GB2397896A - Kinematic mount with array of engaging surface features - Google Patents

Abstract

A kinematic mount comprising a first piece (10) having a plurality of locating points, which may be in the form of a ball bearing (11) and a needle roller bearing (12), and a second piece (13) having an array of surface features (14) for engaging with the locating points. The array of surface features, which may be square pyramid pits, allows the first piece to be engaged e.g. magnetically with the second piece in any one of a plurality of engagement positions defined by the array. A part attached to one of the pieces can therefore be moved in an accurate way between different defined positions without the need for realignment. The kinematic mount can be used for positioning applications in which one piece of the mount is integrated into a sample mounting stage of a positioning apparatus.

Description

TITLE OF THE INVENTION

KINEMATIC MOUNT

BACKGROUND OF THE INVENTION

The invention relates to kinematic mounts and their use in positioning and alignment systems.

Figures lA and 1B show in plan view the two pieces of a conventional kinematic mount. A first piece 1 (Figure 1A) is of a flat rectangular shape and has three V- grooves 2 machined on one of its major faces, aligned at 120 degrees to each other. A second piece 3 (Figure 1B) is also of a flat rectangular shape and has three ball bearings 6 embedded on one of its major faces. The ball bearings 6 are positioned on the second piece such that they will engage in the grooves 2 of the first piece when the two pieces are brought together (see dashed lines in Figure 1A). The first piece 1 can thus be considered female, and the second piece 3 male. It is also known to use hemispherically ended needle rollers instead of ball bearings to provide the same 1 5 function.

Figure 1 C shows an alternative design for the female piece. The same male piece can be used. The female piece 5 has a flat 7, illustrated as being of rectangular area, a triangular pyramid pit 4 and a V-groove 2. This combination provides the same number of constraints as the design of Figure 1A and thus provides the same function.

The above-described kinematic mounts are disclosed in US 5,748,827 [1]. Other two piece kinematic mounts are known in the art, based on similar principles.

Two-piece kinematic mounts are used widely for positioning of optical components, both in production lines and in scientific experiments. Another field of application is in machine shops, where use of a kinematic mount allows workplaces to be removed from milling machines or lathes for inspection, testing or external processing, prior to further machining steps being carried out on the workpiece.

The principal advantage of kinematic mounts is that they allow one piece of the kinematic mount (which will be carrying a workpiece) to be removed from the other piece (which will be fixed to a machine bed, positioning stage etc.) and then replaced at a later time at the same position without realignment. This means that a realignment process for the replaced workpiece can be avoided, thereby saving time and avoiding realignment errors. Depending on the precision of the kinematic mount pieces, a repositioning accuracy of 0.25-10 micrometers is achievable. While this functionality is useful, it is nevertheless limited to replacing the workpiece at its original position.

SUMMARY OF THE INVENTION

According to a first aspect of the invention there is provided a kinematic mounting apparatus comprising a first piece having a plurality of locating points and a second piece having an array of surface features for engaging with the locating points, so that the first piece can engage with the second piece in any one of a plurality of engagement positions defined by the array.

A part attached to one of the pieces can therefore be moved in an accurate way between different defined positions without the need for realignment.

The array can be two-dimensional, so that the engagement positions lie on a two- dimensional grid, or one-dimensional, so that the engagement positions lie along a line. In the case of a two-dimensional array, a square symmetry is preferred. The patterned surface can be conveniently fabricated when the surface features are square pyramid shaped pits. In the case of a one-dimensional array, the surface features can be grooves, wherein V-grooves are convenient to fabricate.

The locating points can be in a variety of forms. They can all be defined by spherical objects, such as ball bearings, or objects with part spherical surfaces, such as hemispherically ended pins. It is also possible to define all the locating points using cylindrically curved surfaces, using objects such as pins or needle rollers, or semicircular cross-section rods. Furthermore, a mixture of these two options can be used, e.g. one ball bearing in combination with one needle roller. Accordingly, the locating points can be defined by at least one spherically curved surface, and/or at least one cylindrically curved surface.

To assist secure engagement of the first and second pieces, the apparatus may further comprise at least one magnet arranged to magnetically clamp the first and second pieces together. Other clamping assistance is also possible, e.g. using vacuum suction.

The kinematic mounting apparatus can be used for a variety of applications. One important application is as part of a positioning system.

Accordingly, a further aspect of the invention provides a positioner comprising a sample mounting stage fur use with a kinematic mount comprising a first piece having a plurality of locating points and a second piece having an array of surface features for engaging with the locating points, so that the first piece can engage with the second piece in any one of a plurality of engagement positions defined by the array, wherein the sample mounting stage includes one of the first and second pieces, the other of the first and second pieces being provided for detachable mounting to the sample mounting stage.

The kinematic mounting apparatus can also be integrated into a wide variety of industrial and scientific processes. One important process relates to aligning a part to a reference position and then moving the part to a different engagement position by picking and placing, either manually or robotically.

Accordingly, a further aspect of the invention provides a method of displacing a part comprising: (a) providing a kinematic mount with a first piece having a plurality of locating points and a second piece having an array of surface features for engaging in the locating points, so that the first piece can engage with the second piece in any one of a plurality of engagement positions defined by the array; (b) securing a part to one of the first and second pieces; (c) securing the other of the first and second pieces to a mount; (d) engaging the first and second pieces; (e) aligning the part to a reference position using external adjustment; and (f) disengaging the first piece from the second piece and reengaging them at a different one of the engagement positions, thereby moving the part by a controlled distance defined by the array of surface features.

BRIEF DESCRIPTION OF THE DRAWINGS

For a better understanding of the invention and to show how the same may be carried into effect reference is now made by way of example to the accompanying drawings in which: Figure 1 A is a schematic plan view of a female piece of a prior art kinematic mount; Figure 1B is a schematic plan view of a male piece of a prior art kinematic mount for engagement in the female piece of Figure 1 A; Figure 1C is a schematic plan view of an alternative prior art female piece to that of Figure 1A; Figure 2A is a perspective view of a male piece of a kinematic mount according to an embodiment of the invention; Figure 2B is a plan view of the male piece of Figure 2A; Figure 3A is a perspective view of a female piece of a kinematic mount according to an embodiment of the invention; Figure 3B is a plan view of the female piece of Figure 3A; Figure 4A is a plan view of an alternative male piece to that of Figure 2A; Figure 4B is a plan view of a further alternative male piece to that of Figure 2A; Figure 5A is a schematic perspective view of a first applications example of a kinematic mount embodying the invention in a first position; Figure SB is a schematic perspective view of the first applications example with the kinematic mount in a second position; Figure 6 is schematic drawing of the system features of the first applications example; Figure 7 is a flow diagram of a process flow that can be carried out in the first applications example; Figure 8 is a schematic plan view of a second applications example using a kinematic mount embodying the invention; Figure 9 is a flow diagram of process flow in the second applications example; Figure 10 is a schematic plan view of a third applications example using a kinematic 1 S mount embodying the invention; and Figure 11 is a flow diagram of process flow in the third applications example.

DETAILED DESCRIPTION

A kinematic mount embodying the invention comprises two pieces. Figures 2A and 2B show a first piece 10 in perspective and plan views respectively, and Figures 3A and 3B show a second piece 13 in perspective and plan view respectively.

Referring to Figures 2A and 2B, the first piece 10 has arranged on one surface thereof a ball bearing l l and a needle roller 12, wherein a plane tangentially passing through the distal surfaces of the ball bearing l l and needle roller 12 lies parallel to and above the surface of the first piece 10 in which these components are mounted. The roller 12 is mounted in a square section groove in the surface, and the ball bearing 11 is mounted in a blind bore in the surface.

Referring to Figures 3A and 3B, the second piece 13 has a square array of surface features 14. These are produced by edge milling the blank to cut two sets of perpendicular V-grooves that coalesce to form the illustrated array of square pyramid shaped surface pits. Parallel grooves cut in one of the directions are still evident at either end of the piece, where no corresponding orthogonal grooves have been machined.

It will be understood that the upper illustrated surfaces of the first and second pieces and 13 can fit together to form the kinematic engagement.

The locating points defined by the bearing 1 1 and roller 12 in the first piece 10 engage tangentially in the sides of the pyramidal pits 14. The bearing 11 will sit in one of the pits, contacting the pyramid sides at four points. The roller 12 will sit in one row of the pits, contacting each pyramid of that row at two points on opposite sides of the pyramid.

By virtue of the array, it will be understood that the first piece can be ratcheted along in increments of one pyramid along either one of the two orthogonal directions in which the pyramids extend. More generally, the first piece can be removed and replaced at any one of a large number of engagement positions defined by the two dimensional array of surface pits. Moreover, since the dimensions of the array are known, the distance between the different engagement positions is known. Therefore, the relative position of one engagement position to another can be known to whatever degree of accuracy the first and second pieces are manufactured to. A high degree of accuracy of relative position of the pieces can thus be ensured. Precision relocation of parts over a two-dimensional grid of points can thus be achieved using simple mechanical means.

Figure 2B is a plan view of the male piece of Figure 2A; Figures 3A & 3B is a perspective view of a female piece of a kinematic mount according to an embodiment of the invention; Figure 3B is a plan view of the female piece of Figure 3A; Figure 4A is a plan view of an alternative male piece 10 in which three ball bearings 11, 15 and 16 are used to define the locating points. Other spherically curved surfaces could be used, for example hemispherical ends of pins.

Figure 4B is a plan view of a further alternative male piece 10 in which two cylindrical pieces 12 and 17, such as needle rollers, are arranged at right angles to each other to define the locating points. Other cylindrically curved surfaces could be used, for example, semicircular section rods.

While the kinematic mount according to the invention allows incremental displacement of a workpiece from a reference position using the tessellating grid of the female piece, it will be appreciated that the standard functionality of a conventional single position kinematic mount is retained. Namely, it is still possible to remove a workpiece, e.g. for testing at a remote site, and then replace it in the same position without having to refind the reference position. The design of the invention thus retains the advantages of conventional kinematic mounts as well as providing the additional functionality of easy precision movement over a pre-defined one- or two- dimensional array of grid points lying in a positioning place.

Engagement of the female and male pieces occurs by gravity only, provided that the i pieces are generally horizontally aligned. This is the simplest fomm of engagement and I will be sufficient for many applications. However, in some applications, stronger engagement will be needed.

To provide stronger engagement of the female and male pieces, pemmanent magnets can be inserted in the female piece 13. The male parts 11, 12 of the male workpiece 10, and/or the whole main body 10, can be made of ferro- or pare-magnetic material which will be attracted by the magnets. Standard ball bearings and needle rollers are made of magnetic steel. To fit the magnets, holes can be machined into the underside of the female piece (as viewed in Figures 3A and 3B) into which cylindrical shaped pemmanent magnets can be inserted, e.g. rare-earth magnets. In principle, electro- magnets could be used instead of pemmanent magnets.

Another engagement option is to use vacuum suction, in which case one or more air passages would be included in either the male or female piece venting into the engagement surface. A flexible skirt around the periphery of one of the pieces could be added to assist generation of a sufficient vacuum.

The above embodiment has a square pyramid pattern for the female piece. This is convenient to machine using standard milling techniques, since it can be achieved with two edge cutting actions. However, in principle any tessellating two dimensional pattern would be suitable. Square or rectangular symmetry patterns are generally preferred, since this provides compatibility with standard positioning software which generally assumes a Cartesian coordinate system. However, in principle, triangular, hexagonal or other symmetry patterns could be used. Moreover, instead of milling, the pattern could be generated by other manufacturing techniques, for example moulding or stamping.

It will also be appreciated that the mating surfaces may be androgynous or bisexual.

For example, the piece 10 could have a pyramid array matched to the pyramid array of i piece 13 so that the two engage. This is however less elegant, since such an I arrangement would be overrefined.

Applications Examples

1 S Example 1

A first applications example is now given. The application relates to alignment and testing of an optical fibre, laser source and solid-state waveguide structure.

Figure SA is a schematic perspective view of the principal optical and mechanical components used in conjunction with the kinematic mount. A solid-state waveguide structure 17 having an input side to be aligned with an output end of an optical fibre 18 leads from a laser source 19. The output end of the optical fibre 18 is encased in a ferrule 20 fixedly attached to the unstructured side of the female piece 13 of the kinematic mount. The female kinematic mount piece 13 is in engagement with the male piece 10, which is in turn secured to a positioning stage 21. The positioning stage 21 is positionable by respective positioning elements thereof (not shown) in three orthogonal linear axes X, Y and Z. The plane defined by the axes X and Y is generally perpendicular to an optical axis of the optical fibre 18 defined by the bore of the ferrule 20, wherein the optical axis is aligned with the Z positioning axis. (In alternative - 1 1 examples a 6-axis positioner with three rotational and three translational degrees of freedom could be used.) The solid-state waveguide structure 17 is mounted on a fixed plinth 22 with the input of the waveguide of the waveguide structure 17 positioned adjacent to the output end of the optical fibre 18 held in the ferrule 20. The solid- state waveguide structure 17 has an output leading to an optical detector (not shown in this figure), for example a p-i-n detector.

Figure 5B is generally the same as Figure 5A but shows the female kinematic mount piece 13 displaced backwards in the Z direction by a selected number of grid repeats.

This displacement can be achieved manually by an operator. The operator can remove the female piece 13 using thumb and forefinger and then displace it back in the Z direction by a desired number or periods. The groove periods are sufficiently large to allow their counting by eye. Alternatively, the displacement could be carried out in an automated fashion by a pick-and-place robot, which only needs to have a relatively coarse placement accuracy of better than half the grid period. Lateral movement in the X direction would of course also be possible.

Figure 6 is a schematic block diagram showing a complete positioning system, including the position control components, as well as those components shown in Figures 5A and 5B. The positioning stage 21 is connected to a positioner controller 24 with electrical leads 25 for conveying control signals, one for each of the positioning axes X, Y and Z. The controller 24 is also connected to receive a signal from the optical detector 23 through an electrical lead 26.

The controller 24 is operable to output control signals to the positioning elements associated with the various positioning axes during signal optimization. Signal optimization is based on the optimization of the signal received by the controller 24 from the optical detector 23 through the electrical lead 26. The electrical lead 26 thus serves to close a feedback loop under which the controller 24 drives the individual positioning elements of the positioning stage 21 for the various axes. A signal optimisation or signal search process may conveniently be generated by a software module which can run on a personal computer 27 interfaced with the controller 24 through an interface connection 28. Signal strength and other data relating to the status I of the current process can be displayed on a monitor 29, and the process can be controlled by means of a keyboard 30, computer mouse 31 or other human interface device.

It will thus be appreciated that movement of the workpiece, i.e. the ferrule 20 carried by the female kinematic mount piece 13 in this example, is carried out by using the positioning stage 21 for continuous fine manipulation, and manual or automated repositioning of the female kinematic mount piece relative to the male piece for incremental coarse manipulation in the XZ plane.

Figure 7 shows the flow of an example process carried out using the components illustrated in Figures 5A, 5B and 6. The process is a manufacturing process to test a batch of laser sources 19 packaged with pigtailed optical fibre 18 terminating in the ferrule 20.

In Step S 1, a first one of the laser modules is secured to the female piece 13.

In Step S2, the female piece 13 is arranged on the male piece.

In Step S3, the ferrule 20 is moved using the positioning stage 21 under control of the controller 24 to align the output of the optical fibre with the waveguide in the solid-state waveguide 17 in the XY plane, and to but the end of the ferrule 20 against the end face of the waveguide 17 in the Z direction. This defines a reference position, and I corresponds to the position illustrated in Figure 5A.

In Step S4, the female kinematic mount piece 13 carrying the ferrule 20 is picked off the positioning stage 21 and moved back in the Z direction by a certain number of grid increments to an offset position, as illustrated in Figure 5B. As indicated by the dashed lines in the flow diagram, between picking and placing the ferrule, it can be subjected to an external process in Step 4.5. An external process might simply be a manual visual inspection, or investigation using a microscope, perhaps at a different location.

In Step S5, the throughput of the waveguide is measured with the ferrule in its offset position using the detector 23. Test data is thus acquired by the PC 27.

In Step S6, the laser module is removed from the kinematic mount and placed in an output bin, since testing is now complete.

In Step S7, if there are more laser modules to test the process flow returns to Step S1.

Otherwise, the process is finished.

Example 2

Figure 8 is a schematic plan view of the optical and mechanical components of a second applications example. A light source 40, such as a laser diode or light emitting diode, emits light which is coupled into an optical fibre 41, the other end of which is secured using a fibre block on one piece of a multi-position locatable kinematic mount 43 according to the invention, which is in turn secured on a multiaxis positioner 42.

Arranged in front of the output of the optical fibre 41 is a long-travel single axis linear positioner (usually referred to as a linear stage) having integrated on its upper surface one piece of a conventional singleposition kinematic mount (hidden in the figure).

The other piece of the kinematic mount, the upper piece, forms a multimount tray 44 having berths for mounting several waveguide modules 45 in set positions spaced apart in the x axis by a fixed offset period P, but in the same position with respect of the y axis, and of course the z axis.

A further multiaxis positioner 48 is arranged on the other side of the linear stage and has mounted on it a multi-position kinematic mount 46 according to the invention carrying a further optical fibre 47. The further optical fibre 47 is arranged on a fibre block to receive light emitted from the waveguide modules 45 and direct it onto a detector 49, such as a photodiode.

Figure 9 is a flow diagram of process flow using the apparatus of Figure 8. Control can be broadly similar to that described in the first applications example.

In Step S21 an operator manually picks several workpieces, e.g. waveguide modules.

In Step S22, the workpieces are placed into berths in the kinematic multimount tray 44.

In Step S23, the populated tray is placed onto the positioner 50 so that the two parts of the kinematic mount engage. An end one of the workplaces is then aligned to provide efficient coupling. This is done in two steps. First the operator manually positions the fibre blocks by eye into approximate alignment with the input/output waveguides of the workpieces using the multi-position kinematic mounts 43 and 46. Second fine adjustment is carried out to maximise light throughput under an automated feedback control algorithm that optimises the alignment using the positioners 42 and 48.

In Step S24 a test run is carried out in which each workpiece 45 is tested in turn, with the long travel stage 44 being used to move between workpieces.

In Step S25, after completion of the test run, the tested workplaces are removed.

In Step S26, control of the process is returned to Step S21 if there are more workpieces to test, otherwise the process is finished.

It is noted that preparation of another multi-mount tray can be carried out by the operator while the testing is running in Step S24. Another advantage of this arrangement is that it can handle modules of different width. For example, silica or semiconductor waveguide modules come in a variety of widths. This can be accommodated in a simple manner in the illustrated example by moving the upper parts of the kinematic mounts 43 and 46 in the y direction, thereby avoiding having to remount the positioners 42 and 48 on the breadboard, bearing in mind that precision positioners will usually have quite a limited travel.

Example 3

Figure 10 is a schematic plan view of the optical and mechanical components of a third applications example. A first sub-assembly comprises a light source 51, such as a laser diode or light emitting diode, emits light which is coupled into an optical fibre 52, the other end of which is secured to an upper piece 55 of a kinematic mount, the lower piece 54 of which is arranged on a multiaxis positioner 53.

A second sub-assembly carries a second optical fibre 59 facing the output end of the first optical fibre 52. The second optical fibre 59 is mounted on an upper piece 57 of a second kinematic mount, the lower piece 58 of which is arranged on a second multiaxis positioner 60. The second optical fibre 59 leads to a detector 61, such as a photodiode, arranged to receive light emitted from the end of the second optical fibre i 59.

A control fixture 56 is arranged between the two fibre-mounting subassemblies and has I a locating berth 50 for mounting samples. The berth 50 is vacant in the illustration.

Figure 11 is a flow diagram of process flow using the apparatus of Figure 10. Control can be broadly similar to that described in the first applications example.

In Step S31, it is decided whether a reference measurement is needed. If no, a test or work sequence is initiated. If yes, the following process steps are carried out.

In Step S32, the optical fibres 52 and 59 are shifted forward, i.e. towards each other, in their kinematic mounts, moving from Position 1 (dashed lines in figure) to Position 2 (solid lines in figure).

In Step S33, a reference measurement is performed. Namely, the optical fibres are brought into alignment under an automated feedback control algorithm that moves the positioners 53 and 60 so as to maximise optical throughput.

Is Step S34, the optical fibres are moved back from Position 2 to Position 1, again using the kinematic mounts. ! Following completion of this procedure for obtaining e reference measurement, a test or work sequence can be initiated by placing a test component, such as a wafer or module onto the berth 50, wherein the optical path in Position 1 will be through a waveguiding part of the test component, rather than directly between the fibres as in Position 2.

The above processes are described by way of example only. Many other processes can be envisaged where the ability to accurately move over a predefined grid of points It will be appreciated that the female and male pieces can swap position without any change in functionality. For example, in the applications example, the female piece could be secured the positioning stage and the male piece to the workpiece (ferrule 20).

Moreover, if single axis adjustment only was needed, then the twodimensional structuring of the female piece (square pyramids in the main embodiment) could be changed to a one-dimensional structuring (e.g. grooves).

It will also be appreciated that the kinematic mount can be used on a fixed mount and does not necessarily have to be mounted on a positioner. Fine motion control, if required, can be carried out by a positioner that is separate from the kinematic mount.

For example, in the second applications example, the long travel stage could be substituted with a kinematic mount according to the invention having a one-dimensional array of surface features, e.g. V-grooves, that allow stepwise translation of the multi mount tray 44 in the x direction, where the kinematic mount grid has a grid period that is a fraction of the sample spacing period P.

RENCE

REFE

[1] US 5,748,827 l - 1 8

Claims (14)

1. A kinematic mounting apparatus comprising a first piece having a plurality of locating points and a second piece having an array of surface features for engaging with the locating points, so that the first piece can engage with the second piece in any one of a plurality of engagement positions defined by the array.

2. The apparatus of claim 1, wherein the array is two-dimensional, so that the engagement positions lie on a two-dimensional grid.

3. The apparatus of claim 2, wherein the two-dimensional array has square symmetry.

4. The apparatus of claim 1 or 2, wherein the surface features are pyramid shaped pits.

5. The apparatus of claim 4, wherein the pyramids are square pyramids.

6. The apparatus of claim l, wherein the array is one-dimensional, so that the engagement positions lie along a line.

7. The apparatus of claim 6, wherein the surface features are grooves.

8. The apparatus of claim 7, wherein the grooves are V-grooves.

9. The apparatus of any one of the preceding claims, wherein at least some of the locating points are defined by at least one spherically curved surface.

10. The apparatus of any one of the preceding claims, wherein at least some of the locating points are defined by at least one cylindrically curved surface.

11. The apparatus of any one of the preceding claims, further comprising at least one magnet arranged to magnetically clamp the first and second pieces together.

12. A positioner comprising a sample mounting stage fur use with a kinematic mount comprising a first piece having a plurality of locating points and a second piece having an array of surface features for engaging with the locating points, so that the first piece can engage with the second piece in any one of a plurality of engagement i positions defined by the array, wherein the sample mounting stage includes one of the first and second pieces, the other of the first and second pieces being provided for detachable mounting to the sample mounting stage.

13. A method of displacing a part comprising: (a) providing a kinematic mount with a first piece having a plurality of locating points and a second piece having an array of surface features for engaging in the locating points, so that the first piece can engage with the second piece in any one of a plurality of engagement positions defined by the array; (b) securing a part to one of the first and second pieces; (c) securing the other of the first and second pieces to a mount; (d) engaging the first and second pieces; (e) aligning the part to a reference position using external adjustment; and (f) disengaging the first piece from the second piece and reengaging them at a different one of the engagement positions, thereby moving the part by a controlled distance defined by the array of surface features.

14. A kinematic mount substantially as hereinbefore described with reference to Figures 2A, 2B, 3A and 3B; Figure 4A; or Figure 4B of the accompanying drawings.

14. A kinematic mount substantially as hereinbefore described with reference to Figures 2A, 2B, 3A and 3B; Figure 4A; or Figure 4B of the accompanying drawings. l - %ú)

Amendments to the claims have been filed as follows 1. A kinematic mounting apparatus comprising a first piece having a plurality of locating points and a second piece having an array of surface features for engaging with the locating points, so that the first piece can engage with the second piece in any one of a plurality of engagement positions offset from each other defined by the array.

2. The apparatus of claim 1, wherein the array is two-dimensional, so that the engagement positions lie on a two-dimensional grid.

3. The apparatus of claim 2, wherein the two-dimensional array has square symmetry.

4. The apparatus of claim 1 or 2, wherein the surface features are pyramid shaped 1 5 pits.

5. The apparatus of claim 4, wherein the pyramids are square pyramids.

6. The apparatus of claim 1, wherein the array is one-dimensional, so that the engagement positions lie along a line.

7. The apparatus of claim 6, wherein the surface features are grooves.

8. The apparatus of claim 7, wherein the grooves are V-grooves.

9. The apparatus of any one of the preceding claims, wherein at least some of the locating points are defined by at least one spherically curved surface.

10. The apparatus of any one of the preceding claims, wherein at least some of the locating points are defined by at least one cylindrically curved surface. A\

11. The apparatus of any one of the preceding claims, further comprising at least one magnet arranged to magnetically clamp the first and second pieces together.

12. A positioner comprising a sample mounting stage for use with a kinematic mount comprising a first piece having a plurality of locating points and a second piece having an array of surface features for engaging with the locating points, so that the first piece can engage with the second piece in any one of a plurality of engagement positions offset from each other defined by the array, wherein the sample mounting stage includes one of the first and second pieces, the other of the first and second pieces being provided for detachable mounting to the sample mounting stage.

13. A method of displacing a part comprising: (a) providing a kinematic mount with a first piece having a plurality of locating points and a second piece having an array of surface features for engaging in the locating points, so that the first piece can engage with the second piece in any one of a plurality of engagement positions offset from each other defined by the array; (b) securing a part to one of the first and second pieces, (c) securing the other of the first and second pieces to a mount; (d) engaging the first and second pieces; (e) aligning the part to a reference position using external adjustment; and (f) disengaging the first piece from the second piece and reengaging them at a different one of the engagement positions, thereby moving the part by a controlled distance defined by the array of surface features.