing theoretical dimensions, which has the advantage of extremely good workability.

[Brief explanation of the drawing]

Figure 1 is a diagram showing a spiral locus drawn by a tool, Figure 2 is an example of a numerically controlled machine tool according to the present invention, and is a diagram in which a numerical control device is added to a plan view of a grinding machine. , b are diagrams showing the positional relationship between processing points and measurement points, Figure 4 is a diagram showing a program sheet with profile data recorded, Figure 5 is a diagram showing a program sheet with sequence data recorded, Figure 6 7 to 9 are flowcharts showing the operation of the arithmetic processing unit, and FIG. 7 is a flowchart showing a routine for decoding commands.
FIG. 8 is a flowchart showing the grinding routine, and FIG. 9 is a flowchart showing the tool position correction routine. 10...Slide table, 11...Bed,
12, 13, 23... Servo motor, 15... Headstock, 17... Tailstock, 21... Tool, 22...
Tool stand, 24...Dimension measurement head, 25...AD
Converter, 26, 27, 28... Drive unit, 30... Numerical control device, 31... Arithmetic processing unit, 32... Memory device, 33... Pulse distribution circuit, 32... Step for correcting tool position, 39
~41... Step of accumulating NC data, 51
... Step for calculating deviation, 52 to 57 ... Step for distributing correction pulses, TR ... Tape reader, W ... Workpiece.

Claims (1)

[Claims] 1 The external shape is determined by rotating the workpiece at a predetermined speed and continuously moving the tool forward and backward in the radial direction of the workpiece in synchronization with the rotation of the workpiece while gradually changing the axial position of the workpiece. This is a numerically controlled machine tool that creates a non-perfect circular shape that varies depending on the axial position, and the numerical control data that sequentially programs the amount of movement of the tool in the radial direction of the workpiece while the workpiece rotates by a predetermined angle is used. Distributes command pulses for moving the tool in the radial direction of the workpiece based on the data storage means and the numerical control data stored in the data storage means, and moves the tool relative to the workpiece axis in a direction parallel to the workpiece axis. pulse distributing means for distributing command pulses to move the workpiece; and tool feeding means for moving the tool using the command pulses distributed from the pulse distributing means to machine the machined surface of the workpiece into a shape commanded by numerical control data. and dimension measurement in which the outer diameter of the workpiece is measured at a position close to the point of machining by the tool in the direction of the axis of the workpiece, and at a measurement position distant from the point of machining by the tool in the direction of rotation of the workpiece. means, a theoretical dimension calculation means for accumulating the amount of movement of the tool programmed as the numerical control data and calculating a theoretical dimension value at the measurement position, and the theoretical dimension calculation means operated at a constant cycle during machining. deviation calculation means for calculating the deviation between the dimension value calculated by the dimension measurement means and the dimension value measured by the dimension measurement means; and the tool according to the deviation calculated by the deviation calculation means. A numerically controlled machine tool comprising a correction pulse distribution means for distributing correction pulses to a feeding means.