The present invention relates to precision motion slideways for use in machines and instruments requiring smooth motion with an accuracy of the order of better than 5 nanometres.

Examples of modern technology requiring this order of accuracy are in the manufacture of optical and electronic components. Many of these components are required to have a surface texture exhibiting minimal roughness, and in fact to have surfaces smooth to almost atomic level. There is therefore a requirement for instruments to be able to measure texture to this level.

In a known instrument test pieces are static and a sensor, mounted on a strip spring arm, is moved over its surface. This can give readings to an accuracy of 0.05 nanometres, but unfortunately the sensor moves along an arc of a circle. Due to the geometry of the mechanism accuracy can be maintained only over relatively small arcs. Also it is in many instances a necessity that measurements be made linearly.

Instruments are also known which do operate in a straight line manner but these are mounted on precision roller races or on bulk polytβtrafluoroethylene (PTPE) bearings, and have an accuracy of only 10 to 20 nanometres.

It is well known that the static and dynamic coefficients of friction between two surfaces differ. The Applicant has discovered that for certain materials, when two surfaces of the substance are in sliding contact, the difference between the static and dynamic coeffi- cients of friction decreases as the combined thickness of the two surface layers decreases below a certain level. Also, as this difference decreases so do the fluctuations in the dynamic coefficient of friction. In this specification the coefficients of friction during conventional use 4f the materials will be referred to as the bulk coefficients, whilst the coefficients of friction between thin layers such that the difference between them is reduced will be referred to as the thin layer coefficients.

The Applicant has also discovered that in a sliding system the motion accuracy (that is, the stability in a direction perpendicular to the plane of motion) and the "quietness" of motion (the constancy of forces generated by friction in the system) are related.

As a result of these discoveries the Applicant has been able to design an instrument capable of providing an accuracy of the order of 0.05 nanometres.

According to the present invention a plain bearing slide system includes a carriage mounted on a slide, the carriage being kinematically constrainable to move along an accurately defined path by means inclu— ding bearing pads between the carriage and the slide, characterised in that the surface of each bearing pad is of convex shape, is coated with a thin layer of substantially non-eroding polymeric material having a convex surface, and is arranged to ride along a counterface in the form of a face of a thin layer of substantially non-eroding polymeric material, the combined thickness of the two layers of substantially non-ercding material being such that the difference between the static and dynamic coefficients of friction is considerably less than the difference between the static and dynamic bulk coefficients of friction and that the fluctuation in dynamic coefficient of friction is corres¬ pondingly reduced.

The convex surfaces are preferably sphericalbutmaybe, forexample, ellipsoidal or paraboloidal. The bearing pads are preferably secured to the carriage.

The non-erosive material may be PTFE, in which case the combined thickness of the layers should be such that the difference between the static and dynamic coefficients of friction is no more than 5% (about 25% of the difference between the bulk coefficients). The combined thickness is preferably not greater than between 2 and 3 micrometres, and the radius of curvature of the bearing surfaces is preferably in the range 500 to 50 millimetres.

Other suitable materials might be, for example, high density polyethylene (HDPE) or ultra high molecular weight polyethylene (UHMWPE). One form of slide system has five pads, the accurately defined path then being a straight line. The carriage preferably has a centre of friction at, or acting along a line through which, a force can be applied parallel to the accurately defined path in order to move the carriage. The system is preferably such that when a test piece is mounted on the carriage the centre of gravity and inertia of carriage and test piece coincide with the centre of friction. Where test pieces all have

the same physical characteristics this can be arranged by adjustment of the weight of the carriage, otherwise the carriage can be designed for the addition of weights so that the combined weight and inertia of carriage and test piece can be adjusted. The thin layer of material on the slide is preferably deposited from material on the pads by a running in or pre-preparation method.

According to another aspect of the invention a process of constructing a plain bearing slide system having a carriage mounted on a slide, the carriage being kinematically constrainable to move along an accurately defined path by means including bearing pads between the carriage and the slide, thesurface of each bearing pad being of convex shape and coated with a thin layer of substantially non-eroding polymeric material, each bearing pad being arranged to ride along a counterface in the form of a surface of a thin layer of non-eroding polymeric material, is characterised in including the steps of; selecting a polymeric substance which, at the planned operating conditions of the slide system, has a thin layer difference of static and dynamic coefficients of friction substantially less than the bulk difference and a correspondingly reduced fluctuation in dynamic coefficient of friction, and which is capable of-providing a counter¬ face having a roughness no greater than 5 nanometres Ra; determining a combined thickness of the thin layers which gives a reduced difference in coefficients of friction and a reduced fluctua¬ tion in dynamic coefficient of friction appropriate to the required accuracy of the instrument; establishing limits of radius of curvature of a surface of the thin layer of each bearing pad within which, in use, any variation of interface contact position or area and any inadequacy in averaging-out geometrical imperfections are not too great to prevent the required accuracy from being attained, and within which interface stress remains within acceptable limits; constructing the bearing pads, assembling the carriage, slide and bearing pads and adding the counterface for each bearing pad such that the combined thickness of the thin layers is no greater than that which has been determined to give coefficient of friction characteristics appropriate to the desired accuracy.

The pads may have a plain convex surface, or may be modulated to have a plurality of projections, all the projections having convex surfaces preferably of a common radius. The projections are preferably cylindrical in shape, but may be square, rectangular or any other convenient shape.

Some embodiments of the invention will now be described, by way of example only, with reference to the accompanying diagrammatic drawings of which:

Figure 1 is an elevation of a carriage and slide assembly according to the invention;

Figure 2 is a plan view of the slide;

Figures 3 (a and b), 4 (a and b), 5 (a and b) and 6 (a and b) are similar elevation (a) and plan (b) views of alternative carriage and slide combinations; Figure 7 is an elevation, in section, of a bearing pad according to the invention;

Figure 8 is an elevation, in section, of another form of bearing pad according to the invention; and

Figures 9 to 12 are records of measurements made using a slide system according to the invention.

A carriage and slide assembly (Figures 1, 2) has a carriage 10 mounted on a prismatic form slide 11 having a groove 12 in which a contact member 13, integral with and situated at the centre of gravity and inertia of the carriage 10, travels freely. The carriage 10 rides on the slide 11 which it contacts through five kinematically disposed bearing pads 14 so positioned that the centre of friction lies along the contact member 13. (The positions which the bearing pads 14 take up on the slide 11 are shown in plan view at 15 on Figure 2) . Each bearing pad 14 (Figure 7) has a stiff backing member 20 having a convex backing surface 21 on which is deposited a thin layer 22, of polymeric material such as PTFE, HOPE, orUHMWPE, having a bearing surface 24.

In use the carriage 10 is moved relative to the slide 11 by means of a motive force (not shown) applied to the contact member 13, and as a result some polymeric material is transferred from the layer 22 to the face of the slide 11 to provide a counterface 2$ in the fom of Λ

surface of a thin layer 23 (Figure 7> of the material. This system rapidly stabilises after which there is no significant interchange or removal of polymeric material. The assembly can then be used for measuring a specimen (not shown) placed on the carriage and moved relative to a measuring device (not shown) such as a fine diamond tipped stylus, laser beam, etc. Testing methods, and various measuring methods are well known in the art, form no part of the present inven¬ tion and will not be discussed further.

In practice elastic deformation of the interface between the layer 22 and thin layer 23 on the slide 11 produces a contact area which averages out slide imperfections and reduces interface stresses.

In an instrument as described above with reference to Figures 1 and 2 bearing pads 14 as described above with reference to Figure 7 were formed from a flat sheet of material having a thin steel backing 20 on which was bonded a porous bronze surface 21, a PTFE_based material being bonded on to the porous bronze to form the layer 22. The material was pressed between an accurately formed hard spherical male former and a plastic (nylon or polypropylene) female former to provide a convex circular surface of 5mm diameter having a radius of curvature slightly smaller than was required. The surface 24 of the layer 22 was then machined to a radius of approximately 300 milli¬ metres (slightly greater than the radius of the steel backing 20 and pcrous bronze surface 21) and thickness of 2 to 3 micrometres at the point of contact, using a concave glass file formed by standard optical workshop techniques.

It was found that, after counterfaces 25 had been established on the slide 11 the combined thicknesses of the layers 22 and associated counterface layers 23 was 2 to 3 micrometres, and the instrument would work to an accuracy of 0,05 nanometres. The instrument was thermally and mechanically stable and would therefore work to this accuracy over a long period of time.

Records obtained using the instrument are shown in Figures 9 to 12. Figure 9 shows start up of the instrument after a stylus had been allowed to stand on a polished specimen of silicon carbide for 6 minutes (Trace A), 10 minutes (Trace B) and 15 minutes (Trace C) respectively. It can be seen that there is a transitional rise of no more than 1.5 nanometres. The rise is due partly to specimen indentation and

partly to slide start up characteristics.

Figure 10(a), (b) and (c) show measurements of a very thin oxide growth layer on a silicon wafer over sampling lengths of 0.05, 0.5 and 5 millimetres respectively. Figures 11 and 12 show traces of the sur- face features of two uncoated indium phosphide wafers. One

(Figure 11) had been polished by the conventional alcohol/bromine system, the other (Figure 12) by a new wafer technology process.

Alternative combinations of carriage 10 and slide 11 are shown in Figures 3 to 6, in which Figures (a) are elevations. Figures (b) are plan views, and in which the same identification numbers are used as in Figures 1 and 2.

In the embodiment of Figure 3 the slide 11 is of off-set prismatic form with the groove 12 positioned in the longer of the two sloping sides. The embodiment of Figures 4 has a slide 11 of triangular form with one face of the triangle truncated and the groove 12 in the un-truncated arm. Figure 5 show an embodiment wherein the slide 11 has an -shaped form with the carriage 10 positioned in the L so as to effectively form a rectangle. In this embodiment the motive force is applied to a position 13(a) on the carriage 10. The embodiment of Figure 6 has a slide 11 of rectangular form with a rectangular spine 111 extending upwardly therefrom and the carriage 10 having a channel 110 which encompasses the spine 111. The motive force is, as with the Embodiment of Figure 5, applied to a position 13(a) on the carriage 10. In each of the embodiments of Figures 5 and 6 two bearing pads (14) are positioned between vertically adjacent faces of slide and carriage and restraining means, as indicated at 30, are required to maintain contact of the pad 14 on the associated slide face. The embodiment of Figures 3 to 6 are arrangements well known in the art and operate in a manner similar to that of the embodiment described above with reference to Figures 1 and 2.

In an alternative form of bearing pad 14 (Figure (8)) the backing 20 is formed integrally with the carriage 10, and the convex surface 24 is defined by a plurality of convex surfaces 44,preferably having a common radius,formed at the ends of protruberances 40 of, for example, square or circular section, each tipped with a thin layer 41 of polymeric material.

The thickness of the layers 22,41 and 23 of the polymeric material depends on the accuracy required from the instrument. For PTFE layers a total thickness across two contacting layers (22 and 23 or 41 and 23 for example) of 100 micrometres or more results in a difference between static and dynamic coefficients of friction of 20% or more. There are also fluctuations in the dynamic coefficient of friction of 10% or more. Moreover, on changing from the static to the sliding state there is a displacement, relative to the thickness, of 20 namometres or more. When the thickness has been reduced to 20 micrometres the difference in coefficients is down to about 10% with fluctuations in the dynamic coefficient of 1 to 2%. With a thickness of 2 to 3 micrometres both the differences between the two coefficients and the fluctuation in dynamic coefficient have been reduced to less than 1%, and the displacement on initiating movement reduced to the order of one nanometre. There is no detectable fluctua¬ tion of the displacement during movement, allowing test pieces mounted on a carriage to be measured to an accuracy of the order 0.05 nanometres. Where such a degree of accuracy is unnecessary, of course, a greater thickness can be used. Instruments giving improved accuracy of measure- ment over conventional instruments using roller or bulk PTF_F5 bearings should, however, have a difference between the static and dynamic coefficients of friction no greater than 5%.

It has been found that kinematic restraint of the carriage 10 relative to the slide 11 is essential for motion accuracy. A geo- metrically ideal restraint would involve point mountings, but this is not practical. Each mounting, in the form of a bearing pad 14, is there¬ fore expanded to provide a convex (preferably spherical but alterna¬ tively, for example, paraboloidal or ellipsoidal) bearing surface 2 , 4. In use each bearing surface 24,44 is locally compressed to form a contact area.

The practical limits of the radius of each bearing surface 24,44 are determined by the contact area. The upper limit is set by the stage at which geometrical imperfections in the surfaces 24,44 and layer 23, result in an inconsistency in interface contact position or area. The lower limit is set by the acceptable limit of interface stress (which tends to physically damage the layers 22, 41 and 23), and by the requirement that the contact area should be large enough to

adequately average out the geometrical imperfections. For PTFE bearings of spherical shape and if the type described with reference to Figure 7 these limits have been found to be 500 millimetres and 50 millimetre respectively. Changes in weight of the carriage 10 appropriate to instruments of this style have negligible effect on these limits. Each layer 23 deposited on the slide 11 from a layer 22 of a pad 14 typically has a thickness of less than 5 nanometres. There is a consequent small change in the radius of the bearing surfaces 24,44 which might have to be allowed for in designing the pads 14. During the deposition process there is negligible overall loss of material. During subsequent operation of the instrument the thickness of the layers 22,41 and 23 remains substantially unchanged, counterface roughness being of the order of 0.5 nanometres Ra (as defined in BS 1134) giving a wear rate of less than 10"ϋ (10 nanometres of thickness per kilometre travelled). The upper limit for counterface roughness in an instrument according to the invention is 5 nanometres Ra. Because the loss of bearing material during running in and opera¬ tion is so small, the material is, at least for the use described, substantially non-eroding. it will be realised that there are many alternatives, within the scope of the invention, to some of the details described above. For example the backing 20 of the bearing pad 14 described with reference to Figure 7 may be formed integrally with the carriage 10 or the backing 20 of the bearing pad 14 described with reference to Figure 8 may be formed separately and then bonded to the carriage 10. An alternative method of finishing the surface layer 22, 41 to ^" the desired radius and thickness is by diamond turning. The thin layer of non- eroding polymeric material may be deposited on the slide 11 directly rather than from pads 14, as illustrated at 42 in Figure 8. The combined thickness of layers 41, 42 should be the same as that of layers 22, 23 to give the accuracies obtained with the instrument described with reference to Figures 1, 2 and 7.

It is well known in the Art that static and dynamic coefficients vary with conditions such as temperature and pressure. The above quoted radius relate to a particular PTPE material operated at tempera¬ tures and pressures at which mechanisms of the described type are conventionally used.

It will also be realised that whilst the mechanism described had PTFE as the substantially non-eroding polymeric substance, many other such substances can be expected to be equally effective. Examples are HOPE and UHMWPE. In designing a machine according to the invention a polymeric substance should be selected which has a thin layer difference of coefficients of friction substantially less than the bulk difference at the planned operating conditions, and which is capable of providing a counterface having a roughness no greater than 5 nanometres Ra and preferably as low as 0.5 nanometres Ra. The combined thickness of the bearing pad and counterface layers (22, 23) which gives a difference in static and dynamic coefficients of friction and a reduced fluctua¬ tion in dynamic coefficient of friction appropriate to the required accuracy of instrument should then be established, usually by an empirical method. The limits of radius of curvature of the convex bearing pad surfaces 24, within which any variation of interface contact position or area and any inadequacy in averaging out geometrical imperfections are not too great to prevent the required accuracy from being attained, and within which interface stress remains acceptable, can then be established.

The plain bearing slide system has been described with reference to a five bearing straight line mechanism. However other mechanisms might be constrained to move on other accurately defined paths. For example a four bearing carriage might be mounted by four bearings on a cylindrical slide and constrained to follow, for example, a helical path.

It will also be realised that whilst a plain bearing slide system according to the invention has been described above, with reference to Figures 1 to 8, as used in a measuring instrument there are many other important uses for the invention; for example in precision machining of work-pieces, and in laser interferometers.