FIELD OF THE INVENTION
This invention relates to deflection of an optical beam for measuring angular displacement or for beam scanning.
BACKGROUND
Measuring an angular displacement of a test object by reflecting a beam of light from the test object and measuring the beam deflection corresponding to the angular displacement is well known in the art. The basic principles of this method are shown on FIG. 1. An incident light beam 102 is reflected from a surface 112 to provide a reflected beam 114. If the angle of surface 112 changes, e.g., to align with dotted line 122, incident light beam 102 would be reflected along path 124. If the angular displacement of the reflective surface is ε, the corresponding difference in reflected beam direction is 2ε, which follows from the geometry of FIG. 1 and the reflection law (θi=θr).
It is also known in the art to measure angular displacement using optical systems that employ diffraction instead of or in addition to reflection. For example, U.S. Pat. No. 4,330,212 considers illumination of a diffraction grating on a test object. Changes in the position of the resulting diffraction pattern correspond to angular displacement of the test object. For example, a roll (rotation about the grating surface normal) of the test object will provide a corresponding roll of the diffraction pattern. A similar approach is also considered in U.S. Pat. No. 7,110,103.
In these references, diffraction is exploited to provide a pattern having multiple spots as opposed to a single reflected beam. Such a multiple spot pattern can provide information on angular displacements that cannot be measured with a single beam approach, such as rotation about the surface normal of the test object.
For the configuration of FIG. 1, the displacement of the reflected beam on a detector is 2εL, assuming a reflected beam working distance of L from the test object to the detector. In some applications, such as space based gravitational wave sensors, it is simultaneously required to provide very high angular sensitivity in combination with a relatively limited working distance L. In such cases, it can be difficult or even impossible to meet the combined sensitivity and working distance requirements, thereby requiring an undesirable design compromise.
Accordingly, it would be an advance in the art to provide optical measurement of angular displacement having improved angular sensitivity.
SUMMARY
An angular magnification effect of diffraction is exploited to provide improved sensing and scanning. This effect is most pronounced for a normal or near-normal incidence angle in combination with a grazing diffraction angle, so such configurations are preferred. Angular sensitivity can be further enhanced because the width of the diffracted beam can be substantially less than the width of the incident beam. Normal incidence configurations with two symmetric diffracted beams are preferred, since rotation and vertical displacement can be readily distinguished. Increased sensitivity to vertical displacement can be provided by incorporating an interferometer into the measurement system. Quad cell detectors can be employed to provide sensitivity to rotation about the grating surface normal. A 2-D grating can be employed to provide sensitivity to angular displacements in two different planes (e.g., pitch and yaw). Combined systems can provide sensitivity to vertical displacement and to all three angular degrees of freedom.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 shows a known arrangement for measuring angular change.
FIG. 2 is a diagram showing principles of the invention.
FIG. 3 is another diagram showing principles of the invention.
FIG. 4 shows a first embodiment of the invention.
FIG. 5 shows a second embodiment of the invention.
FIGS. 6
a-b show operation of the embodiment of FIG. 5.
FIG. 7 shows a third embodiment of the invention.
FIG. 8
a-b show detector configurations suitable for use in embodiments of the invention.
FIG. 9 shows a fourth embodiment of the invention.
FIGS. 10
a-c show several views of a fifth embodiment of the invention.
FIG. 11 shows a sixth embodiment of the invention.
DETAILED DESCRIPTION
To better appreciate the principles of the present invention, it is helpful to first consider the conventional geometry of FIG. 1. Incident beam 102 is reflected by surface 112 to provide a reflected beam 114. If an angular displacement of surface 112 shifts it to a new position 122, the corresponding reflected beam would be beam 124. The surface normal of surface 112 is shown as 116, while the surface normal of the surface 122 is shown as 126. Thus displacing surface 112 by ε causes a corresponding change in the angle of incidence, Δθi, which is equal to ε. From the law of reflection (θi=θr), it follows that the corresponding change in the angle of reflection Δθr is also equal to ε. Because the angle between dotted line 130 and beam path 114 is ε, the angle between surface normal 126 and dotted line 130 is θr. The new reflected beam path 124 must make an angle with respect to surface normal 126 which is larger by Δθr=ε, as shown. Therefore, the difference between reflected beam path 114 and reflected beam path 124 is 2ε. To summarize, in the arrangement of FIG. 1 an angular displacement of ε causes a corresponding change in reflected angle Δθr which is equal to ε. The resulting deflection of the reflected beam from its original path is 2ε.
A key aspect of the present invention is the discovery that substantially higher angular sensitivity can be obtained in suitably selected diffraction geometries. FIG. 2 is a diagram showing principles of the invention. A normally incident beam 202 is diffracted from a diffraction grating 212. In this example, diffracted beams 214 and 216 correspond to 1 and −1 order diffraction respectively. As before, an angular displacement of grating 212 by ε shifts it to a new position shown schematically by dashed line 222. Dotted lines 230 and 232 differ from beam paths 214 and 216 respectively by ε, which is also the change in angle of incidence shown by surface normal 228.
The angular displacement of grating 212 induces corresponding changes in the 1 and −1 order diffraction angles, Δθ1 and Δθ−1 respectively, shown by beam paths 224 and 226. These changes are determined by the grating equation:
d(sinθm−sinθi)=mλ, (1)
where λ is the optical wavelength, m is the diffraction order, θi is the angle of incidence, θm is the mth order diffraction angle, and d is the grating period. For small angular displacements, Eq. 1 gives
Δθm=MΔθi, (2)
where
The change in diffracted angle is related to the change in incidence angle via a magnification factor M, which in general can be less than 1, equal to one, or greater than one. For increased angular sensitivity relative to the conventional arrangement of FIG. 1, M is greater than one, which implies θi<θm, and preferably M>2. More preferably, normal or near-normal incidence is employed (e.g., θi<10°) and diffraction is at a grazing angle (e.g., θm>75°). Significant angular magnification can be obtained. For example, if θi=0° and θm=80°, Δθm=5.8Δθi. As is well known in the art, selection of θi and θm in accordance with the above principles entails selection of θi, λ, d, and the diffraction order m.
An important advantage of the invention is that increased angular sensitivity is provided “for free”, by the diffraction geometry itself. No further complications such as increased laser power are required to realize this advantage. This angular magnification effect does not appear to be appreciated in previous work relating to diffraction for measuring angular displacement.
Angular sensitivity is further enhanced by another consequence of the diffraction geometry, shown on FIG. 3. Here an incident beam 302 is diffracted from grating 212 to provide diffracted beams 304 and 306. As shown, the width Wd of the diffracted beams is substantially less than the width Wi of the incident beam. Analysis shows that the beam widths are related as follows: Wi=MWd, where M is the magnification factor as defined above. Thus for the preferred range M>2, Wd<0.5 Wi follows. In practice, diffracted beam position is typically measured with a position sensitive photodetector (e.g., a split cell detector). Other things being equal, reducing the beam width at such a detector causes a proportionate increase in position sensitivity, because it is easier to precisely detect a small position change of a narrow beam than of a wide beam. In practice, both angular magnification and beam width reduction can be relevant, thereby providing a sensitivity improvement on the order of M2. For θi=0° and θm=80°, M2=33, and for θi=0° and θm=85°, M2=132.
The above-described principles can be applied in various ways to provide improved angular sensors and scanners. FIGS. 4-11 show several examples of embodiments of the invention.
FIG. 4 shows a first embodiment of the invention. In this example, an optical source 402 provides an incident beam 404 of substantially monochromatic optical radiation to a test body 406. A grating 408 is disposed on test object 406 and is illuminated by incident beam 404. Incident beam 404 is incident on grating 408 at an angle of incidence θi with respect to grating normal 409. Grating 408 provides a first diffracted beam 410 responsive to the illumination by incident beam 404. First diffracted beam 410 has a diffraction angle θd1 with respect to grating normal 409. A first diffraction plane is defined by the incident beam and first diffracted beam.
A first detector 412 receives first diffracted beam 410 and provides a first output Z1 responsive to a position of diffracted beam 410 on detector 412 in the first diffraction plane. Such a position sensitive detector can be provided in various known ways, such as with a split cell detecting element 414.
If the angular orientation of test object 406 is changed by a first change ε1 from a reference orientation of test object 406, corresponding changes Δθi and Δθd1 of the incidence and diffraction angles are induced. As described above, Δθi=ε1 and θi and θd1 are selected such that Δθd1 is substantially greater than ε1, thereby enhancing sensitivity due to the above-described angular magnification effect.
FIG. 5 shows a second embodiment of the invention. This example shows a preferred embodiment where normal incidence and symmetric diffraction into 1 and −1 orders are employed. Accordingly, in this embodiment it is preferred that grating 408 have a period such that at the wavelength of incident beam 404, grating 408 diffracts only into orders 0, 1, and −1. This embodiment is similar to the embodiment of FIG. 4, except that grating 408 also provides a second diffracted beam 420 responsive to the illumination by incident beam 404. Second diffracted beam 420 has a diffraction angle θd2 with respect to grating normal 409. Second diffracted beam 420 is also in the first diffraction plane.
A second detector 422 having a split cell detecting element 424 receives second diffracted beam 420 and provides a second output Z2 responsive to a position of diffracted beam 420 on detector 422 in the first diffraction plane. Angular displacement of test object 406 by an angle ε1 also induces a corresponding change Δθd2 of the diffraction angle θd2 of second diffraction beam 420. The change Δθd2 is also substantially greater than ε1 due to the above-described angular magnification effect.
FIGS. 6
a-b show operation of the embodiment of FIG. 5. FIG. 6a shows the result of an angular displacement ε1, where first diffracted beam 410 moves to a new position 610, and second diffracted beam 420 moves to a new position 620. The corresponding outputs of the position sensitive detectors, Z1 and Z2, change in opposite senses, since first beam 410 moves down on first detector 412 while second beam 420 moves up on second detector 422.
FIG. 6
b shows the result of a vertical displacement Δz, where first beam 410 moves to a new position 610′ and second beam 420 moves to a new position 620′. The corresponding outputs of the position sensitive detectors, Z1 and Z2, change in the same sense, since first beam 410 moves up on first detector 412 and second beam 420 also moves up on second detector 422.
Since angular displacement and vertical displacement have different effects on the detector position outputs, these two perturbations of test body position can be readily distinguished, which is a significant advantageous feature of this embodiment. More specifically, a difference of Z1 and Z2 is responsive to angular displacement ε1 but not to vertical displacement Δz, while a sum of Z1 and Z2 is responsive to vertical displacement Δz but not to angular displacement ε1. Angular sensitivity is enhanced by the dual detector configuration, since both detectors can contribute to the rotation signal. Another advantage of this configuration is that differential mode rotation sensing avoids effects due to laser frequency noise, since such effects give rise to common-mode errors. Extraordinary sensitivity has been achieved in this configuration. In one experiment, 0.1 nrad/Hz1/2 sensitivity was demonstrated using a grating having 935 lines/mm at a wavelength of 1064 nm. The diffraction angles were about 84°, and the working distance was 6 cm. This combination of sensitivity and working distance is suitable for demanding applications, such as space based gravitational reference sensors.
FIG. 7 shows a third embodiment of the invention, where the detectors are modified to provide sensitivity to z-axis rotation Δθz in addition to the y-axis rotation sensitivity shown in connection with FIGS. 5 and 6a. FIG. 7 is a top view (i.e., x-y plane) as opposed to the side view (i.e., x-z plane) of FIG. 5. The embodiment of FIG. 7 is similar to the embodiment of FIG. 5, except that detectors 412 and 422 on FIG. 5 are replaced with detectors 712 and 722 having split cell detector elements 714 and 724 respectively. Detectors 712 and 722 provide position sensitive outputs Y1 and Y2 respectively, responsive to a lateral (i.e. y-directed) change in beam position, in addition to outputs Z1 and Z2 as described above. In more general terms, it is convenient to regard the first diffraction plane as defined with respect to a reference angular orientation of the test object. With this convention, detectors 712 and 722 are responsive to deviation of diffracted beams 410 and 420 from the first diffraction plane (e.g., as shown by beam positions 710 and 720 respectively). The difference of Y1 and Y2 is responsive to the angular displacement Δθz. This embodiment provides measurement capability for Δθz in combination with the enhanced measurement sensitivity for angular displacement ε1 of the embodiment of FIG. 5.
FIGS. 8
a-b show detector configurations suitable for use in embodiments of the invention. FIG. 8a shows a split cell detector 802 having two separate detector elements 804 and 806. Such a detector is suitable for use in the embodiments of FIGS. 4 and 5, to provide vertical position outputs such as Z1 and Z2. FIG. 8b shows a quad cell detector 808 having four separate detector elements 810, 812, 814, and 816. Such a detector is suitable for use in the embodiment of FIG. 7, to simultaneously provide vertical and lateral outputs, such as (Z1, Y1) from detector 712 and (Z2, Y2) from detector 714. Split cell and quad cell detectors suitable for use in embodiments of the invention are well known in the art. Any other method of observing or measuring diffracted beam position can also be employed in practicing the invention.
FIG. 9 shows a fourth embodiment of the invention, where an interferometer is added to the angular sensor to provide more sensitive capability for measuring vertical displacement Δz. This embodiment is similar to the embodiment of FIG. 5, except that a Littrow grating 902 is disposed in the incident beam path. Littrow grating 902 performs two functions. The first function is directing beam 404 to illuminate grating 408 as a normally incident beam 904. The second function is retro-diffracting light traveling along path 906 to propagate along path 908. Any grating arranged to provide such retro-diffraction can be referred to as a Littrow grating. Methods for selecting the grating period and the angles of incidence and diffraction for a given optical wavelength to provide a Littrow grating as described are well known in the art. Since the 0 order light from grating 408 (i.e., the specular reflection) propagates along path 906, the combination of grating 408 and Littrow grating 902 form an optical resonator, with light recirculating along paths 906 and 908.
With this arrangement, the total power in diffracted beams 410 and 420 will depend periodically on the vertical displacement Δz, in accordance with well known principles of interferometry. Detectors 412 and 422 can provide power outputs P1 and P2 which are responsive to Δz. Determination of Δz can be based on four measured quantities: P1, P2, Z1, and Z2.
To maximize sensitivity, it is preferable for the interferometer in the example of FIG. 9 to be impedance matched. More specifically, let η0 be the 0 order diffraction efficiency of grating 408 (i.e., the fraction of normally incident light that is diffracted into 0 order). Let ηL be the Littrow efficiency of grating 902 (i.e., the fraction of incident light along path 906 that is retro-diffracted along path 908). Impedance matching the resonator entails selecting the various design parameters such that η0=ηL.
Sensitivity to all three angular degrees of freedom can be provided by employing principles of the invention in connection with a two-dimensional grating on the test object. FIGS. 10a-c show several views of such an embodiment of the invention.
FIG. 10
a is a top view showing a 2-D grating 1008 on test object 406. Normally incident light on grating 1008 is diffracted into a first diffracted beam 410, a second diffracted beam 420, a third diffracted beam 1010, and a fourth diffracted beam 1020. Beams 410, 420, 1010, and 1020 are received by first, second, third and fourth detectors 712, 722, 1012, and 1022 respectively. These detectors can be quad cell detectors. Detectors 712 and 722 provide outputs Y1 and Y2 responsive to lateral deviation of the first and second diffracted beam from the first diffraction plane, as described in connection with FIG. 7. Similarly, detectors 1012 and 1022 provide outputs X3 and X4 responsive to lateral deviation of the third and fourth diffracted beams from a second diffraction plane. The second diffraction plane (y-z plane) in this example is perpendicular to the first diffraction plane (x-z plane). The differences Y1−Y2 and X3−X4 are responsive to rotation about the grating surface normal Δθz.
FIG. 10
b shows a view in the first diffraction plane (x-z plane). First diffracted beam 410 has a diffraction angle θd1 with respect to grating normal 409. Second diffracted beam 420 has a diffraction angle θd2 with respect to grating normal 409. An angular displacement ε1 of test object 406 in the first diffraction plane (i.e., a rotation about the y axis) provides corresponding changes Δθd1 and Δθd2 in the diffraction angles. The changes Δθd1 and Δθd2 are substantially greater than ε1 due to the above-described angular magnification effect. Detectors 712 and 722 provide position sensitive outputs Z1 and Z2 respectively. As described above in connection with FIGS. 6a-b, the sum Z1+Z2 is responsive to vertical displacement Δz and the difference Z1−Z2 is responsive to angular displacement ε1.
FIG. 10
c shows a view in the second diffraction plane (y-z plane). Third diffracted beam 1010 has a diffraction angle θd3 with respect to grating normal 409. Fourth diffracted beam 1020 has a diffraction angle θd4 with respect to grating normal 409. An angular displacement ε2 of test object 406 in the second diffraction plane (i.e., a rotation about the x axis) provides corresponding changes Δθd3 and Δθd4 in the diffraction angles. The changes Δθd3 and Δθd4 are substantially greater than ε2 due to the above-described angular magnification effect. Detectors 1012 and 1022 provide position sensitive outputs Z3 and Z4 respectively. As described above in connection with FIGS. 6a-b, the sum Z3+Z4 is responsive to vertical displacement Δz and the difference Z3−Z4 is responsive to angular displacement ε2.
The combined sum Z1+Z2+Z3+Z4 can be used as a measure of Δz. Vertical displacement sensing in the embodiment of FIGS. 10a-c can also be enhanced by combining it with vertical interferometry as described in connection with FIG. 9.
Principles of the invention are applicable to angular scanning as well as to angular displacement sensing as described above. FIG. 11 shows a beam deflector or scanner in accordance with an embodiment of the invention. An optical source 1102 provides a substantially monochromatic beam 1104 of optical radiation to a diffraction grating 1108 disposed on a deflector 1106. Beam 1104 is incident on grating 1108 at an angle of incidence θi, and a diffracted beam 1110 is diffracted from grating 1108 at a diffraction angle θd. As described above, an angular displacement ε of deflector 1106 provides corresponding changes Δθi and Δθd in the incidence and diffraction angle. The change Δθd is substantially larger than ε, and Δθi=ε. An angle control 1112 can be employed to vary the angular displacement of deflector 1106. The resulting deflection in diffracted beam angle is >2ε. More specifically, the diffracted beam deflection is ε(1+M), where M>1 is the above-defined magnification factor. A conventional mirror-based beam deflector would provide an reflected beam deflection of 2ε. Such angular magnification can provide increased bandwidth in applications such as laser printing laser scanning, and optical switching.
If scanning is performed over a significant range of angles, it may be important to account for the nonlinearity of the dependence of Δθd on ε. Such nonlinearity can be accounted for by appropriate pre-compensation of angle control 1112. For example, an inverse-cosine pre-compensation can be employed to drive ε in a nonlinear manner such that Δθd changes linearly during a scan.