WO1991011679A1

WO1991011679A1 - Method of scanning a workpiece surface

Info

Publication number: WO1991011679A1
Application number: PCT/GB1991/000135
Authority: WO
Inventors: David Roberts Mcmurtry
Original assignee: Renishaw Plc
Priority date: 1990-01-31
Filing date: 1991-01-30
Publication date: 1991-08-08
Also published as: GB9002154D0

Abstract

A method of scanning the surface (S1) of a workpiece (W) using a coordinate positioning machine equipped with a touch trigger probe (20) is disclosed. The probe is driven in a circular path toward, or away from the surface (S1) dependent upon whether or not the stylus (22) of the probe (20) is displaced from its rest position. The radius of curvature of the circular path on which the probe (20) is driven is determined in accordance with an angle between lines joining the last three points on the surface (S1) measured by the machine. The radius of curvature of the circular path is reduced with increasingly acute angles. The speed of the probe is also determined in accordance with the angle between lines joining the last three points measured by the machine; typically as the product of a maximum designated speed and the scalar product between vectors defined by lines joining the previous three measured points.

Description

Method of scanning a workpiece surface

BACKGROUND TO THE INVENTION

1. FIELD OF THE INVENTION

The present invention relates to a method of scanning a workpiece surface in order to obtain data on, for example, the contour of the surface.

Such a scanning operation is usually performed using a coordinate positioning machine (e.g. a coordinate measuring machine or machine tool) having a head moveable with, for example three linear degrees of freedom relative to a table (or bed) upon which the object whose surface is to be scanned is supported. The position of the head relative to a reference position (or datum) on the table is measured by transducers. The head usually carries a measuring probe which is also equipped with one or more transducers which determine the distance between the probe and the surface (and thus the distance between the head and surface) ; thus by summing the machine and probe transducers the position of the surface relative to the reference position can be determined. However, in order to scan (i.e. take many measurements along the surface e.g. to determine its contour or flatness) , the probe must be moved over the surface while all the time remaining close enough to the surface to measure its position, and yet sufficiently far away from the surface to avoid collision.

1_mm DESCRIPTION OF RELATED ART

It is known (e.g. from US 4,769,763) to use an analogue, or measuring probe for scanning. Such a probe typically supports a stylus for movement with 3 linear degrees of freedom relative to the head, and has a transducer for measuring displacement of the stylus in each of three orthogonal directions. A scanning operation according to this teaching is performed by using a vector defined by deflection of the stylus from a null position to determine the direction in which to drive the head.

SUMMARY OF THE INVENTION

According to a first aspect of the present invention, there is provided a method of controlling a coordinate positioning machine to measure the coordinate position of a plurality of points in a designated plane on the surface of a workpiece, the machine having a head moveable relative to a table on which the workpiece is supported, and a probe attached to the head for sensing whether the distance between the head and the surface is greater or less than a threshold distance, the method comprising the steps of:

a) driving the probe into proximity with a part of the surface; b) determining whether the distance between the head and the surface is greater or less than the threshold distance; c) if the distance between the surface and the head is less than the threshold distance, driving the head on a curved path in said plane away from the surface; d) if the distance between the surface and the head is greater than the threshold distance, driving the head on a curved path in said plane toward the surface; e) measuring the position of the surface, during a least alternate transitions of the surface-to-head distance across the threshold distance; and f) reiterating steps (c) to (e)

Preferably, the curved movement of the head (and thus the probe) will have a form which may eventually return the head to its initial position, for example an ellipse or a circle. The head thus describes a "porpoising¹¹ movement along the surface, and the macroscopic direction of probe travel is substantially parallel to the surface. Preferably, the method further comprises the steps of determining a speed for driving the probe as a function of the angle between the last three measured surface points; and driving the probe at the determined speed.

Preferably, the radius of curvature of the circular path with which the head is driven is determined in accordance with the angle between the previous three measured points; the more acute the angle (i.e. the greater the curvature of the surface to be measured) the smaller the radius of curvature of the circular path.

The above method may for example, be implemented using a touch trigger probe as described in e.g. US 4,153,998 which supports a stylus in a rest position. Such a probe automatically outputs a signal as the stylus is deflected from its rest position, and a further signal when the stylus returns to the rest position. Alternatively an analogue touch probe may be used (by adapting the probe to output a signal when a stylus carried by the probe deflects from a rest position by a nominal amount) . The method of the present invention may also be used with an optical probe.

A second aspect of the present invention provides a method of controlling a coordinate positioning machine, to measure the coordinate position of a plurality of points in a designated plane on the surface of a workpiece, the machine having a head moveable relative to a table on which the workpiece is supported, and a probe attached to the head for sensing whether the distance between the head and the surface is greater or less than a threshold distance, the method comprising the steps of:

a) driving the probe in a principal path direction in said plane, and substantially parallel to a part of the surface; b) determining whether the distance between the head and the surface is greater or less than the threshold distance; c) if the distance between the head and the surface is less than the threshold distance, operating the machine to additionally drive the head in a subsidiary path direction in said plane away from the surface; d) if the distance between the head and the surface is greater than the threshold distance, operating the machine to additionally drive the head in a subsidiary path direction in said plane toward the surface; e) measuring the position of the surface, during at least alternate transitions of the surface-to-head distance across the threshold distance; and f) reiterating steps (c) to (e)

Preferably the method additionally comprises the steps of determining the principal path direction upon the basis of the two previously measured points. In one example the principal path direction is parallel to the direction of a vector joining the two previously measured points.

In one example, the movement of the head is perpendicular to the instantaneous principal path direction, and may be an accelerated movement resulting in a curved path of the head toward or away from the surface.

Typically, the initial principal path direction for the scanning procedure outlined above is determined by measuring the coordinate position of two points on the surface of the workpiece, and subsequently driving the probe in the direction parallel to a vector joining the two points, and thus substantially parallel to the surface. BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention will now be described, by way of example, and with reference to the accompanying drawings in which:

Fig 1 shows an example of a coordinate measuring machine carrying a touch probe; and

Fig 2 is a path diagram showing a typical scanning operation according to a first embodiment of the method of the present invention; and

Fig 3 is a path diagram showing a scanning operation according to a second embodiment of the present invention.

DESCRIPTION OF PREFERRED EMBODIMENTS

Referring now to Fig 1, a coordinate positioning machine 10 has a head 12 moveable with three linear degrees of freedom (X,Y,Z) relative to a table 14. The head 12 is movable relative to a carriage 16 in a Z direction, the carriage 16 is mounted on, and movable relative to a bridge 18 in an X direction. The bridge 18 is moveable over the surface of the table 14 in a Y direction. Transducers (not shown) provided on the machine measure the displacement of each of the head 12, carriage 16 and bridge 18 in the Z, X & Y direction respectively relative to a datum, or reference position. A probe 20 is provided on the head, and supports a stylus 22 for movement from a rest position. The probe 20 outputs a step signal when contact between the stylus 22 and the workpiece is detected by the probe as a result of a small deflection of the stylus 22 from a rest position. Such a probe is therefore a "digital" probe having two possible outputs, a "high" output when the stylus is in a state of deflection from its rest position, and a "low" output when the stylus is in its rest position. This type of probe is traditionally used to take point measurements of a surface; the rising edge of the high output from the probe being indicative of the instant of contact between the stylus 22 and the surface, is sent to the machine control which, upon receipt of this signal reads the outputs of the machine transducers. Such a probe is known in the art as a touch-trigger probe and is described in more detail in our US patent 4,153,998.

A method of using such a probe to scan the profile of a workpiece W in a pre-designated X,Y plane will now be described with reference to Fig 2. All movements of the head are made in pre-designated XY plane. Referring to Fig 2, the probe 20 (as viewed along the Z direction, i.e. orthogonal to the X,Y plane) is driven in a direction substantially perpendicular to the surface SI of the workpiece W until the stylus 22 comes into contact with surface at a point Pi. The high output signal from the probe, which occurs as a result of contact between the stylus 22 and the surface at the point PI, is sent to the machine control which subsequently drives the probe 20 along a curved path 24 until the stylus 22 loses contact with the surface SI at point P2. Loss of contact between the stylus 22 and the surface SI is indicated by the output of the probe going"from high to low. Motion of the head 10 of the machine is then braked until the probe 20 comes to a rest at point P3. The probe 20 is then driven back toward the surface SI until it contacts the surface SI once again at or near the point P2, whereupon the output of the probe 20 once again goes high. This time, the high output of the probe 20 is used to brake the motion of the machine until the head 10 of the machine comes to rest at the point P4. The probe 20 is then driven from the point P4 in a direction substantially parallel to a vector defined by a line joining the points PI and P2, since this is a reasonable indication of the initial direction of the surface SI. It is not essential to start the scanning operation in this way; the probe may simply be positioned in proximity with the surface and the routine described below initialised. It is however preferable to initiate the operation thus, since a smoother start-up movement is obtained.

Thereafter, movement of the probe 20 is controlled in accordance with the following routine. At a time interval T seconds ("the update" time) after the probe has been driven from the point P4, the machine control determines whether the stylus 22 of the probe is in a rest position (i.e. the probe output is low) , or whether it is displaced from its rest position (i.e. the probe output is high). If the stylus is displaced from its rest position, then the probe 20 is driven on a circular path in a direction away from (i.e. anti-clockwise as viewed from Fig 2) the surface SI. This interrogation cycle is repeated every T seconds, and since the typical update time for a machine control is of the order of 50 milliseconds, such a cycle will be repeated many times before the stylus 22 of the probe loses contact with the surface SI at point P5, and the probe output once again goes low. The next interrogation cycle after the probe output has gone low determines that the stylus 22 is now in a rest position. The machine control now operates the machine to drive the probe 20 in a circular path toward (i.e. clockwise from above) the surface SI. Once again this interrogation cycle will be repeated many times until the stylus 22 is brought into contact with the surface SI at point P6. A subsequent interrogation of the probe output is once again made and the process is reiterated.

As mentioned above, the rising edge of the high output from the probe may be used to determine a position at which the stylus of the probe contacts the surface relative to the datum of the machine. This is done by sending the output pulse from the probe to an interface (not shown) which causes the outputs of the transducers measuring the position of the head 12, carriage 16 and bridge 18 (relative to the datum) to be frozen. Thus, every time the probe output goes from low to high data in the form of a set of coordinates indicating the position of the surface is generated. It is also possible to acquire data points when the probe output goes from high to low (i.e. the stylus reseats into its rest position) however these have a lower repeatability than points which are determined on the basis of stylus contact with a workpiece (i.e. the stylus unseating and the probe output going from low to high) .

The relative position of the last three points on the surface SI at which the output of the probe 20 goes from low to high (or vice versa) may be used to determine the radius of the circular path on which the probe 20 is to be driven. In general terms, the more acute the angle between the last three measured points (i.e. the more curved the surface) , the smaller the radius of curvature of the circular path will be. This helps to minimise the occurrences of conditions in which controlling the motion of the probe in accordance with the above routine does not provide a relatively smooth movement of the probe 20 along a surface. An example of this variation in path curvature is shown in Fig 2, where at point P9 the stylus 22 loses contact with the surface SI and, in accordance with the routine discussed above, the probe is driven along a circular path in a clockwise direction. Because the point P9 is adjacent a right angle vertex in the surface of the workpiece W between two surfaces SI, and S2, when the stylus 22 eventually comes into contact with the surface S2 of the workpiece, it will be travelling partially in a direction towards to the surface SI. Thus, if the probe 20 is driven towards the point Pll on a circular path of similar radius of curvature to the path linking points P9 and P10, but in an anti-clockwise direction, it is possible that the probe may become damaged because the stylus cannot deflect sufficiently to accommodate motion of the probe along the said path while remaining in contact with the surface S2. To obviate this problem the machine control reduces the radius of curvature of the path of the probe between the points P10 and Pll in accordance with the angle joining the lines P8 and P9, and P9 and P10. The closer such an angle is to a straight line the larger the radius of curvature, whereas the more acute the angle, the smaller the radius of curvature. From the point P12 onwards the radius of curvature of the path of the probe 20 is very similar to that shown for movement of the probe 20 along surface SI, reflecting the linearity between the previous three measured points.

In addition to controlling the radius of curvature of the path, the accelerations which the head of the machine experiences may be reduced by driving the head at a speed which is dependent upon the angle of lines joining the previously measured points. Generally, the speed is reduced with increasing the angle between previously measured points; one specific example is that the speed is determined as a maximum designated speed multiplied by the scalar product of two vectors joining the last three measured points (e.g. vectors joining points P8 and P9, and P10 and Pll) . The advantages of this can be readily seen in the above example, where, applying the above criterion, the probe has a relatively high speed when on a circular path of large radius (e.g. between points P9 and P10) , and a relatively low speed when on a path of small radius (e.g. between points P10 and Pll) .

An alternative method of controlling the machine to achieve a similar result will now be described with reference to Fig 3. Once again, a scanning operation in a pre- designated X,Y plane will be described as viewed in the Z direction with all movements being constrained to occur in this plane. The probe 20 is driven toward the surface si* until the stylus 22 contacts the surface SI* and point PI¹. As described above, at this instant the probe output goes high, and this causes the machine control to drive the probe in an arc until the stylus 22 loses contact with the surface SI' at point P2• ; motion of the head of the machine 10 is then braked and the probe 20 comes to rest at point P3* . The probe 20 is then backed into the surface SI* once again, the stylus 22 contacting the surface SI' at, or very near, the point P3' ; the motion of the head is once again braked to bring the probe 20 to rest at point P4^». The probe is then driven from point P4• in a direction parallel to a vector defined by a line joining the points PI' and P2•; this direction is known as the principal path direction. As described in the previous embodiment, the machine then repeatedly performs an interrogation cycle to determine whether or not the stylus 22 is unseated from its rest position (i.e. whether the output of the probe is high or low) . However, the present embodiment differs from the embodiment described above in that if the probe output is high, as is the case when the probe is being driven from the point P4 in the principal path direction, the machine control introduces a perturbation movement in a direction orthogonal to the principal path direction and in a direction away from the workpiece surface. In this example, the perturbation movement is with an increasing speed up to a pre-designated maximum. The resultant movement of the probe 20 is thus along a curved path, as shown in Fig 3 (N.B. the perturbation movement may also be made at constant speed, resulting in a series of substantially linear paths) . When the stylus 22 loses contact with the surface at the point P5' and the probe output goes low the machine determines a new principal path direction in accordance with a vector defined by a line joining the points P4' and P5'. Furthermore, the interrogation cycle now determines that the stylus has returned to its rest position (since the output of the probe is now low) and decelerates motion of the probe in the previous perturbation direction to zero speed, in order that the probe may be driven with an updated perturbation direction, this time towards the surface SI. The resultant movement is thus a curved path as in Fig 3 between the points P5' and P6' . This procedure is then simply reiterated until the probe passes through the point P9' . Unlike the method of control described in the first embodiment, the control routine described above will not automatically bring the stylus into contact with the surface S2' of the workpiece W' after the probe has passed through the point P9' . Thus, to make the transition from the surface SI' which extends at least partially in a first plane, and the surface S2' which extends in an orthogonal plane, the following routine must be employed. When, in the course of movement of the probe in accordance with the control routine described above, the angle between a line joining the two previously measured points and a line joining the instantaneous probe position Q* to the previously measured point exceeds a predetermined threshold (which in the example in Fig 3 is approximately 25°) , a recovery routine takes over from the usual control routine, and the speed of movement of the probe in the principal path direction is reduced. The probe is then driven along a circular path 30 (or any other curved path which returns the probe to its previous position) until the stylus 22 comes into contact with the surface S2* and the point P10* . Motion of the probe 20 is then braked until the probe comes to rest at the point Pll¹. A new principal path direction defined by a line joining the vectors P9¹ and P10' is then determined, and the probe may once again be driven in accordance with the main routine described above along the surface S2• . If driving the probe along the circular path 30 from the point Q* does not result in the stylus 22 coming into contact with a surface, then the recovery routine simply increases the radius of the path 30 once the probe has returned to the point Q' .

Claims

1) a method of controlling a coordinate positioning machine to measure the coordinate position of a plurality of points in a designated plane on the surface of a workpiece, the machine having a head moveable relative to a table on which the workpiece is supported, and a probe attached to the head for sensing whether the distance between the head and the surface is greater or less than a threshold distance, the method comprising the steps of:

2. A method according to claim 1 wherein the curved path on which the head is driven is a circle.

3. A method according to claim 2 further comprising the step of determining a radius of curvature for the circular path on the basis of an angle between lines joining the last three measured points.

4. A method according to claim 1 further comprising the step of determining a speed for driving the head on the basis of an angle between lines joining the last three measured points.

5. A method according to claim 4 wherein the speed is determined as the product of a maximum designated speed and the scalar product of a pair of vectors defined by lines joining the last three measure points.

6. A method according to claim 1, further comprising the step of initiating the scanning operation by determining the position of two points on the surface, and initially driving the head in a direction in said plane which is parallel to a vector defined by a line joining the previous two measured points.