US20010006421A1 - Device for measuring translation, rotation or velocity via light beam interference - Google Patents

Abstract

The device for measuring translation, rotation or velocity includes at least a light source, a light detector, a first grating and a second grating, the first grating being mobile relative to the second grating. A incident beam reaches the first grating where it is diffracted in two beams whose directions are interchanged by the second grating, the resulting beams being then again diffracted by the first grating in an output diffraction direction where they interfere together. Both gratings are used in reflexion.

Description

• The present invention concerns a device for measuring translation, rotation or velocity via interference of light beams diffracted by diffraction gratings which are substantially parallel to each other.[0001]
• European application 0 672 891 discloses a device for measuring relative displacements between a head unit and a scale. This device is of the type where all diffraction gratings have the same spatial period or pitch P. The head unit has a light-emitting element (source), a cylindrical lens to condense the light beam provided by the source and a first diffraction grating used in transmission for splitting the light beam. The resulting diffracted beams fall onto a second grating arranged on the scale where they are diffracted in reflexion. The head unit further comprises a third grating used in transmission for mixing the diffracted beams coming back from the scale and a light-receiving element (photodetector). In all embodiments, the source and the photodetector are spatially separated respectively from the first and third gratings so that the head unit has relatively large dimensions. The distance between the mixing grating and the photodetector is actually needed because there is a plurality of interfering beams coming out of this mixing grating. Further, it is to be noted that for each diffraction event, at least one diffracted beam is not used. The unused diffracted beams represent a loss of light power, generate noise, and may lead to spurious interferences. The efficiency of such a measuring device is thus relatively low.[0002]
• U.S. Pat. No. 5,424,833 discloses a measuring device of another type wherein the first and third gratings are replaced by an unique index grating used in transmission with a pitch twice as large as the pitch of the scale grating. Thus, the scale grating, which is longer than the index grating, has a pitch or spatial period smaller than that of this index grating. Further, all embodiments in this document are arranged so that the incident beam falling on the index grating has a main propagating direction comprised in a plane perpendicular to the moving direction of the scale grating and thus parallel to the lines of both gratings. In order to spatially separate the light source and the photodetector, this document proposes, in a first embodiment, to have said main propagating direction oblique relative to the direction perpendicular to the index grating in said perpendicular plane. In a second embodiment, the incident beam falls perpendicularly onto the index grating and a beam splitter is used which deflects the interference beam coming back normally from the index grating into a direction different from the light source. The first embodiment needs an extended space in a direction perpendicular to the moving direction (measurement direction) and to the direction perpendicular to the gratings. The second embodiment has the following drawbacks: it needs an extended space between the source and the index grating, it is less efficient, and it involves more parts.[0003]
• European application 0 603 905 discloses a measuring device wherein two gratings are formed on the scale, a first one for splitting the light beam coming from the source and a second one with a pitch twice smaller for interchanging the directions of the two used beams diffracted by the first grating. The mixing grating used in transmission is attached to the photodetector. This arrangement is not very efficient because its resolution is twice as small as the resolution of the device of U.S. Pat. No. 5,424,833 for gratings having pitches identical to those of the latter. Further, the scale is transparent and either its two main surfaces are arranged for diffracting and/or reflecting light beams, or an additional mirror is needed. The scale is thus relatively difficult to manufacture.[0004]
• An object of the invention is to provide an optical device for measuring relative movements which has great measuring accuracy while remaining of relatively simple construction.[0005]
• Another object of the invention is to provide such a measuring device the arrangement of whose various parts, in particular the scale or longer grating, can be made within relatively large manufacturing tolerances without adversely affecting the accuracy of measurements.[0006]
• Another object of the invention is to provide a measuring device of this type wherein the variation in wavelength of the source and of its angular spectrum have no influence on the accuracy of measurements.[0007]
• Another object of the invention is to provide a device of this type allowing a very flat arrangement which can easily be miniaturised.[0008]
• A particular object of the invention is to provide a device of this type at least partially integrated in a silicon or semiconductor substrate.[0009]
• The invention therefore concerns a device for measuring translation, rotation or velocity via light diffraction including a light source, at least one light detector, a first grating or first and fourth gratings of the same spatial period and located substantially in a same first plane, and a second grating or second and third gratings of the same spatial period and located substantially in a same second plane; the first and, where appropriate, fourth gratings being mobile along a given direction of displacement relative to the second and, where appropriate, third gratings, this device being arranged so that a first light beam generated by said source defines a beam incident upon said first grating where this incident beam is diffracted into at least a second beam and a third beam; so that these second and third beams then reach at least partially said second grating or, where appropriate, said second and third gratings respectively, where they are respectively diffracted into at least fourth and fifth beams whose propagating directions are interchanged respectively with the propagating directions of said second and third beams; so that these fourth and fifth beams then reach at least partially said first grating or, where appropriate, said fourth grating where they are respectively diffracted in a same output diffraction direction so that they interfere, said light detector being arranged to detect at least partially light resulting from said interference; the first, second and, where appropriate, third and/or fourth gratings being used in reflexion.[0010]
• The features of this measuring device allows an easy miniaturisation and its integration by microelectronic and microsystem technologies.[0011]
• According to a preferred embodiment, said first and, where appropriate, fourth gratings belong to a portion of the device which is mobile relative to said incident beam, said second and, where appropriate, third gratings being fixed relative to this incident beam.[0012]
• According to a particular embodiment, the first and, where appropriate, fourth gratings have a pitch or spatial period which is twice as large as that of the second and, where appropriate, third gratings, said second and third beams being diffracted respectively into the <<+1>> and <<−1>> orders, said fourth and fifth beams being diffracted respectively into the <<−1>> and <<+1>> orders and these fourth and fifth beams being respectively diffracted into the <<+1>> and <<−1>> orders in said same output diffraction direction by said first or, where appropriate, fourth grating.[0013]
• According to a preferred feature of the measuring device according to the invention, the light from said incident beam forming said second, third, fourth and fifth beams and finally detected by the detector reaches said first grating at an angle of incidence which is not zero in a plane perpendicular to lines forming the gratings, this angle of incidence being sufficient so that the light source providing said light and the detection region of the detector receiving said light are spatially separated from each other in projection in a plane perpendicular to said lines.[0014]
• According to a particular feature, said output diffraction direction defines an angle, in said plane perpendicular to lines forming the gratings, which has a value substantially equal to the angle of incidence of the incident beam multiplied by <<−1>> relatively to an axis perpendicular to said gratings, only light interfering along this output diffraction direction being used for measuring a displacement. Thus, the optical arrangement is fully symmetrical and so reciprocal.[0015]
• Other objects, particular features and advantages of the present invention will appear more clearly upon reading the following detailed description, made with reference to the annexed drawings, which are given by way of non-limiting example, in which:[0016]
• FIG. 1 shows schematically an optical device for measuring a relative displacement,[0017]
• FIGS. 2 and 3 show schematically a first embodiment of a measuring device, according to the invention;[0018]
• FIG. 4 shows schematically the spatial distribution of the light beams used for the displacement measurement of a second embodiment;[0019]
• FIGS. 5 and 6 show schematically a third embodiment of a measuring device, according to the invention;[0020]
• FIGS. 7, 8 and[0021]9 show schematically three other embodiments of the invention;
• FIG. 10 shows schematically and partially an embodiment allowing an absolute measurement of the relative position between a mobile scale and the fixed portion of the displacement measuring device;[0022]
• FIGS.[0023]11 to15 show schematically various alternatives for defining a reference position of the mobile scale of the displacement measuring device;
• FIGS. 16 and 17 show schematically two other embodiments of the invention;[0024]
• FIGS. 18 and 19 show schematically an embodiment allowing measurement of displacement along two orthogonal directions;[0025]
• FIG. 20 shows schematically another embodiment of the invention in which the beam emitted by the light source and the interfered beam propagate parallel to the measured displacement direction.[0026]
• FIG. 1 shows a translation measuring device including a[0027]light source2 which supplies a first beam FI, which reaches a firsttransparent structure4 on one surface of which is arranged afirst grating6 of period Λ. Beam FI is diffracted into the <<+1>> and <<−1>> orders and generates twobeams8 and10.Beams8 and10 reach respectively second andthird gratings12 and14 where they are reflected and diffracted respectively into the <<−1>> and <<+1>> orders.Beams16 and18 resulting from these two diffractions propagate symmetrically tobeams10 and8 and are joined together as they reach afourth grating20 where they are diffracted, respectively into the <<+1>> and <<−1>> orders, along a same first direction of diffraction offset angularly by angle α relative to an axis perpendicular to grating20, this angle α being identical in absolute value to angle of incidence α of beam FI incident uponfirst grating6.
• The two beams generated by the diffraction of[0028]beams16 and18 ingrating20, along the aforementioned first direction, interfere and together form a beam FR which again passes throughtransparent structure4 and is then directed towardslight detector22 arranged for measuring the variation in the luminous intensity of beam FR resulting from said interference. The first and fourth gratings are situated in a same first general plane and arranged on a same face oftransparent structure4. Likewise, second andthird gratings12 and14 are arranged in a same second general plane of the device.Grating14 is arranged at the surface of areflective support24 which is fixed relative tostructure4, while grating12 is arranged at a surface of a mobilereflective support26 moving along a direction X parallel to the aforementioned first and second general planes. In this embodiment,mobile portion28, formed ofsupport26 and grating12 remains in a fixed position along axis Z during measured displacements.
• The path travelled by[0029]beams8 and16, on the one hand, andbeams10 and18 on the other hand, are identical. Consequently, the phase shift between the twobeams16 and18 incident upon grating20 depends solely upon the displacement ofmobile portion28. Those skilled in the art know how to calculate the phase shift generated by a displacement along axis X of thismobile portion28 forbeam16 generated by the diffraction ofbeam8 ingrating12, this phase shift increasing proportionally with the displacement of movingportion28 and the luminous intensity of beam FR detected bydetector22 varying periodically. Measurement of this periodic variation in the luminous intensity of beam FR allows the displacement ofmobile portion28 to be determined with great accuracy.
• [0030]Gratings6 and20 have a spatial period Λ andgratings12 and14 have a period which is substantially two times smaller, i.e. substantially equal to Λ/2 and preferably equal to Λ/2. This ratio between the spatial periods of the different gratings allows two reciprocal optical paths to be obtained defining a symmetry relative to axis Z. Indeed, due to the particular arrangement of the aforementioned different spatial periods an incident beam FI at point A ofgrating6 generates twodiffracted beams8 and10 which are diffracted respectively at points B1 and B2 along two directions which are symmetrical to the directions ofbeams8 and10 relative to axis Z. Consequently,beams16 and18 meet at point C situated on grating20. There is thus perfect superposition of the two beams interfering along said first direction of diffraction.
• It will be noted however that the four gratings can be situated in different general planes if required as long as the relative displacements are effected in displacement planes parallel to these general planes. However, such an arrangement loses certain of the advantages of the device of FIG. 1, in particular its independence relative to the wavelength λ of beam FI and its angle of incidence α. This is why, although such a solution is not excluded, an arrangement in accordance with FIG. 1 is preferred. Those skilled in the art can demonstrate mathematically that the intensity of beam FR resulting from the interference is independent of angle α and the wavelength of beam FI when[0031]gratings6 and20 are situated in a first general plane andgratings12 and14 are situated in a second general plane of the device. This feature is particularly advantageous for light sources emitting with a certain divergence or numerical aperture in a spectral band of a certain width, i.e. non monochromatic.
• According to a particular feature of the present invention, beam FI incident upon[0032]first grating6 has an angle of incidence α which is not zero. Consequently, in the plane of FIG. 1 which is parallel to the direction of displacement ofmobile portion28 and perpendicular tolines30,31,32 and33 ofgratings6,20,12 and14, the point of incidence A on grating6 and the point of interference C on grating20 are separated spatially so thatsource2 anddetector22 are separated spatially in projection in this plane and can thus be arranged so as to be globally aligned along a direction parallel to direction of displacement X. This allows very flat measuring devices to be obtained given that the source and the detector can both be arranged in a plane parallel to the measured displacement direction.
• Another consequence of non-zero incidence angle a is to prevent the spurious z-dependent modulation signal due to self-mixing when the source is a semiconductor laser.[0033]
• The device according to FIG. 1 is favourable for measuring a relative displacement between two bodies situated in a same general plane.[0034]
• Given that only diffraction orders <<+1>> and <<−1>> of grating[0035]6 are useful, thisgrating6 is arranged so that the majority of the luminous intensity of beam FI is diffracted into these two diffraction orders to form respectively beams8 and10. In particular, the light emitted into diffraction order <<0>> is minimised. Likewise, in the event that the second diffraction order may intervene, grating6 is arranged so that the light diffracted into this second order is relatively weak.
• By way of example, for a wavelength Λ=0.67 μm and an angle of incidence α=10°,[0036]diffraction grating6 is formed indielectric layer36 of refractive index approximately n=2.2, in particular made of Ta₂O₅or TiO₂deposited by a technique known to those skilled in the art, onglass substrate4, the total thickness E₁of this layer being comprised between 0.4 and 0.5 μm. The depth P₁of the grooves situated betweenlines30 of grating6 is comprised between 0.30 and 0.35 μm. Transmission of approximately 80% of the total luminous energy of beam FI is thus obtained in diffractedbeams8 and10. Defining thegrating6 inlayer36 composed of a high index dielectric material is particularly advantageous since it allows a large diffraction efficiency of the <<+1>> and <<−1>> orders to be obtained with a shallower groove depth P₁than in a lower index layer, or than directly in thetransparent structure4.
• Those skilled in the art can also optimise the profile of the section of grating[0037]6 along the transverse plane of FIG. 1 to further increase this selective transmission of the luminous energy or define other grating profilers in layers of different transparent materials such as SiO₂or polymers or solgels. It will be noted that, given that the diffraction events at point C form a reciprocal situation with the diffractions at point A, a difference in the percentage transmitted into the <<+1>> and <<−1>> orders at point A is re-established during diffraction at point C at angle α so that the contributions ofbeams16 and18 along the direction of diffraction selected are identical, which leads to maximum contrast for the interference. It will also be noted that the diffraction efficiency in the aforementioned example is substantially independent of the polarisation of the incident light. The light diffracted into <<0>> order is practically zero. With a period Λ=1 μm, diffraction orders greater than 1 do not exist.
• Those skilled in the art will choose for[0038]reflection gratings12 and14 a corrugated metal surface. It is known that such metal gratings exhibit high diffraction efficiency forbeams8 and10 of TM polarization only. High diffraction efficiency for the TE polarization requires a large groove depth which is very difficult to obtain in practice when the period is of the order of the wavelength. Furthermore, it is practically very difficult to obtain such metal grating exhibiting comparable large diffraction efficiency for both TE and TM polarizations ofbeams8 and10 as is requested in case the light source is unpolarized. An object of the invention is to provide high diffraction efficiency for the TE polarization, and for both TE and TM polarizations, by using a grating structure comprizing aflat mirror substrate26 or24, adielectric layer38 and40, the grating12 or14 being realized in thedielectric layer38 or40. Such structure associates the diffraction of grating12 or14 with the reflection of thereflective substrate26 or24 in order to give rise to constructive interference effects in the direction ofbeam16 or18.
• In a particular example,[0039]gratings12 and14 are both formed of a dielectric layer respectively38,40 also having a refractive index n=2.2. With a total thickness E₂=0.34 μm and a depth P₂=0.18 μm for the grooves situated betweenlines32 and33, the luminous intensity diffracted into the <<−1>> order for grating12 and the <<+1>> order for grating14 is approximately 50%, the remainder being essentially diffracted into the <<0>> order. Given thatbeam8 is diffracted to the right of the direction perpendicular to grating6, the light diffracted into the <<0>> order by grating12 does not disturb the measurement in any way since it is not received bydetector22. Likewise, the light diffracted at B2 into the <<0>> order reaches grating20 at a distance from point C comparable to the distance separating point C from point A. It is thus easy to arrangedetector22 so that the light diffracted at point B2 into the <<0>> order is not detected. This fact favours in particular a ratio between wavelength λ and period Λ generating propagation ofbeams8 and10 to the right and left of the direction perpendicular to grating6 respectively.
• The arrangement of[0040]gratings12 and14 described in the example hereinbefore is provided for a situation in which the light received is not polarised. However, if the light is TE polarised (electric field vector parallel to the grating lines), thickness E₂ofgratings12 and14 is approximately 0.1 μm, while the depth P₂is situated at around 0.08 μm and can even be equal to thickness E₂. Substrates24 and26 are made for example of aluminium or coated with an aluminium film or another suitable metal. Under these conditions, approximately 80% of the luminous intensity ofbeams8 and10 is diffracted respectively inbeams16 and18. For a TM polarisation (electric field vector perpendicular to the grating lines), one can omit the dielectric layer and the aluminium substrate is micro-machined with a groove depth of approximately 0.12 μm. In a variant, substrate of any type is micro-machined, then coated with a metal film. Thus, the luminous intensity diffracted inbeams16 and18 is approximately 70%. Again, the profiles ofgratings12 and14 in the plane of FIG. 1 can be optimised by those skilled in the art so as to increase the transmission of luminous energy in the respective useful directions, in proportions substantially equal but not necessarily equal at points B1 and B2. Other layer materials like other oxides, fluorides, polymers, solgels can be chosen and deposited or coated by different techniques like vacuum deposition, spinning, dipping, in which the grating can be achieve by dry or wet etching, lift-off, photo inscription or moulding techniques.
• [0041]Dielectric layer42 of grating20 has a thickness E₁and a groove depth P₁substantially identical to those of grating6 so as to assure reciprocity of the diffraction event at C relative to the diffractive event at A. The diffraction efficiencies at C correspond to those given hereinbefore for the diffractions occurring at A.
• Finally, in a variant,[0042]transparent structure4 is in two portions which are mobile in relation to each other and carry respectively the first andfourth gratings6 and20, while the second andthird gratings12 and14 are both attached to one of these two portions.
• FIGS. 2 and 3 show a first embodiment of the invention. Beam FI generated by a source which is not shown passes through[0043]transparent structure44 and reaches grating46, at an angle of incidence α, where it is diffracted into the <<+1>> and <<−1>> orders to formbeams8 and10, as in the first embodiment. However, this second embodiment differs from the first in thatbeam8 is diffracted to the left of the direction perpendicular to grating46. By way of example, the light wavelength λ=0.67 μm, angle of incidence α=20° and period Λ=2 μm.
• [0044]Beams8 and10 reach grating48 arranged at the surface ofreflective substrate50.Beams8 and10 are respectively diffracted by grating48 into diffraction orders <<−1>> and <<+1>> to form respectively beams16 and18 which are joined as they reach again grating46 where they are diffracted along a same diffraction direction, at an angle α relative to the direction perpendicular to grating46. Beam FR resulting from this interference again passes throughtransparent structure4 prior to being detected at least partially by a detector which is not shown.
• It will be noted that[0045]substrate50 is here stationary relative to the source and the detector, whilestructure44 is mobile along direction X. The luminous intensity of beam FR varies periodically as a function of the displacement ofstructure44 relative tosubstrate50. This detected luminous intensity and the periodic variation therein allows the relative displacement betweenstructure44 andsubstrate50 to be accurately determined.
• In order to optimise the transmission of the luminous energy of beam FI in diffracted[0046]beams8 and10 and also in order to optimise the transmission of the luminous energy of thesebeams16 and18 in beam FR, for α,λ and Λ given hereinbefore, grating46 is formed of adielectric layer52 of refractive index n=2.2 approximately and having a thickness E₁comprised between 0.35 and 0.40 μm with a groove depth P₁equal to approximately 0.24 μm. It will be noted that this grating structure and these values are given by way of non-limiting example and have been determined for atransparent structure44 with an index of approximately n=1.5. Under these conditions, approximately 60% of the luminous energy of beam FI is transmitted in diffractedbeams8 and10 in substantially equal parts, independently of the polarisation of the light. The luminous intensity transmitted into the <<0>> order is low. It is approximately zero for TE polarisation while it reaches approximately 5% for TM polarisation.
• In the event that the light is not polarised,[0047]second grating48 is formed by adielectric layer54 of refractive index n=2.2 having a total thickness E₂comprised between 0.25 and 0.30 μm with a groove depth P₂=0.22 μm. As in FIG. 1, a high efficiency grating comprising adielectric layer54 and areflective substrate50 is provided, the grating48 being made in said dielectric layer. Approximately 55% of the luminous intensity ofbeams8 and10 is diffracted respectively inbeams16 and18. Preferably, the refractive index of the dielectric layers mentioned is greater than 1.8. For the sole TE polarised light, the luminous intensity diffracted into the useful orders at grating48 can be increased to approximately 70% with a thickness E₂slightly greater than 0.30 μm. Under these conditions, it is possible to obtain 70% of the energy transmitted inbeams16 and18 while the luminous energy diffracted into the <<0>> order is very low; which is not the case for TE polarisation when thickness E₂is less than 0.30 μm.
• The numerical example given here thus allows the luminous energy transmitted into diffraction order <<0>> in grating[0048]46 to be reduced to the maximum and also, although to a lesser extent, in grating48. Then, the light transmitted into the second diffraction order is relatively small. Consequently, the only significant interference is that generated by the diffraction ofbeams16 and18 in grating46 respectively into the <<+1>> and <<−1>> orders, at angle of diffraction α. This favourable situation results essentially from the fact that the transmission ofbeams16 and18 into the <<0>> order of diffraction and the orders greater than the first order of diffraction at point C is relatively low, or even zero. Thus, a detector situated in proximity to point C essentially receives beam FR as a light signal varying alternately as a function of the displacement ofsubstrate44. The other contributions received by this detector generate a substantially constant signal independent of the relative displacement betweensubstrate50 andstructure44.
• In the example given here, the light is essentially transmitted in the useful orders and the low intensity of the light transmitted into the <<0>> order of diffraction at points A and B[0049]1 allow any light generating a constant signal to be reduced to the maximum for the luminous intensity received by the detector. It will also be noted that given that the diffraction at point C into the <<0>> order is relatively low, any interference with a diffraction into the second order can generate only a small luminous variation and thus a minor disturbance for the measurement signal propagating at angle α and formed by beam FR. In the examples given hereinbefore, most of the luminous intensity ofbeams16 and18 is diffracted respectively into the <<+1>> and <<−1>> orders, the amplitudes of the diffracted beams into other orders being small or zero. It is to be noted that no particular care must be taken of the luminous intensity in the zero and second orders when the light source is a broadband source like a Light Emitting Diode (LED) since their contribution in the detected signal only amounts to a DC component because of the short coherence length of a LED.
• In order to be able to determine the direction of relative displacement between[0050]structure44 andsubstrate50, grating48 has been divided into two regions R1 and R2 along the direction perpendicular to direction of displacement X (FIG. 3). In region R2, grating48 is also divided into two distinct regions R3 and R4. In region R3, lines58 of grating48 are in phase over the two regions R1 and R2. However, in region R4, lines58 have a discontinuity given that the part of these lines situated in region R2 is offset by Λ/8 relative to the part of these lines situated in region R1.Grating48 is arranged relative to the light source so thatbeam8 reaches grating48 in region R3 whilebeam10 reaches in region R4. In these conditions those skilled in the art can calculate that the offset of Λ/8 provided in region R4 finally generates a phase shift of Π/4 betweenbeams16 and18 incident upon grating46 at point C. Consequently, the luminous intensity resulting from the interference originating from region R1 has a phase shift of Π/2 relative to the interference originating from region R2. By separately detecting the contributions from regions R1 and R2, the detector receives two alternating luminous intensity signals phase shifted by rV2 in relation to each other. In a variant, it is possible to provide three gratings in parallel with an offset of Λ/6 to give three luminous intensity signals phase shifted by 120°. Ifbeams8 and10 are not spatially separated when they reach grating48, region R2 does not have to be separated into regions R3 and R4. Region R2 as a whole is offset by Λ/16 with respect to region R1 in order to provide an optical intensity phaseshift of Π/2, or by Λ/12 for a phaseshift of 120°.Grating48 can also be devided into four regions similar to R1 and R2 with three regions having respectively offsets of Λ/16, Λ/18, 3Λ/16 relative to the last one in order to obtain the full set of four quadrature optical power signals.
• Thus, on the basis of these two, or three or four separately detected signals, the electronic system of the measuring device can determine the direction of relative displacement between[0051]structure44 andsubstrate50 and interpolate finely within the electric period Λ/4 of the luminous intensity resulting from said interference to further increase the accuracy of the measurement. It will be noted that, in the case of the device of FIG. 1, this electric period is Λ/2.
• It will be noted that a variation in the spacing between this[0052]structure44 andsubstrate50, i.e. a variation in thedistance separating gratings46 and48 has no influence on the measurement of the displacement along axis X, the two optical paths between points A and C remaining identical and the phase shift between the two contributions forming beam FR and originating respectively frombeams16 and18 remaining dependent solely on the relative displacement along axis X.
• Finally, it will be noted that the phase shift for a given displacement is twice as large in this second embodiment than in the first embodiment of FIG. 1.[0053]
• FIG. 4 shows schematically a second embodiment in which[0054]transparent structure44 is stationary relative tosource2 anddetector22,reflective substrate50 being mobile.Gratings46 and48 are the same as those described with reference to FIG. 2. FIG. 4 is given to allow the light useful for the displacement measurement provided bysource2 to be visualised. Thissource2 generates a beam FI which has a divergence or numerical aperture and which reaches grating46 at an angle of incidence varying continuously within a range of given values. It will be noted that this range of values can include the value α=0, i.e. an incidence perpendicular to grating46. This beam FI generatesbeams8,10,16,18 and FR as described hereinbefore. The numerical aperture of beam FI generates a divergence of these diffraction beams.
• Since[0055]detector22 is arranged relative tosource2 so that their projections in a plane perpendicular to the lines ofgratings46 and48 are not superposed, although they are globally aligned along a substantially parallel direction to the direction of displacement, only the light which is comprised in a partial beam FI* and illuminates region RA of grating46 (comprised between the two arrows in the drawing) forms the partial beam useful for the displacement measurement. According to the invention, the totality of light FI* incident upon region RA has an angle of incidence which is not zero, but sufficiently large for the light finally incident upondetection element80 to be spatially separated from the light forming beam FI*, in projection in a plane perpendicular to the lines ofgratings46 and48 corresponding to the plane of the drawing of FIG. 4. Whendetection element80 is situated in direct proximity to region RC wherepartial beams16* and18* arrive which generate partial beam FI* detected bydetector22, this condition corresponds to a spatial separation of regions RA and RC of grating46. Beam FI* which is useful for the displacement measurement thus generatespartial beams8* and10*, which reach grating48 respectively in regions RB1 and RB2. From there they are diffracted to formpartial beams16* and18* and are joined in region RC of grating46 where they are diffracted along a same direction to form partial beam FR* of beam FR.
• In conclusion, whatever the divergence or numerical aperture of beam FI, only partial beam FI* contributes to the displacement measurement and only regions R[0056]1, FB1, RB2 and RC define the active regions ofgratings46 and48 in which the optimising conditions for maximum diffraction efficiency and maximum contrast of the detected interference signal must be fulfilled. It will also be noted that the light forming beam FI* can have a wide spectrum.
• Hereinafter, the numerical references already described will not be described again in detail, since they were only given as an example. It is indeed an object of the invention that the gratings can be manufactured with large tolerances without affecting the measurement accuracy.[0057]
• With reference to FIGS. 5 and 6 a third embodiment of the invention will be described hereinafter, wherein an angular displacement of a[0058]wheel60 is measured, said wheel having at its periphery a grating62 formed oflines64 parallel to the axis of rotation ofwheel60.Grating62 defines a scale of period Λ. Facing grating62 there is provided a measuringhead66 formed of atransparent structure68 having on its face opposite grating62 adiffraction grating70 having a period Λ/2. The ratio of the period of grating70 to the period of grating62 is substantially ½when the angle between the normals to grating62 at points A et C is close to zero. This ratio is smaller than ½when the radius ofwheel60 is small and when the spacing between gratings is large. On the other face ofstructure68 are arranged alight source72 and adetector74. Beam FI generated bysource72 passes throughstructure68 and reaches grating62 where it is diffracted in reflection essentially into the two orders of diffraction <<+1>> and <<−1>>. Beam FR, resulting from the interference ofbeams16 and18 diffracted in reflection at angle α at point C, again passes throughstructure68 prior to being detected bydetector74.Grating70 is formed in areflective substrate76 deposited at the surface oftransparent structure68.
• An incremental angle of rotation of[0059]wheel60 corresponds to period Λ of grating62. Thus, for every displacement of grating62 relative to measuringhead66 there is a corresponding angle at centre ofwheel60. Consequently, the processing of the alternating luminous signal detected bydetector74 allows an angle of rotation ofwheel60 to be accurately determined.
• As in the second embodiment, the direction of rotation of[0060]wheel60 can be detected. In order to do this, grating70 shown in plane in FIG. 6 has two regions R1 and R2 in which thelines78 of grating70 are offset by Λ/16. This offset provided at points B1 and B2 finally generates an optical intensity phase shift of Π/2 in beam FR between the two contributions originating from regions R1 and R2.
• FIG. 7 shows a fourth miniaturised embodiment which is partially integrated in a[0061]semiconductor substrate82. Thissubstrate82 has anaperture84 wherein is arranged a collimation ball for the light emitted byelectroluminescent diode88 arranged at or close to the surface ofball86.Diode88 is arranged so that the central axis of beamFI leaving ball86 has an angle of incidence which is not zero when reaching grating90 of period Λ. On the face ofsubstrate82 situated facing grating90 arranged onreflective substrate112 there is provided a reflection grating92 of period Λ/2. This grating92 can be either micro-machined directly insubstrate82, in particular in silicon, or be obtained by deposition of one or more layers by deposition techniques known to those skilled in the art. In particular, it is possible to deposit a metal layer followed by a dielectric layer. The lines of grating92 can be obtained either by micro-machining the dielectric layer or by a two phase deposition, the deposition effected in the second phase forming the lines of grating92. The resulting beam FR originating from diffraction ofbeams16 and18 in grating90 is finally detected bydetector98 integrated insubstrate82. Such detectors are known to those skilled in the art, as is the electronic circuit used for processing the light signals received by saiddetector98.
• It will be noted that the light detector can be formed by a unit which is materially distinct from[0062]substrate82, in particular by a detection unit preceded by a focusing element. In such case, this detection assembly is arranged either in another aperture, or in a recess provided on the face of thisstructure82 situated opposite grating90.
• FIG. 8 shows a fifth miniaturised and partially integrated embodiment.[0063]Semiconductor substrate82 comprisingintegrated detector98 has arecess100 in which is arranged the source formed ofelectroluminescent diode88 andtransparent ball86. The bottom ofrecess100 is closed by atransparent layer102, in particular made of SiO₂or Si₃N₄, arranged on one face ofsubstrate82 on the side ofdetector98. At the surface of thislayer102 is arranged a dielectric layer defining grating104 of period Λ. Facing grating104 is arranged reflection grating106 of period Λ/2 at the surface of a mobilereflective scale108.
• FIG. 9 shows a sixth entirely integrated embodiment. The displacement measuring head is formed by[0064]semiconductor substrate82 in which are integrateddetector98 andlight source110. Preferably,source110 is directly integrated insubstrate82. In a variant,source110 can be manufactured separately and arranged at the surface ofsubstrate82 or in a recess provided for the source. Althoughsource110 emits with a large numerical aperture in several directions, only a portion of the beam generated defines beam FI diffracted bygratings90 and92 is finally detected byintegrated detector98. The optical paths of the two end beams FIA and FIB of partial beam FI have been shown so as to visualise the spatial distribution of the different diffracted beams useful for the relative displacement measurement betweensubstrates82 and112. The two end rays of each beam are referenced respectively by the letters <<A>> and <<B>> after the previously used numerical reference. This sixth embodiment allows an ultimate miniaturisation of the measuring device according to the invention and the integration thereof in mechanical and micromechanical devices.
• FIG. 10 shows schematically a seventh embodiment of the invention which differs from the sixth in that, in place of a[0065]single grating90, threegratings90A,90B and90C are provided, arranged next to each other and having respectively three different, although relatively close, spatial periods Λ1, Λ2 and Λ3.Grating92 is also replaced by three gratings (not shown) situated facing the threegratings90A,90B and90C, and each having a spatial period which is two times smaller than the spatial period of the grating which it faces. For each of the pairs of gratings, the application of the optical principle disclosed in the present invention is identical. By selecting appropriate values for Λ1, Λ2 and Λ3, the light intensities, received by a detector having three distinct detection zones for the three pairs of gratings, define a signal corresponding to a single relative position betweensubstrate82 andsubstrate112. Such a device thus enables the absolute position of the mobile portion to be defined relative to the fixed portion of the device. This constitutes an application of the Vernier principle. The device can contain N paths of different periods to assure univocal coding of each measured relative position betweensubstrates82 and112.
• FIGS.[0066]11 to14 show schematically four alternative embodiments of the mobile portion relative to the light source and the detector each able to be arranged in any of the embodiments described hereinbefore to define at least one reference position between the fixed portion and the mobile portion of the displacement measuring device.
• According to the variant of FIG. 11, in addition to base grating[0067]116 of constant period Λ or Λ/2, there is provided beside this latter another grating118 of variable spatial period and decreasing to substantially an identical period to that of grating116, able to perform identically to grating116 on a certain number of lines, to increase again. The reference position REF is defined by the symmetrical axis of grating118. The variant of FIG. 12 differs from that of FIG. 11 in that a grating120 is provided beside grating116 whose period varies by increasing or decreasing passing from a value higher than the value of the period of grating116 to a lower value than the latter. Reference position REF corresponds to the middle position of the place of coincidence between the periods ofgratings116 and120 able to extend over a certain number of lines.
• When the light beam sweeps grating[0068]118 of FIG. 11 or grating120 of FIG. 12, an interference signal is generated on passing across the reference region allowing the displacement measuring detector or another detector to determine reference position REF. This originates from the fact that grating118 or120 has only in the reference region a period having aratio 1/2 or 2/1 with the grating situated opposite on the fixed portion of the displacement measuring device. In other words, there is coding of an absolute or reference position by mutual spatial coherence of the two gratings.
• FIG. 13 shows another variant wherein there is provided beside grating[0069]116 a grating122 of decreasing then increasing variable period passing from a period higher than that of grating116 to a lower period. Grating122 has symmetry relative to reference position REF situated between two interference signals occurring at two reference positions REF1 and REF2 where the period is identical to that of grating116. Grating122 thus allows two reference positions REF1 and REF2 to be determined, which allows the detected signal processing means to define with great accuracy the central reference position REF.
• In FIGS.[0070]11 to13 it will be noted that in the event that grating116 has a period Λ/2, the mutual coherence at the reference location must be verified at least partially for the diffraction events at the diffraction points or regions ofincident beams8 and10. Consequently, the variant of FIG. 13 can only define one reference position with a spacing between these two points or regions substantially equal to the distance between REF1 and REF2.
• FIG. 14 shows another alternative embodiment wherein the mobile portion relative to the light source includes grating[0071]126 of period Λ/2. Asecond grating128 is provided beside grating126, these twogratings126 and128 being arranged facing the grating of constant period Λ. Grating128 is formed oflines130 defining a period Λ/2 with two discontinuities defining an phase shift or offsetting of one portion of grating128 relative to thecorresponding lines132 of grating126. Grating128 thus has a first offset of Λ/4 increasing a space between twolines130 to 3Λ/4. At a certain distance from this offset a second offset of Λ/4 is provided decreasing from period Λ/2, generating a space Λ/4 between twoother lines130.
• FIG. 15 shows the variation in the luminous intensity detected by a detector as a function of the displacement of grating[0072]128 when the light beam passes through the region including the two offsets of opposite directions described hereinbefore. First, the component AC of the intensity I of beam FR defined hereinbefore decreases given that one increasing portion of this beam includes an interference product having a phase difference of Π. When more than half of the first offset of grating128 has been passed through or the second phase jump is reached, the component AC of intensity I again increases to the maximum before again decreasing and then increasing towards the initial mean value.Graph134 of FIG. 15 thus defines three reference points F1, F2 and F3 allowing three reference positions to be defined or, using a processing unit, central reference position F2 to be accurately defined. It will be noted here that it is possible in another variant to provide a single phase jump of Λ/4 thus generating a single minimum in the AC component of intensity I.
• FIG. 16 shows another embodiment of the invention which is particularly advantageous and able to be miniaturised. The device includes on the one hand a[0073]substrate82 on one face of which is arranged a light source, in particular an electroluminescent diode or a light source integrated in a semiconductor region ofsubstrate82 and known to those skilled in the art. As in the embodiment of FIG. 9, thissource110 can be a porous silicon light emitting zone, an electroluminescent polymeric emitter in a recess zone, or a LED chip bonded ontosubstrate82. This embodiment differs essentially from the sixth embodiment in that a partial beam propagating to the right oflight source110 and another partial beam propagating to the left of said source are used for the displacement measurement. Thus, to the left and right ofsource110 are provided twogratings92 and92′ of period Λ/2. On either side of these two reflection gratings are arranged twolight detectors98 and98′ integrated in regions ofsemiconductor substrate82. The optical paths of the beams diffracted to the left and right ofsource110 and the two partial beams used for the displacement measurement are substantially symmetrical. Facing the face ofsubstrate82 havinggratings92 and92′ is arranged a grating90 of period Λ on areflective substrate112.
• In order to determine the direction of displacement of grating[0074]90 and to interpolate in a period of the detected luminous intensity signal, a variant provides an offset of Λ(m/4+{fraction (1/16)}) betweengratings92 and92′ where m is an integer number. Consequently, the alternating signal detected bydetector98 is phase shifted by Π/2 relative to the alternating signal detected bydetector98′. However, in order to be free of any dilatation problem, it is preferable to provide two additional gratings phase shifted or offset by Λ/16 on each side ofsource110. The mention of possible expansion leads us to mention here an application of the device according to the invention to temperature measurements by expansion of the substrate formed of materials determined for such application. This is important in rotating or translating mechanical systems where the temperature of the moving parts has to be monitored as a criterion for the system's safety or lifetime.
• FIG. 17 shows another particularly advantageous embodiment which differs to that described in FIG. 16 in that an[0075]opening100 is provided in thesilicon substrate82 in which acollimation ball86 is arranged and adiode88 arranged at the surface or at a distance of saidball86 so that the direction defined by the centre ofdiode88 and the centre ofball86 is substantially perpendicular to adiffraction grating140 arranged so as to close opening100 on the side of the surface ofsubstrate82 havingdiffraction gratings92 and92′. The light supplied bydiode88 is collimated byball86 so that most of the light reaches grating140 with a substantially perpendicular direction. Grating140 has a spatial period and a profile determined so that most of the luminous intensity incident upon grating140 is diffracted substantially in equal parts into the <<+1>> and <<−1>> diffraction orders. The angle of diffraction in the air with respect to the direction perpendicular to grating140 is for example comprised between 20° and 50°. Thus, most of the luminous intensity provided bydiode88 is transmitted in useful beams FI and FI′. Grating140 can be formed in a SiO₂or Si₃N₄layer or in a multi-layered structure including in particular a superficial dielectric layer of index n greater than 2.0.Gratings92 and92′ are formed at the surface ofsubstrate82 by deposition of ametal layer142 followed by deposition of adielectric layer144, for example SiO₂or Si₃N₄. Alternatively, the grating can be first etched intosubstrate82 followed by metal deposition.
• In a variant, it is possible to provide a polarisation element between[0076]ball86 andgrating140. In another variant, it is possible to provide a transparent layer formed insubstrate82 and defining the bottom ofrecess100. On this transparent layer is deposited a dielectric layer in which are formed grating140 andgratings92,92′. It will be noted that any light source may be provided in this embodiment, fixed tosubstrate82 or at a distance from the latter. Preferably, the incident light over grating140 is substantially collimated. However, even for a diverging source, grating140 allows transmission into the <<0>> diffraction order to be limited and thus the luminous intensity to be concentrated along directions defining a non zero angle of incidence on grating90.
• FIGS. 18 and 19 show another embodiment of the invention allowing a displacement along two orthogonal axes of displacement X and Y to be measured. The arrangement along axis X, Y respectively corresponds to the embodiment described hereinbefore in FIG. 17. A[0077]bi-directional grating150 diffracting along directions X and Y is arranged onreflective substrate112. Thisbi-directional grating150 is formed of a set ofstuds152 defining grating lines along axes X and Y respectively. It may also be formed by a set of recesses or square hollows, regularly distributed along axes X and Y.Bi-directional grating150 shown in FIG. 18 is mobile relative to the portion forming the measuring head shown in FIG. 19 and corresponding to the portion associated with the source. The measuring head includes on one of its faces arranged facing grating150, abi-directional grating140A having the same function as grating140 along the two directions X and Y.Grating140A diffracts a light of normal incidence essentially into the first diffraction order in directions X and Y.Dotted line154 represents an opening in the measuring head while the light source supplying a substantially collimated beam is represented bydotted line156. Grating140A is formed of studs orsquare hollows158 aligned along the two directions X and Y. The measuring head further includes fourgratings92,92′,92A and92A′ of period Λ/2 and at least fourdetectors98,98′,98A and98A′ arranged so as to allow optical paths along the two directions X and Y as shown in the embodiment of FIG. 17 for a unidirectional displacement along axis X.
• It will be noted that, in a less perfected variant, it is possible to use a diverging source, in particular the[0078]source110 shown in FIG. 16, and to omitdiffraction grating140A. It will also be noted that the embodiments shown in FIGS.1 to8 can each also be arranged in a bi-directional displacement device. In order to do this, the light source in particular is arranged so as to emit light along the two directions X and Y in a direction of propagation which is not perpendicular to the diffraction grating of period A similar tobi-directional grating150 shown in FIG. 18. In the case of a collimated beam, in particular a laser beam, this beam will be oriented in a non-perpendicular way with respect to the measuring device grating and will have a direction, in projection in the plane X-Y, median to axes X and Y.
• Another use according to the invention of the devices corresponding to FIGS. 2, 4,[0079]5,7,8,9,16,17,18 or19 is the measurement of the relative velocity V along direction X between two gratings, by measuring the instantaneous frequency f of the modulated signal detected in the direction of beam FR by at least one detector. The relationship between f and V is given by V=Λf/4. It allows a direct measurement of the velocity without resorting to phase measurement and period counting.
• A further embodiment of the invention for velocity measurement corresponds to FIGS. 2, 4 or[0080]8 whereby grating48 or106 is the rough surface of the movingsubstrate50 or108 whose Fourier component along coordinate X corresponding to the spatial frequency of period Λ/2 has non-zero amplitude.Substrate50 or108 can be a moving band or wire. Among all the beams scattered in all directions at points B1 and B2 illuminated bybeams8 and10, only those diffracted indirections16 and18 by the spatial frequency corresponding to the spatial period Λ/2 will interfere after recombination along beam FR by grating46 or104. Two conditions may preferably be fulfilled for a high constructive interference to take place along the beam FR. The first condition is that the rough surface ofsubstrate50 or108 is placed at a distance from grating46 wherebeams8 and10 have a non-zero spatial overlap on said surface. The second condition is that the length difference AB2−AB1 (FIGS. 1 and 2) betweenbeams10 and8 is smaller than the coherence length ofsource2. This interference appears as a peak of frequency f in the temporal frequency spectrum of the optical power detected by at least one detector, f being related to the instantaneous velocity V ofsubstrate50 or108 by V=Λf/4. Those familiar with the art will easily locate f in the frequency spectrum by resorting to spectral analysis instruments dedicated to Doppler velocimetry. The advantages of the device according to the invention for velocity measurement are the miniaturization, the possible small spacing between the readout head, comprising the light source, the detector and the grating of period Λ, and the moving substrate. Another advantage is the possibility of using a Light Emitting Diode.
• A further embodiment of the invention for velocity measurement relates to the previous embodiment where grating[0081]48 is the surface, exhibiting a non-zero spatial component at period Λ/2, of asubstrate50 moving at velocity V. The distinct characteristics with respect to the previous embodiment is that thetransparent grating46 of period Λ no longer has a fixed position relative to the source and to the detector, but translates at a constant and known velocity v_ralong X, v_rbeing larger than the maximum which V can have. In one variant, grating46 is a radial grating made at the surface of a large radius disk rotating in a plane parallel to the displacement direction X and normal to the plane of incidence of beam F1. In a second variant, grating46 is a closed grating band rotating on two drums having their rotation axis normal to the incidence plane, the movement of grating46 between the source/detector assembly and thesubstrate50 being rectilinear and in the X direction.Grating46 is for instance made by embossing in a polymeric foil. The frequency f of the modulated optical power signal measured by the detector is related to the velocities V and v_rthrough f=4/Λ(V+v_r). This embodiments allows the accurate and fast measurement of the velocity V even in case V is close to zero. As a consequence, this embodiment allows an accurate determination of the length of a finite displacement L inclusive of its slow beginning and of its slow end by integrating the velocity V over time t.$L={\int }_{t_0}^{\mathrm{t1}}Vt=\frac{\mathrm{<figure-callout\; label="\Lambda "\; filenames="US20010006421A1-20010705-D00001.png"\; state="\{\{state\}\}">\Lambda </figure-callout>}}{4}{\int }_{t_0}^{{\mathrm{<figure-callout\; label="t"\; filenames="US20010006421A1-20010705-D00001.png,US20010006421A1-20010705-D00002.png"\; state="\{\{state\}\}">t</figure-callout>}}_1}ft-v_r\left(t_1-t_0\right)$

  
• where t[0082]₀and t₁are the starting and stop times of the displacement. The device according to the invention can therefore be advantageously used to measure the length of long strands of wire, bands, ribbons or sheets of different materials.
• FIG. 20 shows an embodiment of a measuring device with a[0083]mobile scale160 allowing a maximum measurement range for a given grating length and having in addition the advantage that the whole set of the gratings, source, detector(s) and optical paths used for the measurement is entirely contained in a closed case (a tube for example), without the mobile grating associated withscale160 exiting the case, while the displacement range of this scale (a metal rod for example) can reach a value only slightly smaller than the length of theinner cavity164 ofcase162, and without thescale160 supporting the source and the detector. In order to do this, alight source166 emits a beam FI along a direction essentially parallel to direction of displacement X.Rod160 has in its upper portion aplane168 inclined at an angle greater than 45° relative to axis X. Thisinclined plane168 defines a mirror for beam FI, which is reflected in the direction of afixed grating170 of period Λ arranged on a wall ofcavity164. Beam FI thus reaches grating170 at an angle of incidence which is not zero according to the invention. Scale orrod160 also includes areflective surface162 defining a grating174 of period Λ/2. Following grating174 is arranged aninclined plane176 defining a mirror. Thisinclined plane176 defines an angle, relative to a direction perpendicular togratings170 and174, equal to the angle defined betweeninclined plane168 and direction X. Thus, the resulting beam FR is reflected along a direction parallel to axis X and is directed towardsdetector178.
• Those skilled in the art will understand that it is possible to invert the arrangement of[0084]source166 anddetector178, the optical paths remaining the same and the light propagating in a reverse direction to that shown in FIG. 20. In order to assure a stable displacement along axis X, twobearings180 and182 are provided at the opposite end to that where the source and the detector are arranged. It will be noted that any other guide means, in particular a slide can be provided as an alternative arrangement.
• Other variants using mirrors to deviate and orient incident beam FI and resulting beam FR can be designed by those skilled in the art while remaining within the scope of the present invention and, in particular, of the embodiment described with reference to FIG. 20.[0085]
• Finally, it will be noted that the gratings can be formed in various ways by various methods known to those skilled in the art, in particular by a periodic variation in the refractive index at the surface of a plane dielectric layer. Moulding and embossing techniques may also be envisaged. The profiles of the transverse sections of the diffraction gratings can be optimised for each particular device in order to increase the efficiency of the displacement measurement according to the principle of the invention.[0086]

Claims (30)

What is claimed is:

1. A device for measuring translation, rotation or velocity via light diffraction including a light source, at least one light detector, a first grating or a first grating and a fourth grating of the same spatial period and located substantially in a same first plane, and a second grating or a second grating and a third grating of the same spatial period and located substantially in a same second plane; the second and, where appropriate, third gratings being mobile along a given direction of displacement relative to the first and, where appropriate, fourth gratings, this device being arranged so that a first light beam generated by said source defined a beam incident upon said first grating where this incident beam is diffracted into at least a second beam and a third beam; so that these second and third beams then reach at least partially said second grating or said second and third gratings respectively, where they are respectively diffracted into at least fourth and fifth beams whose propagating directions are interchanged respectively with the propagating directions of said second and third beams; so that these fourth and fifth beams then reach at least partially said first grating or, when appropriate, said fourth grating where they are respectively diffracted in a same output diffraction direction so that they interfere at least partially, said light detector being arranged to detect at least partially light resulting from said interference; wherein said first and second gratings and, where appropriate, said third and/or fourth gratings are used in reflexion.

2. The device of

claim 1

, wherein said first and, where appropriate, fourth gratings belong to a portion of the device which is mobile relative to said incident beam, said second and, where appropriate, third gratings being fixed relative to this incident beam.

3. The device of

claim 2

, wherein said second grating and, where appropriate, said third grating are arranged between said source and said detector.

4. The device of

claim 3

, wherein said second and, where appropriate, third gratings form together with said source and said detector a measuring head of this device, said first grating defining a scale of said device.

5. The device of

claim 4

, wherein said detector is integrated in a region of a semiconductor substrate bearing said second grating and, where appropriate, said third grating.

6. The device of

claim 4

or

5

, wherein said light source is integrated or arranged in a region of a semiconductor substrate bearing said second and, where appropriate, said third grating.

7. The device of

claim 1

, wherein the second and, where appropriate, third gratings have a spatial period which is twice as small as that of the first and, where appropriate, fourth gratings, said second and third beams being diffracted respectively into the <<+1>> and <<−1>> orders, said fourth and fifth beams being diffracted respectively into the <<−1>> and <<+1>> orders, and these fourth and fifth beams being respectively diffracted into the <<+1>> and <<1>> orders in said same output diffraction direction by said first or, where appropriate, fourth grating.

8. The device of

claim 2

, wherein the second and, where appropriate, third gratings have a spatial period which is twice as small as that of the first and, where appropriate, fourth gratings, said second and third beams being diffracted respectively into the <<+1>> and <<−1>> orders, said fourth and fifth beams being diffracted respectively into the <<−1>> and <<+1>> orders, and these fourth and fifth beams being respectively diffracted into the <<+1>> and <<−1>> orders in said same output diffraction direction by said first or, where appropriate, fourth grating.

9. The device of

claim 7

, wherein said output diffraction direction defines an angle, in a plane perpendicular to lines forming the gratings, which has a value substantially equal to the angle of incidence of the incident beam multiplied by <<−1>> relatively to an axis perpendicular to said gratings, only light interfering along this output diffraction direction being used for measuring a relative displacement.

10. The device of

claim 8

, wherein said output diffraction direction defines an angle, in a plane perpendicular to lines forming the gratings, which has a value substantially equal to the angle of incidence of the incident beam multiplied by <<−1>> relatively to an axis perpendicular to said gratings, only light interfering along this output diffraction direction being used for measuring a relative displacement.

11. The device of

claim 9

or

10

, wherein the light from said incident beam forming said second, third, fourth and fifth beams and finally detected by the detector reaches said first grating at an angle of incidence which is not zero in a plane perpendicular to lines forming the gratings, this angle of incidence being sufficient so that the source providing said light and the detection region of the detector receiving said light are spatially separated from each other in projection in a plane perpendicular to said lines.

12. The device of

claim 7

or

8

, wherein a diffraction region of said first or fourth grating, from which originates said light resulting from said interference and detected by the light detector, is arranged so that other interference, along different diffraction directions to said first direction and originating from different diffraction orders of said fourth and fifth beams than respectively <<+1>> and <<−1>>, have at least one of the two contributions of these fourth and fifth beams whose amplitude is considerably less than the amplitudes of the fourth and fifth beams diffracted along said first diffraction direction in said diffraction region.

13. The device of

claim 7

or

8

, wherein said first grating is arranged, in a region of said first grating receiving the light from said first beam finally detected by said detector, so that the <<0>> diffraction order is relatively low, said first beam being diffracted in this region mostly into said <<+1>> and <<−1>> orders.

14. The device of

claim 1

or

7

, wherein said first grating and, where appropriate, said fourth grating are formed in a dielectric layer of index n greater than 1.8, so as to achieve a larger diffraction efficiency with shallower grating grooves.

15. The device of

claim 1

or

7

, wherein said second grating and, where appropriate, said third grating are formed in a dielectric layer on top of a reflective substrate, so as to achieve a large diffraction efficiency for the TE polarization.

16. The device of

claim 1

or

7

, wherein said first and second gratings, where appropriate said third and/or fourth gratings are each formed of several longitudinal secondary gratings of close but different frequencies allowing an absolute displacement measurement over at least one range of measurement.

17. The device of

claim 1

or

7

, wherein it further includes at least one diffraction grating of increasing and/or decreasing period, arranged beside at least one of said first and second gratings, where appropriate said third and fourth gratings so as to define at least one reference position for said detector or for another detector provided for this purpose.

18. The device of

claim 1

or

7

, wherein it further includes at least one diffraction grating having at least one offset or phase jump in the arrangement of its lines so as to define at least one reference position for said detector or for another detector provided for this purpose.

19. The device of

claim 1

or

7

, wherein it is arranged for measuring the relative velocity between said first and second gratings, the sole measurement of the frequency of the detected luminous intensity modulation providing said relative velocity.

20. The device of

claim 1

or

7

, wherein at least one grating among said first and second gratings, and where appropriate said third and fourth gratings has a region where its lines are offset or phase shifted relative to the rest of this grating or is formed of at least two secondary gratings of the same period and of phase shifted or offset lines between these secondary gratings, this phase shift or offset being provided so that said light resulting from said interference has two partial beams or two distinct beams whose alternating luminous intensity signals, which varies as a function of the relative position between a first portion attached to said source and a second mobile portion relative to said first portion, are phase shifted, in particular by Π/2, to allow detection of the relative displacement direction between these first and second portions and interpolation in an electric period of the luminous intensity signals.

21. The device of

claim 4

, wherein said light source is formed of an electroluminescent diode.

22. The device of

claim 21

, wherein it includes an optical collimation element arranged between said source and said first grating.

23. The device of

claim 1

or

7

, wherein said source emits light forming a first partial beam incident upon said first grating at a positive angle of incidence and another partial beam incident upon said first grating at a negative angle of incidence, said first and second gratings, and where appropriate, said third and fourth gratings, being provided on either side of the two regions of incidence of said first and second partial beams incident upon said first grating so as to form on either side said first to fourth beams and to generate on either side said interference between said fourth and fifth diffracted beams, the light resulting from this interference being detected on either side by at least two detectors also arranged on either said of said regions of incidence.

24. The device of claims7 or8, wherein said source emits light forming a first partial beam incident upon said first grating at a positive angle of incidence and another partial beam incident upon said first grating at a negative angle of incidence, said first and second gratings, and where appropriate, said third and fourth gratings, being provided on either side of the two regions of incidence of said first and second partial beams incident upon said first grating so as to form on either side said first to fourth beams and to generate on either side said interference between said fourth and fifth diffracted beams, the light resulting from this interference being detected on either side by at least two detectors also arranged on either said of said regions of incidence.

25. The device of

claim 23

, wherein said source is attached to said second and, where appropriate, third gratings of which useful portions situated on either side of said source are offset or phase shifted relative to each other so that the alternating light signals resulting from said interference and detected respectively by the two detectors are phase shifted, in particular by Π/2, in relation to each other.

26. The device of

claim 23

, wherein it further includes a fifth diffraction grating arranged between said source and said first grating, this fifth grating diffracting mostly into the <<+1>> and <<−1>> orders respectively on either side of a direction perpendicular to said first grating.

27. The device of

claim 26

, wherein said source provides a substantially collimated beam propagating along a direction substantially perpendicular to said first grating.

28. The device of

claim 1

or

7

, wherein at least said first or second grating defines a bi-directional diffraction grating of the same spatial period along said two orthogonal axes.

29. The device of

claim 1

, wherein it includes at least first and second reflective surfaces, the first reflective surface being arranged to deviate said first beam, originating from said source and propagating substantially along said displacement direction, in the direction of said first grating in order to provide said incident beam, said second reflective surface being arranged to reflect said light interfering along said output diffraction direction in a direction substantially parallel to said displacement direction before being received by said detector.

30. The device of

claim 29

, wherein said source and said detector are attached to said first and, where appropriate, fourth gratings and said first and second reflective surfaces being formed on a rod supporting said second and, where appropriate, third gratings.

Images