GB2136163A - Light Beam Stabilizer - Google Patents

Abstract

Apparatus for reducing the effects of pointing instabilities in a light source, e.g. a laser comprises two coaxial lenses 9, 12 sharing a confocal point 7 and comprises a rotatable parallel dielectric (e.g., glass) plate 34 positioned between the lenses and intersecting the axis. A collimated, incoming laser beam 3 is projected to a first of the lenses, focused through the plate to the confocal point, and received and recollimated by a second of the lenses. A beamsplitter 23 diverts part of the recollimated beam toward a photosensor 26. Angular deviation of the incoming laser beam from the axis results in an angular deviation of the recollimated beam as well as in failure by the first lens to focus the laser beam at the confocal point and, further, in a translation of the diverted part of the recollimated beam across the photosensor. A resulting error signal is fed to a motor 36 which rotates the plate to restore focus at the confocal point and to thus reduce angular deviation of the recollimated beam. <IMAGE>

Description

SPECIFICATION Light Beam Stabilizers The invention relates to light beam stabilizers and, more particularly, to such stabilizers which sense an angular deviation of a light beam from a predetermined path and then reduce the deviation.

The light beams projected by some lasers can sporadically deviate at any time in the angular directions in which the light beams travel. In other lasers, such deviations can also occur, but chiefly during warm-up. Such deviations are commonly termed "pointing instabilities". Whiie these instabilities are generally small, of the order of 10 milliradians, in precision optical work they become significant.

It is an object of the present invention to provide a new and improved light beam stabilizer, particularly one for reducing pointing instability effects in lasers.

In one form of the present invention, a detection means detects an angular deviation of an entering light beam and generates an error signal in response. A correction means coupled to the detection means receives the error signal and acts to reduce the angular deviation.

In the accompanying drawings, by way of example only: Figure 1 is a schematic illustration of one light beam stabilizer embodying the present invention; Figure 2 is an illustration of part of the apparatus of Figure 1; Figure 2A illustrates shifting of a light beam by the parallel plate of Figure 2; Figure 3 illustrates a perspective view of a light beam stabilizer embodying the present invention; Figures 4A-B illustrate optical paths in the stabilizer of Figure 1; and Figure 4C illustrates the operation of the plate 39 in Figure 1.

As shown in Figure 1, an entering light beam 3 (which is preferably a laser light beam and is thus highly collimated), is focused as a converging beam 5 to a focus point 7 by a first lens means 9.

Light from the focus point 7 travels towards a second lens means 12 as a diverging light beam 1 5 is recollimated by the second lens means 12 into an exit beam 18 in which the rays, such as rays 18A and 1 8B, are substantially parallel, that is, collimated. A portion 21 of the exit beam 18 is diverted out of the exit beam 18 by a diverting means such as a beamsplitter 23, focused by a third lens 23A, and directed toward a photosensor 26 to form an image on a focal plane of the third lens 23A which is shown as dotted lines 27.

The photosensor 26 preferably comprises two photosensors 26A and 26B separated by a space 26C. The output of each of photosensors 26A and 26B is connected to respective inputs 28A and 28B of a servo amplifier 30 which responds to the difference between the signals presented to the inputs 28A and 28B. The output 33 of the servo amplifier 30 is connected to an actuator means, such as an electric motor 36 taking the form of a D'Arsonval movement. The motor 36 functions to rotate a steering means such as a parallel dielectric plate 39, which is positioned in the path of the converging beam 5 and which is preferably an optically flat glass sheet, so that the parallel plate 39 can occupy the positions shown by phantom outlines 39A and 39B, as well as intermediate positions.

The first and second lens means 9 and 12 share a common focal point 40 which is herein termed a confocal point. That is, distance 43 equals distance 46 and these distances equal the focal lengths of respective lens means 9 and 1 2.

Further, the first lens means 9, the confocal point 40, and the second lens means 12 are positioned on a common optical axis 49 and are thus coaxial.

Still further, as shown in Figure 1 , the confocal point 40 is coincident in space with the focus point 7 under the particular positioning of the components of that Figure and the position of the entering light beam 3.

With reference to Figure 2, the operation of the above-described stabilizer is explained as follows.

Should the entering beam 3 in Figure 1 deviate from a predetermined path, such as from a path defined by the axis 49, and instead follow a deviant path indicated as 3A in Figure 2, the entering beam 3A will be focused by the first lens means 9 as a converging beam 5A, which is a different beam than the converging beam 5 of Figure 1. Upon passing through the parallel plate 39, the converging beam 5A of Figure 2 will be focused at focus point 7A which is not coincident with confocal point 40. Thus, the focus point 7A of the deviant entering beam 3A strays from the confocal point 40. This deviation causes the diverging beam 1 5A to follow the path shown with the result that the exit beam 1 8A is a different beam than the exit beam 1 8 in Figure 1.

Consequently, the light beam portion 21A deflected by the beamsplitter 23 will deviate to follow the path shown as 21 A.

As shown in Figure 2, the deflected portion 21A of beam 1 9A does not intersect the photosensor 26 because Figure 2 is exaggerated for ease of illustration. Preferably, the deflected portion 21 A is offset on the photosensor 26 slightly, as deflected portion 21 B is shown, so that the photosensors 26A and 26B produce unequal signals due to unequal illumination.

Otherwise, a deflected beam such as 21 A would illuminate neither photosensor and the resulting difference signal of zero would be possibly confused with a difference signal of zero produced by the equally illuminated photosensors of Figure 1.

The difference in output signals is amplified by the servo amplifier 30 and the amplified difference is transmitted to the actuating means 36, causing the actuating means 36 to rotate the parallel plate 39 to a position such as that shown in phantom outline 39B. The direction of rotation is in accordance with the algebraic sign of the difference signal. This rotation, by changing the angle at which the converging beam 5A strikes the plate 39 and by changing the amount of plate dielectric material through which the converging beam 5A passes, steers the converging beam 5A to a position such that the focus point 7A in Figure 2 becomes substantially coincident with the confocal point 40. An intermediate stage in this process is shown in Figure 2 in which the focus point 7B is shown en route to the focal point 40.The focus point 7B, the diverging beam 15B, the exit beam 18B, and the deflected beam portion 21 B are all shown in phantom outline in Figure 2 while moving toward their solid outline counterparts 7, 1 5, 18 and 21 in Figure 1.

Therefore, as shown in Figure 2, a portion 5B of the converging beam 5A is moved by the rotation of plate 39 in a direction indicated by arrow 101. Thus, the diverging beam 1 5A is also moved in the direction of arrow 101 by the rotation of the parallel plate 39 until both photosensors 26A and 26B in Figure 2 are substantially equally illuminated so that the difference between their output signals is zero so that the output of the servo amplifier 30 is likewise zero. At this point in time, motion of the parallel plate ceases.

As shown in Figure 2, plate 39, by its rotation to occupy phantom outline 39B, shifts the beam 1 5A to the position 1 5B which is, in this case, nearer to the axis 49 than beam ISA. Further, it is noted that the incoming light beam 3A in Figure 2 is traveling upward, toward the top of the Figure, as the light beam 3A travels toward the right.

Thus, the converging beam 5A and the diverging beam 15A are also traveling upward. Under these circumstances, in order to move the focus point 7A to coincide with the confocal point 40, the portion 5B of the converging beam 5A which is moved in the direction of arrow 101 must be moved toward the axis 49, and some of the portion 5B must cross the axis 49. As shown in Figure 2A, a greater portion of the converging beam 5AA (portion 102) is moved to the lower side of the axis 49 upon rotation of the plate 39 to position 39B than the portion 103 of converging beam 5A which was below the axis 49 before rotation.The reason for this is clear: since the converging light beam 5A is traveling upward as it progresses rightward toward the focus point 7A, and since the rotation of the parallel plate 39 shifts the convering beam 5A parallel to itself (to phantom 5AA), then in order to focus at a focus point 7B which is nearer to the confocal point 40 than is focus point 7A, part 102 of the converging beam 5AA must be shifted across the axis 49.

Stated differently, the light ray (not shown) at the center of the converging beam 5A must be shifted across the axis 49 and done so while within the plate 39 itself. Otherwise, this central ray, in traveling upward as it progresses rightward, will not cross the confocal point 40. The plate 39 must be thick enough and rotated sufficiently to accomplish this shifting of the central ray.

As shown in Figure 3, a D'Arsonval-type movement 104 supports the parallel plate 39.

This D'Arsonval movement is schematically shown in Figure 1 as a pair of gears 105 driven by the motor 36. The plate 39 is supported on a shaft 106 which in turn is supported by a bearing such as a watchmaker's jewel (not shown) contained in a housing 109. The shaft 106 is rotatable about an axis 107 which intersects the optical axis 49. A permanent magnet 112 surrounds the shaft 106 without contacting it.

The permanent magnet 112 provides a first, fixed, magnetic field vector indicated by arrow 113. A wire coil 11 5 is fastened to the shaft 106, surrounds but does not contact the permanent magnet 11 2, is free to rotate with the shaft 106, and is electrically connected to the output 33 of the servo amplifier 30 in Figure 1 (connection not shown in Figure 3). The coil 11 5 provides a second, rotatable, magnetic field vector 1 16 in response to the current of the output 33.

Because of the properties of magnetic fields, the first and second magnetic fields 113 and 1 6 tend to align themselves, so that the existence of the second magnetic field 1 16 tends to rotate the plate 39. However, a spring (not shown) exerts a force biasing the plate 39 into a position perpendicular to the axis 49. Thus, for rotation of the plate 39 to occur, the second magnetic field vector 11 6 must overcome the spring force.

When the spring force is overcome so that rotation does occur, rotation continues until the spring force again controls, since the rotation causes tightening of the spring and a resulting increase in spring force.

A schematic view of some details of the operation of the present stabilizer is shown in Figures 4A-C. In Figure 4A, entering light beams 3C-F are shown as taking different paths which deviate in angle from the optical axis 49. This can result from pointing instabilities when laser light is used. The light beams 3C-F in effect rotate about point 3K. The beams 3C-F are focused by the first lens means 9 as converging beams 5C- F (not shown as pairs of converging lines as in Figure 1 but as single lines) which have focus points 7C-F located in a confocal plane 1 33 which is perpendicular to the optical axis 49 and intersecting with the confocal point 40.The converging beams 5C--F continue past their respective focus points 7C-F as diverging beams 15C--F which are collimated as exit beams 1 8C-F which are substantially parallel to the corresponding incoming beams 3C-F. (The parallelism is not shown for ease of illustration).

Beamsplitter 23 deflects from exit beams 1 8C-F corresponding deflected portions 21 C-F which cast images (not shown) onto different positions of photosensor 26 (not shown in Figure 4A).

As shown in Figure 4B, the faces 144 and 146 of parallel plate 39 are parallel. Thus, a converging beam such as 5CC will leave the plate 39 as a converging beam SCCC which is traveling in the same general direction as converging beam 5CC, but shifted or steered in the direction of arrow 148 which is parallel to the faces 144 and 146. Also, the greater the angle of incidence 1 50 which the converging beam 5CC makes with the normal 1 52 of the surface, the greater will be the shift of beam SCCC in the direction of the arrow 148. One reason for this is the greater amount of materiai of plate 39 through which the converging beam 5CC must travel.

Thus, with respect to the entering beams 3C F in Figures 4A-B, the converging beams 5C--F and 5CC are directed to the confocal point 40 by the rotation of the plate 39 to the positions shown by phantom outlines 39C-F in Figure 4C.

In passing through the confocal point 40, the converging beams SCC-FF are focused by second lens means 12 as exiting beams 1 8C-F which are substantially parallel to the optical axis 49, although displaced siightly to the sides of the optical axis 49.

Of course, the actual light beams used will not behave like the geometrically perfect lines such as 5C mentioned in the above discussion. Thus, the discussion relating to Figures 4A-C represents a theoretical ideal used for explanatory purposes.

Accordingly, from one point of view, a rotatable plate 39 intersects an optical axis 49 and is rotatable in the vicinity of a first magnetic field vector 113 and about an axis 107 which also intersects the optical axis 49. The rotatable plate 39 has associated with it a second magnetic field vector 11 6 and rotation of the plate 39 is induced by the tendency of the second magnetic field vector 11 6 to align itself with the first magnetic field vector 113. The second vector 11 6 is rotatable and changeable as to magnitude and direction. The magnitude and direction depend upon the amplified difference signal produced by the servo amplifier 30 and the amplified difference signal is indicative of the degree of angular deviation of a light beam 3 from the optical axis 49.The rotation of the rotatable plate 39 reduces the influence which the angular deviation of the entering light beam 3 has upon the exit beam 18 in Figure 1. The rotation continues until the exit beam 1 8 produces a deflected light beam 21 which equally iluminates both photosensors 26A and 26B.

The use of two photosensors 26A and 268 separated by a space 26C has been described.

The space 26C functions as a dead zone in the sense that no signal is produced in response to light falling upon the dead zone. Thus, in one form of the invention, if the image cast by the properly aligned exit beam 18 is small enough to fit entirely within the dead zone, then no signal at all is produced by either photosensor 26A or 26B.

Such a situation can be desirable to reduce oscillations in the situation where high precision is desired. Further, photosensors 26A and 26B generally have differing sheet resistances at different positions on their surfaces and they also generally have thermal gradients across them such that the Fermi levels of different regions of the photodetectors will be different. Thus, even if both photosensors are identically illuminated, there will probably exist minute differences in the signals produced. Therefore, it is preferable to treat the exit beam 18 as being correctly positioned when the deflected beam portion 21 falls upon the dead zone so that neither photosensor 26A nor 26B is illuminated.Thus, a zero output signal from both photosensors 26A and 26B is interpreted as indicating that the exit beam 1 8 is properly aligned, subject to the exception discussed above in connection with the exit beam 1 8A of Figure 2. However, it is foreseen that, in the future with the advent of improved photosensors, a photosensor lacking a dead zone can be used. However, it is presently believed that dead zone can be eliminated only at the cost of some accuracy.

From one point of view, the first and the second lens means 9 and 12, the beamsplitter 23, and the photosensor 26 in Figure 1 comprise a sensing means for detecting the angular position of the exit beam 1 8, with respect to the optical axis 49 as a reference. This is tantamount to detecting the position of the focus point 7. The servo amplifier 30, the motor 36, and the plate 39 can be viewed as a correction means for reducing the influence of the angular deviation of the entering beam 3 upon the exit beam 18, but without affecting the position of the source (not shown) of the entering light beam 3. The position of the exit beam 1 8 is determined by the position of the image 27 at the intersection of the exit beam 18 with the photosensor 26.Thus, insofar as the surface 1 23 in Figure 2 of the photosensor 26 can be viewed as a geometric plane, the angular displacement of the entering beam 3 is transformed into a lateral displacement of an image in a plane. Further, while this geometric plane associated with photosensor 26 is not located within the exit beam 1 8 itself, no difference in principle is seen from the position of the photosensor 26 shown in Figure 1 and the position shown as phantom outline 1 26 in that Figure. Of course, if the photosensor 26 is positioned as phantom outline 126, the exit beam 1 8 would be at least partially obstructed.

Parallel plate 39can also be positioned on the opposite side of the confocal point 40, namely at the position indicated by phantom outline 129, rather than on the side shown by the solid outline 39.

The present invention can be utilized in conjunction with the inventions described in our copending patent applications: "Optical Inspection System", serial number filed , 1 983, and "Optical Projector", serial number ,filed 1983.

A stabilizer has been described which senses angular deviations of an entering light beam from a predetermined path and in reponse modifies the entering beam to produce a light beam derived from it, namely exit beam 18, which travels along and substantially parallel to a predetermined path, namely, the optical axis 49.

Numerous substitutions and modifications can be undertaken within the scope of the present invention including the use of lenses 9 and 12 of different sizes from each other. In addition, the above discussion describes a stabilizer which stabilizes the entering light beam 3 as to angular deviations in a single plane, namely the plane of the paper on which Figure 1 is drawn. The addition of a second plate analogous to the plate 39 and positioned at phantom outline 1 29 can serve, together with plate 31, to stabilize against deviations of the light beam 3 in all planes. The second plate (not shown) rotates about an axis which is perpendicular to that of the plate 39. The photosensor 26 is replaced with a quadrant-type photosensor (not shown) having four elements, two of which are connected to the servo amplifier 30 which controls the rotation of the plate 39.

The other two of the four elements are connected to an analogous servo amplifier (not shown) which drives an actuating means (not shown) which is analogous to the actuating means 36.

The actuating means 36 rotates the plate 39 about one axis and the analogous actuating means rotates the analogous plate about an axis perpendicular to the axis of the plate 39. Thus, angular deviation of the entering beam 3 in any plane can be compensated.

It has been state that the first and second lens means 9 and 12 are coaxial. This is not a strict requirement. Two other arrangements, as well as possibly others, will suffice. A first arrangement positions these two lenses so that both of their focal planes are coincident, thus making both optical axes either coincident or parallel. A second arrangement positions the two lenses so that both of their focal points are coincident, thus making both optical axes either colinear or intersecting.

Claims (1)

1. Light beam stabilizer, comprising: (a) detection means for detecting an angular deviation of an entering light beam from a predetermined reference, and for generating an error signal in response, and (b) correction means coupled to the detection means for substantially removing the influence of the angular deviation in a light beam derived from the entering light beam in response to the error signal.

2. Stabilizer according to claim 1 in which the detection means comprises: (a) a first and a second lens means having a confocal point, the first lens means being for receiving the entering light beam of 1 (a) and the second lens means being for transmitting the derived light beam of 1(b), and (b) a photosensitive element for generating a signal indicative of the position of the intersection of the derived light beam with a predetermined reference plane.

3. Stabilizer according to claim 2 in which the correction means comprises a D'Arsonval movement driving a rotatable parallel plate through which the entering light beam passes.

4. Apparatus having an optical axis and for stabilizing an entering light beam, comprising: (a) a first lens means for focusing the entering light beam to a focus point on a confocal plane, the position of the point being determined by the angle between the entering light beam and the optical axis, (b) sensing means for sensing the position of the focus point and for generating an error signal indicative thereof, (c) a second lens means for receiving light from the confocal plane and for transmitting the received light as a substantially collimated light beam, and (d) steering means for steering the light focused by the first lens means to a predetermined point in response to the error signal.

5. Apparatus for stabilizing a light beam produced by a source, comprising: (a) sensing means for sensing the angular deviation of the path of the light beam from a predetermined path, and (b) correction means coupled to the sensing means for steering the light beam toward the predetermined path without moving the source of light in response to the deviation sensed.

6. Apparatus according to claim 5 in which the sensing means comprises a deflector for deflecting part of the light beam along a second path and photodetection means for detecting a displacement of the deflected part from a predetermined position.

7. Apparatus according to claim 5 in which the correction means comprises a rotatable parallel plate.

8. Apparatus having an optical axis and for stabilizing an entering light beam, comprising: (a) a first lens means for focusing the entering light beam to a point whose location is dependent upon the angle between the entering light beam and the optical axis, (b) sensing means for (i) receiving the focused light beam of (a) and transmitting it as a substantially collimated exit beam, (ii) sensing the location of the point of (a) and generating an error signal indicative thereof, and (c) correction means for steering the entering light of (a) to a predetermined point in response to the error signal.

9. Apparatus having an optical axis and for stabilizing an entering light beam, comprising: (a) a first lens means for providing a focused beam in response to the entering light beam, (b) a second lens means for providing a substantially collimated beam in response to the focused light beam, (c) a beamsplitter positioned in the path of the collimated beam of (b) for diverting some of this collimated beam to a photosensing means which can generate an error signal in response to angular deviation of the entering light beam, from the optical axis, (d) an amplifier coupled to the photosensing means for generating an output signal in response to the error signal of (c), and (e) a rotatable plate positioned in the path of the focused beam of (a) and coupled to the amplifier of-(d) for rotating the focused beam in response to the output signal of (d).

10. Apparatus according to claim 9 in which the first and second lens means have the same focal length.

1 Apparatus according to claim 10 in which the first and second lens means are coaxial and share a common focal point.

1 2. Apparatus according to claim 9 in which the rotatable plate comprises an optically flat glass plate.

13. Apparatus according to claim 9 in which the photosensing means includes two photosensitive elements separated by a space and which produce as an output signal two signals whose difference is indicative of the deviation of the diverted beam of (c) from a predetermined path.

14. Method of stabilizing a light beam, comprising the steps of: (a) focusing the light beam through a plate using a first lens means to a focus point on a confocal plane and thence to a second lens means for transmission of an exit light beam, (b) sensing a deviation of the focus point from a predetermined point, (c) rotating the plate in response to the deviation to reduce the deviation.

15. Method according to claim 14 in which the plate comprises two parallel faces.

16. Method according to claim 14 in which the first and second lens means are positioned from the confocal plane at distances equal to their respective focal lengths.

17. Method according to claim 14 in which the sensing step of (b) includes the deflection of part of the exit beam to a photosensor.

18. Method according to claim 14 in which the photosensor comprises two photosensitive elements, each producing a signal, the difference between which is indicative of the deviation of (b).

1 9. A light beam stabilizer substantially as herein described with reference to the accompanying drawings.