The present invention relates to a beam director, and in particular to a beam director for a 3D printer including a first rotating reflector and a second rotating conical reflector.
A beam director for use in 3D printers, such as the one disclosed in U.S. Pat. No. 9,435,998, which is incorporated herein by reference, comprises a first mirror rotating about its longitudinal axis, with a reflective surface at an acute angle to the longitudinal axis. Accordingly, a beam transmitted along the longitudinal axis may be redirected onto a second mirror, and then to a work surface, which is typically perpendicular to the longitudinal axis.
The second mirror may take the form of a rotating flat mirror or a stationary arcuate mirror, which is used to reflect the beam along straight or arcuate paths on the work surface.
Lower case x, y and z denotes local beam coordinates:
In this application we define a lower case x, y and z coordinates system that is cartesian coordinates that are local to the beam. Lower case z denotes the direction of the beam while x and y denotes the beam size. Therefore, any reference to lower case x, y or z axis will be referring to local coordinates of the beam.
Upper case X, Y and Z denotes global system coordinates shown in
With reference to
The beam will keep moving in the Z direction (90° shift); caused by the 45° cone.
When the source beam 3 is collimated, the beam x component 3B will no longer be parallel to the Z direction, as the cone curve will bring into focus the x component because of the rule of deflection as shown in
The cone curve along the radius of 132 will add an optical diversion in the x axis that is proportional to the radius of the conical reflector 132. As an example, this will add a focal point proximate to the work surface in the x axis only, thus causing the departing beam 107A from the conical reflector 132 to have different focal point for x and for y. This is an undesired result: the desired result is either a collimated beam or a focusing beam in both x and y axis. To clarify; see
Accordingly, an object of the present invention is to address the optical components for handling the beam of the prior art by providing corrective elements, whereby the beam has the same dimension in the first and second directions when incident on the work surface or when the beam keeps the proportion between the first and second directions when incident on the work surface.
The present invention relates to a beam director comprising:
a rotatable first reflector rotatable about a longitudinal axis for receiving a beam from a beam source along the longitudinal axis, the first reflector including a reflective surface at an acute angle to the longitudinal axis for reflecting the beam;
an actuator for rotating the first reflector about the longitudinal axis, whereby the first reflector rotates and reflects the beam at a constant angle to the longitudinal axis;
a second conical reflector rotated by the actuator and rotating in a circle around the longitudinal axis of first reflector as first reflector rotates; constantly facing the first reflector at a constant angle; the second reflector further configured to reflect the beam towards a work surface at a constant angle thereto,
whereby when the beam is activated and the actuator rotates the first and second reflectors, the beam strikes the rotating first reflector rotating the beam and reflecting the beam to the second reflector, which reflects the beam to the work surface; the beam then following a curve path relative to the work surface, tracing out an arc on the work surface.
With reference to
For the standard conical reflector 2, the beam 3 leaving the conical reflector 2 will get distorted because of the differences in focal lengths between the x and y axes. As a result the round beam launched from the light source 3 will become a distorted elliptical shape denotes by (xfinal, yfinal) as it hits the work area 1, as shown in
This distortion is a result of the curved surface of the conical reflector 2 in the x axis shown in
Solution(s):
Method I:
Adding a cylindrical lens 12 between the light source and the work surface 1, preferably between the light source and the first reflector 4 to correct the beam 3 in the y direction so it will match the x component at the work area 1. Shown in
As an example:
R=100 mm
Distance to work area (7)=100 mm; this is also the focus fx of the x component. We should choose a cylindrical lens with focal lens of more than 200 mm. We can choose cylindrical mirror with f=250 mm and marked it as fy. Now, positioning the cylindrical lens in a distance of 50 mm from the rotating mirror 4 will provide a sound solution where the fx and fy will land at the working area.
Method II:
In this method the x dimension will diverge prior to hitting the second reflector. When the expanded beam 3B will hit the second reflector it will be corrected back to a collimated beam.
Adding a diverging cylindrical lens 12 between the light source and the work surface 1, preferably between the light source and the first reflector 4, to diverge the x dimension of the beam 3 to produce a diverging 3A beam on the x dimension. After reflecting from first reflector 4, 3B will continue to diverge when hitting the second reflector 2. The second reflector 2 will converge the diverging beam. The convergence rate is dictated by the focal point of the second reflector (R) while the initial divergence rate is dictated by the cylindrical lens. Similarly to method I, the divergence rate can be calculated to achieve a symmetrical beam on the work surface.
Method III:
Conic Sections:
Conic sections are mathematical curves (parabolas, hyperbolas, circles etc.) that satisfy quadratic algebraic expressions (See Optical System Design, Robert E. Fischer Second Edition, Chapter 7 Page 117). Geometrically they are equivalent to the intersection of a cone with a plane, hence the name. When a conic section is rotated about an axis, it sweeps out a surface in three dimensions (paraboloid, hyperboloid, sphere or ellipsoid). Surfaces of this type are very useful in optics and defined by the following equation in the Vertex origin Cartesian coords:
z(r)=Cr2/(1+√{square root over ((1−(1+Asph)C2r2)}
Where: r2=x2+y2
where:
C: is the curvature of a surface at the coords origin; 1/radius
r radius is the distance from Z axis to the surface
Asph: asphericity, departure from spherical profile Asph=Shape−1
By using the conic section equation in this method to modify the reflective surface of the conical reflector 2 in the x direction by providing the y surface with matching curvature to the x curvature. Enforcing a deflection of 90° by choosing the surface slope/derivative to be 45° or 135°.
For convenience we enforced 135°, the invention allows a wider range of deflection angles.
z(r)=Cr2/(1+√{square root over ((1−(Asph+1)C2r2)}) (1)
r2=x2+y2 (2)
z′(r)=−1 (3)
z(r)=−Cr√{square root over ((1−C2(Asph+1)r2)}/(C2(Asp+1)r2−1) (4)
C2r2=1−C2(Asph+1)r2 (3)(4)
Asph=1/C2r2−2 (3)(4)
Selecting work distance between rotating reflector 2 to second reflector 5
r=R (5)
Asph(C)=1/C2R2−2 (5)(3)(4)
Asph(C)=1/C2R2−2 (6)
Selecting C=1/R will resolve to Asph=−1 which is a paraboloid.
The above specifications will deliver a sound solution to correct beam 3 as it reaches the work surface 3C.
As the beam is circulating the Conic Section it will satisfy the requirement for reflector 2 to deflect the beam when beam 3A is circling. The deflection of deflection of 90° is achieved by choosing the surface slope/derivative to be 45° or 135° (derivative to be 1 or −1).
As an example let's use R=125 mm→Curvature=0.008 while Asph=−1
The results are clearly demonstrated where the beam 3 on the surface 1 represented by xfinal and yfinal and showing a nice round beam
Implementing same matching curvature is one option. However, selecting other values for y curvature can be used when the source beam 3 is elliptical as it is the case with many laser diodes.
The scanning system is further improved as:
Similarly, Z(r) the curve function can alternatively be represented by polynomial coefficients to fulfill the imposed conditions.
z(r)=Cr2/(1+√{square root over (1−(Asph+1)C2r2)}+A2r2+A4r4+ . . . +Ajrj
Additionally or alternatively, we can use only polynomials for the complete definition of the surface while imposing the required specifications.
z(r)=A0r0+A1r1+A2r2+A3r3+ . . . +Ajrj
Polynomial terms are useful on their own, without curvature or asphericity, in polynomial optics such as Schmidt correctors. More often they are combined with curvature and asphericity to provide small high-order corrections to a surface. Usually, only the first few even coefficients are sufficient as A2, A4, and A6 since most optical surfaces will be very nearly approximated by the conic aspheric profile.
z(r)=Cr2/(1+√{square root over ((1−Asph+1)C2r2)}+A2r2+A4r4+ . . . +Ajrj
Method IV:
A surface of the conical reflector 2 that fulfills the required conditions for any size beam and working radius. This surface can be formed as a composition of multiple sliced surfaces. Where each slice “i” has its own “Ri”, this way we use the aspherical equation for each slice with its own parameter. C will now be C(Ri) as C depends on R.
with a Z(r) function where C is the curvature will be defined as C(R(r)). C is a variable that is dependent on r which depends on work distance from the origin z.
One way to define this surface is by setting multiple R points into a spreadsheet and calculating the C results by using the conical curve surface equation to create the slice.
The aspherical equation now is simplified by setting Asph=−1 and setting C=1/ri where ri is the work distance (the place the beam will be deflected):
z(r)=Cr2/(1+√{square root over ((Asph+1)C2r2)})
Setting C(ri)=1/ri
z(ri)=ri/(1+√{square root over (1))}
z(ri)=ri/2
Now for each ri we calculate the slice surface z for it.
Similarly, we can define the surface with further correction using polynomials where we calculate the polynomial coefficients be setting multiple points for R.
z(r)=Cr2/(1+√{square root over ((1−Asph+1)C2r2)}+A2r2+A4r4+ . . . +Ajrj
Incorporating the focus properties into the mirror 4 will improve aberration and calibration of the system as a single element performs two functions.
The present application claims priority from U.S. Patent Applications 61/416,763 filed Nov. 3, 2016, which is incorporated herein by reference.
Number | Name | Date | Kind |
---|---|---|---|
4699447 | Howard | Oct 1987 | A |
5177347 | Wike, Jr. | Jan 1993 | A |
5742402 | Kobayashi | Apr 1998 | A |
6085980 | Dickson | Jul 2000 | A |
9435998 | Bibas | Sep 2016 | B1 |
Number | Date | Country | |
---|---|---|---|
20180120556 A1 | May 2018 | US |
Number | Date | Country | |
---|---|---|---|
62416763 | Nov 2016 | US |