U.S. patent application number 11/643839 was filed with the patent office on 2008-04-24 for device for concentrating or collimating radiant energy.
Invention is credited to Pablo Benitez Gimenez, Juan Carlos Minano Dominguez.
Application Number | 20080092879 11/643839 |
Document ID | / |
Family ID | 8310804 |
Filed Date | 2008-04-24 |
United States Patent
Application |
20080092879 |
Kind Code |
A1 |
Minano Dominguez; Juan Carlos ;
et al. |
April 24, 2008 |
Device for concentrating or collimating radiant energy
Abstract
This invention consists in a nonimaging device for concentration
or collimation of radiation on a receiver or from an emitter (14),
depending on the case. The device is made up of the lens (50),
which surrounds the receiver and consists of the aspheric surface
(21), and the lens (15), whose upper refractive surface (16) may be
aspheric, while the lower surface is aspheric (17) in its central
portion (between points 18 and 19) and has a structure with
discontinuous slope (20) in its external portion, in which the
faces (22) fundamentally refract the rays while the faces (23)
reflect them by total internal reflection. The design method
provides that the device properties of concentration/collimation
are noticeably superior to those of the existing inventions.
Possible applications of this lens include: radiation sensors,
illumination systems with LEDs, wireless optical communications and
photovoltaic solar energy.
Inventors: |
Minano Dominguez; Juan Carlos;
(Madrid, ES) ; Benitez Gimenez; Pablo; (Madrid,
ES) |
Correspondence
Address: |
CLARK & BRODY
1090 VERMONT AVENUE, NW
SUITE 250
WASHINGTON
DC
20005
US
|
Family ID: |
8310804 |
Appl. No.: |
11/643839 |
Filed: |
December 22, 2006 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
10148736 |
Oct 15, 2002 |
7160522 |
|
|
PCT/ES00/00459 |
Dec 1, 2000 |
|
|
|
11643839 |
Dec 22, 2006 |
|
|
|
Current U.S.
Class: |
126/699 |
Current CPC
Class: |
G02B 3/08 20130101; Y02E
10/44 20130101; F24S 23/00 20180501; G02B 19/0061 20130101; G01J
1/0411 20130101; G01J 1/04 20130101; G02B 19/0042 20130101; G02B
19/0028 20130101; G02B 17/006 20130101; F24S 23/31 20180501; G02B
19/0076 20130101 |
Class at
Publication: |
126/699 |
International
Class: |
F24J 2/08 20060101
F24J002/08 |
Foreign Application Data
Date |
Code |
Application Number |
Dec 2, 1999 |
ES |
P9902661 |
Claims
1. Device for concentrating radiant energy wherein said device is
axisymmetrical or cylindrical and is configured to transform the
edge rays of an input extended ray bundle into edge rays of an
output extended ray bundle that illuminates a receiver, the optical
spread of said input and output bundles being greater than ten
degrees at one or more of the optical surfaces of said device and
said bundles being defined in the plane of a cross-section by: a) a
lens L.sub.1 comprising on one side a refractive aspheric surface,
S.sub.1, on which the input bundle impinges, and on the other side,
S.sub.2, another refractive aspheric surface in its central region
and a discontinuous-slope structure in its external region, said
discontinuous-slope structure cross-section comprising teeth with
two aspheric faces such that one of them, V, is configured to be
parallel to the flow lines of the bundle transmitted by said
surface S.sub.1, and the other face, T, is configured to reflect
the bundle by total internal reflection toward the face V where it
is refracted so that no ray intercepts the adjacent tooth and that
the nearest edge ray to do so is tangent to that tooth, and b) A
second lens L.sub.2 that surrounds the receiver composed of a
refractive aspheric surface on which the bundle transmitted by the
lens L.sub.1 impinges.
2. Device for concentrating radiant energy according to claim 1
wherein said receiver comprises an optoelectronic receiver.
3. Device for concentrating radiant energy according to claim 1
wherein said receiver is one of a photodiode, a phototransistor or
a photovoltaic cell.
4. Device for concentrating radiant energy according to claim 1
wherein the profiles of the faces of the teeth have at each point a
slope modified by an angle of less than 2 degrees.
5. Device for concentrating radiant energy according to claim 1
wherein the profile of S.sub.1 is circular or flat.
6. Device for concentrating radiant energy according to claim 1
wherein the surfaces S.sub.1 and S.sub.2 of the lens are
interchanged, so that the teeth appear inverted.
7. Device for concentrating radiant energy according to claim 1
wherein S.sub.1 has a saw-toothed profile that diverts the input
bundle to modify the direction of the flow lines.
8. Device for concentrating radiant energy according to claim 1
wherein S.sub.1 or the refractive surface of the central portion of
S.sub.2, or both, are discontinuous-slope Fresnel structures.
9. Device for concentrating radiant energy according to claim 1
wherein the lens L.sub.2 comprises two different dielectric
materials separated by a spheric or aspheric refractive
surface.
10. Device for concentrating radiant energy according to claim 1
wherein the cross-sections of the teeth of S.sub.2 have faces with
circular or rectilinear profiles.
11. Device for concentrating radiant energy according to claim 1
comprising an optically inactive portion joining the two lenses so
that they constitute a single part that includes an interior
space.
12. Device for concentrating radiant energy according to claim 1
fixed to a dielectric plate.
13. Device according to claim 1 wherein at least one of said
surfaces S.sub.1, S.sub.2, or S.sub.3 comprises a Cartesian
oval.
14. Device for collimating radiant energy wherein said device is
axisymmetrical or cylindrical and is configured to transform the
edge rays of an input extended ray bundle generated by an emitter
into edge rays of an output extended ray bundle, the optical spread
of said input and output bundles being greater than ten degrees at
one or more of the optical surfaces of said device, both of said
bundles being defined in the plane of a cross-section, by: a) a
lens L.sub.2 that surrounds the emitter and comprises a refractive
aspheric surface S.sub.3 on which the input bundle impinges, and b)
a second lens L.sub.1 comprising on one side a refractive aspheric
surface, S.sub.1, from which the output bundle leaves, and on the
other side, S.sub.2, another refractive aspheric surface in its
central region and a discontinuous-slope structure in its external
region, said discontinuous-slope structure cross-section comprising
teeth with two aspheric faces such that on one of them, V, the
bundle transmitted by said surface S.sub.3 is refracted so that all
the rays are reflected by total internal reflection on the other
face, T, that the edge ray nearest to not being reflected is
tangent to the profile of the tooth, and that the face V is
parallel to the flow lines of the bundle transmitted toward
S.sub.1.
15. Device for collimating radiant energy according to claim 14
wherein said emitter comprises an optoelectronic emitter.
16. Device for collimating radiant energy according to claim 14
wherein said emitter is one of an LED, an IRED or a laser.
17. Device for collimating radiant energy according to claim 14
wherein the profiles of the faces of the teeth have at each point a
slope modified by an angle of less than 2 degrees.
18. Device for collimating radiant energy according to claim 14
wherein the profile of S.sub.1 is circular or flat.
19. Device for collimating radiant energy according to claim 14
wherein the surfaces S.sub.1 and S.sub.2 of the lens are
interchanged, so that the teeth appear inverted.
20. Device for collimating radiant energy according to claim 14
wherein S.sub.1 has a saw-toothed profile that diverts the input
bundle to modify the direction of the flow lines.
21. Device for collimating radiant energy according to claim 14
wherein S.sub.1 or the refractive surface of the central portion of
S.sub.2, or both, are discontinuous-slope Fresnel structures.
22. Device for collimating radiant energy according to claim 14
wherein the lens L.sub.2 comprises two different dielectric
materials separated by a spheric or aspheric refractive
surface.
23. Device for collimating radiant energy according to claim 14
wherein the cross-sections of the teeth of S.sub.2 have faces with
circular or rectilinear profiles.
24. Device for collimating radiant energy according to claim 14
comprising an optically inactive portion joining the two lenses so
that they constitute a single part that includes an interior
space.
25. Device for collimating radiant energy according to claim 14
fixed to a dielectric plate.
26. Device according to claim 14 wherein at least one of said
surfaces S.sub.1, S.sub.2, or S.sub.3 comprises a Cartesian
oval.
27. Device for collimating radiant energy, said device being
axisymmetrical or cylindrical and configured to transform the edge
rays of an input extended ray bundle generated by an emitter into
edge rays of an output extended ray bundle, both bundles being
defined in the plane of a cross-section, excluding the axisymmetric
case in which the bundles of rays are chosen to provide uniform
irradiance in three dimensions at the exit aperture when the
angular spread of the ray bundles on passing through all of the
optical surfaces is smaller than 10.degree., by: a) a lens L.sub.2
that surrounds the emitter comprising a refractive aspheric surface
S.sub.3 on which the input bundle impinges, and b) a second lens
L.sub.1 comprising on one side a refractive aspheric surface,
S.sub.1, from which the output bundle leaves, and on the other
side, S.sub.2, another refractive aspheric surface in its central
region and a discontinuous-slope structure in its external region,
whose cross-section is formed of teeth with two aspheric faces such
that on one of them, V, the bundle transmitted by surface S.sub.3
is refracted so that all the rays are reflected by total internal
reflection on the other face, T, that the edge ray nearest to not
being reflected is tangent to the profile of the tooth, and that
the face V is parallel to the flow lines of the bundle transmitted
toward S.sub.1.
28. Device according to claim 27 wherein at least one of said
surfaces S.sub.1, S.sub.2, or S.sub.3 comprises a Cartesian
oval.
29. Device for concentrating radiant energy wherein said device is
axisymmetrical or cylindrical and is configured to transform the
edge rays of an input extended ray bundle into edge rays of an
output extended ray bundle that illuminates a receiver, the optical
spread of at least one of said input and output bundles being
greater than ten degrees at one or more of the optical surfaces of
said device and said bundles being defined in the plane of a
cross-section by: a) a lens L.sub.1 comprising on one side a
refractive aspheric surface, S.sub.1, on which the input bundle
impinges, and on the other side, S.sub.2, another refractive
aspheric surface in its central region and a discontinuous-slope
structure in its external region, said discontinuous-slope
structure cross-section comprising teeth with two aspheric faces
such that one of them, V, is configured to be parallel to the flow
lines of the bundle transmitted by said surface S.sub.1, and the
other face, T, is configured to reflect the bundle by total
internal reflection toward the face V where it is refracted so that
no ray intercepts the adjacent tooth and that the nearest edge ray
to do so is tangent to that tooth, and b) A second lens L.sub.2
that surrounds the receiver composed of a refractive aspheric
surface on which the bundle transmitted by the lens L.sub.1
impinges.
30. Device for collimating radiant energy wherein said device is
axisymmetrical or cylindrical and is configured to transform the
edge rays of an input extended ray bundle generated by an emitter
into edge rays of an output extended ray bundle, the optical spread
of at least one of said input and output bundles being greater than
ten degrees at one or more of the optical surfaces of said device,
both of said bundles being defined in the plane of a cross-section,
by: a) a lens L.sub.2 that surrounds the emitter and comprises a
refractive aspheric surface S.sub.3 on which the input bundle
impinges, and b) a second lens L.sub.1 comprising on one side a
refractive aspheric surface, S.sub.1, from which the output bundle
leaves, and on the other side, S.sub.2, another refractive aspheric
surface in its central region and a discontinuous-slope structure
in its external region, said discontinuous-slope structure
cross-section comprising teeth with two aspheric faces such that on
one of them, V, the bundle transmitted by said surface S.sub.3 is
refracted so that all the rays are reflected by total internal
reflection on the other face, T, that the edge ray nearest to not
being reflected is tangent to the profile of the tooth, and that
the face V is parallel to the flow lines of the bundle transmitted
toward S.sub.1.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application is a continuation of U.S. patent
application Ser. No. 10/148,736, which was filed on Oct. 15, 2002
and was a 35 U.S.C. 371 National Stage of PCT/ES00/00459 of Dec. 1,
2000.
TECHNICAL FIELD
[0002] This invention relates to the field of optical systems;
specifically, that of Nonimaging Optics.
BACKGROUND OF THE INVENTION
[0003] There exist previous inventions related to the present
invention, all related to one another, for which various patents
have been taken out (U.S. Pat. Nos. 4,337,759; 5,404,869;
5,577,493). While in a general way some of the possible geometries
of the present invention are qualitatively similar to those of
these previous inventions, there are several fundamental
differences that make this invention novel, and rule out any
conflict with the others. These differences lead to the optical
surfaces of the invention being substantially different, due to the
fact that the conditions imposed on their design are different, and
therefore also their resulting optical performance. In particular,
the invention presented here can work very close (>95%) to the
thermodynamic limit of concentration/collimation, while the
previous inventions, not based on the tools of Nonimaging Optics,
are well short of this limit (<80%) when the angular spread of
the ray bundles on passing through any of the optical surfaces is
large (>10 degrees).
[0004] The related patents are: the patent of Popovich et al. U.S.
Pat. No. 4,337,759, July/1982; that of W. A. Parkyn, Jr. et al.,
U.S. Pat. No. 5,404,869, April/1995, and lastly, that of W. A.
Parkyn, Jr. et al., U.S. Pat. No. 5,577,493, November/1996.
[0005] The designs of all the mentioned inventions are not based
(in contrast to this one) on the edge-ray theorem of Nonimaging
Optics, so that their functioning is limited with the extended
bundles produced by many emitters and receivers used in practice.
U.S. Pat. No. 4,337,759, July/1982 and U.S. Pat. No. 5,404,869,
April/1995 consider only the central ray of the bundles in the
design. U.S. Pat. No. 5,577,493, November/1996 considers the
so-called first-order optics around the central ray (Luneburg,
1964), which provides an order of approximation superior to the
previously-mentioned device, but even so, the performance
attributed to it by its inventors for producing constant irradiance
is only accurate for bundles with very small angular spread.
[0006] Furthermore, the invention protected by U.S. Pat. No.
5,577,493, November/1996 is axisymmetrical and considers as output
bundle that produces uniform irradiance in 3D at the exit aperture.
This bundle is only a particular case of those considered in the
present patent.
SUMMARY OF THE INVENTION
[0007] This invention includes a nonimaging concentration or
collimation device made of two aspheric lenses, one of them
containing a structure with discontinuous slope (i.e., faceted),
that concentrates the radiation incident on a receiver or
collimates the radiation from an emitter, depending on the case.
The design method of this concentrator is based on the nonimaging
design method of Simultaneous Multiple Surfaces or SMS (Minano,
Gonzalez, 1992).
[0008] For the design of this invention two extended (e.g., not
punctual) ray bundles are coupled in two-dimensional geometry (2D).
The actual three-dimensional (3D) devices are obtained by
rotational symmetry (axisymmetrical) or translational symmetry
(cylindrical), and their operation is analyzed a posteriori. Common
examples of ray bundles (FIG. 1) are: (type 1) that composed of
rays impinging on a segment (1) forming an angle inferior to a
given angle (2) (called the acceptance angle of the bundle) with
the perpendicular to this segment, and (type 2) that composed of
the rays that intercept two given segments (3). Both types of
bundle can be defined in a more general way (type 3) if the
segments are substituted by arbitrary curves. FIG. 1 shows, in
addition to two bundles of type 1 in FIG. 1(a) and type 2 in FIG.
1(b), a bundle of type 3, in FIG. 1(c) composed of the rays that
intercept a rectangle (4) and a semicircumference (5) (this bundle
is useful for modeling an LED or an IRED). Another bundle of rays
(type 4) can be described, with a more general character than those
of types 1 and 2 (which includes them as particular cases), as that
composed of the rays that impinge on a segment with an angle of
incidence between two specified angles for each point of the
segment.
[0009] The design of the present invention is based on the
so-called edge-ray theorem of Nonimaging Optics (Welford, Winston,
1989), which states that to couple two bundles associated with the
emitter and the receiver it is necessary and sufficient to couple
the subsets of edge rays of the two. The use of this theorem is the
key to obtaining devices that work very close to the thermodynamic
limit with bundles with wide angular spread. For example, the edge
rays of the bundles in FIG. 1 are, for the type 1 bundle, those
that impinge on the segment with an angle of incidence equal to the
acceptance angle of the bundle and those that pass through the
edges (6) and (7) of the segment; for the type 2 bundle, those that
pass through any of the edges (8), (9), (10) and (11) of the two
given segments; and for the type 3 bundle, those that are tangent
to the rectangle and those that pass through edges (12) and (13) of
the semicircumference.
BRIEF DESCRIPTION OF THE DRAWINGS
[0010] FIG. 1(a) illustrates a known type 1 bundle comprising rays
that impinge on a segment (1) of edges (6) and (7) forming an angle
inferior to the acceptance angle of the bundle (2) with the
perpendicular to that segment.
[0011] FIG. 1(b) illustrates a known type 2 bundle comprising rays
that intercept two given segments (3). The edge rays of this bundle
are those passing through any of the edges (8, 9, 10 and 11) of the
two given segments.
[0012] FIG. 1(c) illustrates a known type 3 bundle composed of the
rays that intercept a rectangle (4) and a semicircle (5) of edges
(12) and (13).
[0013] FIG. 2 illustrates a basic working principle of the
invention as concentrator of radiation on a receiver (14). It
comprises a lens (50) that surrounds the receiver comprising a
refractive aspheric surface (21), and a second lens (15) whose
upper side is a refractive aspheric surface (16) and whose lower
side consists of another refractive aspheric surface (17) in its
central region (between points 18 and 19) and a discontinuous-slope
structure (20) in its external region; whose faces (22)
fundamentally refract the rays and the faces (23) reflect them by
total internal reflection.
[0014] FIG. 3 illustrates a system of Cartesian coordinates (31)
and initial geometric parameters for carrying out the chosen design
for concentrating radiation on a receiver. The input bundle is
defined by the acceptance angle (24) and by the entry aperture
defined by the edges (25) and (26) of the surface S.sub.2. The
output bundle is defined by the segment of edges (27) and (51),
which is the receiver and it is illuminated from surface S.sub.3,
whose edges are (29) and (30), with an angle of illumination
limited to the acceptance angle (28).
[0015] FIGS. 4(a) and 4(b) illustrate teeth of the surface S.sub.2
designed in the first phase for a device in accordance with the
invention acting as a concentrator as in FIG. 4(a) or as a
collimator as in FIG. 4(b). As they are of infinitesimal size
(enlarged in the figure), the adjacent teeth are identical, and the
edge-ray bundles are parallel. It is desired that the light
incident through segment (33), of edges (34) and (35), with slope
between that of rays (36) and (37), is transmitted optimally
through segment (38), of edges (39) and (40), with slope between
that of rays (41) and (42), which form angles (54) and (55) with
the horizontal line, respectively. The geometry of the tooth with
respect to its macroscopic tangent vector (32) is defined by the
angles (56) and (57).
[0016] FIG. 5 illustrates a lens L.sub.2 produced with two
different dielectric materials separated by a spheric or aspheric
refractive surface (43).
[0017] FIG. 6 illustrates a lens wherein surface S.sub.1 can be
substituted by a discontinuous-slope Fresnel structure (44), which
reduces weight and absorption.
[0018] FIG. 7 illustrates a design of the surface S.sub.1 as a
discontinuous-slope structure (45) with saw-toothed profile, which
allows transmission losses to be minimized when the faces V do not
coincide with flow lines (46) of the bundle transmitted by the
continuous surface S.sub.1.
[0019] FIG. 8 illustrates a lens wherein the central portion of
S.sub.2 can be substituted by a discontinuous-slope Fresnel
structure (47).
[0020] FIG. 9 illustrates a configuration wherein surfaces S.sub.1
and S.sub.2 of the previous figures are interchanged, so that the
teeth appear inverted (48).
[0021] FIG. 10 illustrates a device having an optically inactive
portion (49) that joins the two lenses, and in such a way that they
constitute a single piece that includes an interior space (53).
[0022] FIG. 11 illustrates a device in accordance with the
invention with aspheric faces (54) in an advanced mode, which
allows them to be larger while maintaining excellent
performance.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0023] A possible configuration of the invented device is that
shown in FIG. 2, which also shows its basic working principle as a
concentrator of radiation on a receiver (14). The lens (15) L.sub.1
has two active faces: the upper refractive surface (16), referred
to as S.sub.1, which is in general aspheric, and the lower one,
S.sub.2, which consists of another refractive aspheric surface (17)
in its central portion (between points (18) and (19), which we
shall call P and P', respectively) and a structure with
discontinuous slope (20) in its external portion. The lens (50)
L.sub.2 surrounds the receiver and consists of the refractive
aspheric surface (21), which we shall call S.sub.3. The collected
rays that impinge on the central portion (17) undergo three
consecutive refractions before reaching the receiver. On the other
hand, the collected rays that impinge on the more external portion
(20) undergo the following incidences before reaching the receiver:
a first refraction on the surface S.sub.1, a (possible) total
internal reflection on the face (22) (which we shall call face V)
of the teeth of S.sub.2, a total internal reflection on the face
(23) of those teeth (which we shall call face T), a second
refraction on face V, and finally, a third refraction on S.sub.3.
The total internal reflection occurs when the angle of incidence of
the ray with the normal to the surface is greater than the
so-called critical angle of the interface, which is given by
sin.sup.-1(1/n), n being the refractive index of the lens
L.sub.1.
[0024] Particular cases are those in which the profile of S.sub.1
is circular or flat. The latter case is of especial interest in
certain applications, such as photovoltaic concentration, since it
permits the grouping of a set of devices fixed to a dielectric
plate, such as a flat piece of glass, which acts as a reference to
provide parallelism between the devices, as protection against the
elements and as a filter for ultraviolet radiation.
[0025] In the design the surfaces S.sub.2 and S.sub.3 are
calculated from the specification of the profile of the surface
S.sub.1 and of the input and output bundles. The definition of the
input bundle can be made before its refraction on S.sub.1, so that
its definition would be independent of that of that surface. For
example, it could be a type 1 bundle with acceptance .alpha. and
with the edges of the segment coincident with the extreme points of
the surface S.sub.1. Another possibility, which could be
interesting in practice, is that of defining the input bundle after
its refraction on S.sub.1, which allows, for example, the segment
crossed by the rays of the bundle to be that defined by the two
extreme points of the surface S.sub.2. This implies that the
specifications of the bundle and of the surface are interdependent:
if we wish to define the bundle as that composed of the rays that
impinge within the acceptance .alpha. before the refraction on
S.sub.1 and with the edges of the segment coincident with the two
extreme points of the surface S.sub.2, it will be necessary, in
general, to carry out a ray-tracing on the surface S.sub.1. In the
case that the surface S.sub.1 is flat, this ray-tracing is
unnecessary, since the refraction in this dioptric is trivial, and
the specification of the bundle after the refraction is therefore
immediate by application of Snell's Law: it will be a type 1 bundle
with acceptance angle equal to .alpha.'=sin.sup.-1(1/n sin
.alpha.), n being the refractive index of the lens L.sub.1.
[0026] In order to simplify the explanation, and by way of an
example, let us suppose that S.sub.1 is a plane, that the input and
output bundles are both type 1, and that the two bundles are
symmetrical with respect to an axis, as FIG. 3 shows. For the other
types of bundle the procedure is analogous. The input bundle
(specified after the refraction on S.sub.1) is defined by the
acceptance angle (24) with value .alpha.', and by the edges (25)
and (26) of the surface S.sub.2, which we shall call I and I', and
which determine the segment we shall refer to as the entry
aperture. The output bundle is defined by the receiver, which is
the segment of edges (27) and (51), called respectively R and R',
and by the angle of illumination limited to the acceptance angle
(28) of value .beta. (the normal consideration when the sensitivity
of the receiver is low for very grazing angles, as is common in
photodiodes or solar cells). The edges O and O' of the surface
S.sub.3 are the symmetrical points (29) and (30). This figure also
shows the system of Cartesian coordinates (31) that will be used
for the description, and whose origin is centered on the
receiver.
[0027] Input design parameters (apart from the profile of the
surface S.sub.1) are the angles .alpha. and .beta., the distance
RR', the refractive index of the dielectric materials to be used (n
for the lens L.sub.1 and n' for L.sub.2), the ordinate of point I,
the abscissa of point O and the abscissa of point P. The ordinate
of point O is calculated immediately from its abscissa, the
distance RR' and the angle .beta.. However, the calculation of the
abscissa of point I and of the ordinate of point P will be obtained
later, as the result of the design.
[0028] The design procedure consists of three phases. In the first
phase the design conditions for the teeth of the surface S.sub.2
(which will be different for concentration and collimation) are
chosen, supposing that they are of infinitesimal size. With these
conditions the calculation is made of the expressions that
constitute the individual design of teeth for the different angles
of incidence with respect to the mean normal vector of the tooth.
Designed simultaneously in the second phase, with the SMS method,
are the surfaces S.sub.2 and S.sub.3 that couple the output and
input bundles, taking into account the expressions calculated in
the first phase. Lastly, in the third phase, the teeth of the
surface S.sub.2 are generated with finite size (as manufactured in
practice) on the basis of the infinitesimal teeth calculated in the
previous phase.
[0029] There are different possible design modes, according to the
level of complexity of the finite-size teeth of the surface S.sub.2
both in their design in the third phase and in their manufacture.
Thus, we can define as basic mode that in which the profiles of the
T faces are rectilinear, as standard mode that in which these
profiles are arcs of circumference, and as advanced mode that in
which they are aspheric. The three modes converge on one another
when the size of the teeth is very small (providing an operating
quality coincident with that predicted for infinitesimal teeth),
but their performance degrades differently when the size of the
teeth is greater. In increasing order of quality are the basic,
standard, and advanced modes. Since the design of the standard and
advanced modes is carried out from the basic mode, we shall begin
by describing this before proceeding with the explanation of the
others.
[0030] Let us consider for the first phase the description of a
tooth designed in the first quadrant operating as a concentrator as
shown in FIG. 4(a). Given that the size of the tooth is
infinitesimal (enlarged in the figure), this means that, in the
scale of the figure, adjacent teeth are identical, and that the
wavefronts associated with the edge rays are flat. The vector (32),
which we shall call t, is the macroscopic tangent vector of the
surface S.sub.2. It is desired that the light incident through
segment (33), of edges (34) and (35), with slope between that of
rays (36) and (37), which we shall call, respectively, e(+) and
e(-), is transmitted optimally through segment (38), of edges (39)
and (40), with slope between that of rays (41) and (42), which we
shall call, respectively, i(-) and i(+). To this end the following
design characteristics will be imposed: (1) that no undesired
incidences occur, and (2) that the irradiance on leaving the tooth
is as uniform as possible. Both characteristics are obtained on
demanding the two following conditions. On the one hand, that face
V (also identified as face 22 of FIG. 2) is parallel to the
bisector of the impinging bundle, which coincides with the
so-called flow line of the bundle (Welford, Winston, 1989). Face V
situated in this way has the property of reflecting (through total
internal reflection) the bundle without its geometry being
modified. On the other hand, it should be demanded that the ray
e(-) that impinges at point (34), after the total internal
reflection on face T and the refraction on face V, is transformed
into the ray i(-) that passes through point (40). Note that rays
i(-) transformed from rays e(-) pass through all the points of
segment (38), but that rays i(+) emerge from only a portion of
segment (38) (for this reason the irradiance is not uniform in
(38), though it as uniform as possible, as required by condition
(2)). Nevertheless, in the second phase the rays i(+) and i(-) will
be used as though they emerged from the whole of segment (38),
which means that it will not be possible to reach the thermodynamic
limit of concentration/collimation (although the invention comes
very close to doing so).
[0031] These two conditions for the design of the infinitesimal
teeth, which guarantee their optimum functioning, constitute
another innovation with respect to the above-mentioned related
patents, none of which includes these conditions.
[0032] FIG. 4(b) shows a tooth for the basic design operating as a
collimator. As it can be seen, the difference with respect to the
case of FIG. 4(a), in which it was designed as a concentrator, lies
in the second imposed condition: in this case it is the ray e(+)
that impinges at (34) that must be transformed into the ray i(+)
that passes through point (40).
[0033] On imposing the two mentioned conditions it is deduced that
face V is vertical, and the following expressions relating the
angles involved are obtained by trigonometric calculations: (a) tan
.delta.=tan .psi.+(sin .psi.)(n.sup.2-sin.sup.2 .psi.).sup.-2+tan
.gamma. (Eq. 1.a) (b) n cos(2.delta.-.alpha.')+sin .phi.=0 (Eq.
1.b) (c) n cos(2.delta.+.alpha.')+sin .phi.'=0 (Eq. 1.c)
[0034] where .phi., .phi.', .delta. and .gamma. are, respectively,
the angles (54), (55), (56) and (57) shown in FIGS. 4(a) and 4(b),
n is the refractive index of the lens and .psi..ident..phi. in the
design of the concentrator and .psi..ident..phi.' in that of the
collimator.
[0035] In the second phase, in which the profiles of the surfaces
S.sub.1 and S.sub.2 are designed, the following steps are observed:
[0036] a) Select a value for the abscissa of point I (this value
will be recalculated later). [0037] b) Through the (inverse)
application of Snell's Law, calculate the vector tangent to S.sub.3
at point O with the condition that the ray that impinges from I
must be refracted at O toward R. [0038] c) Calculate the angle
.delta. of the infinitesimal tooth situated at point I with the
condition that the ray i(+) associated with the tooth is directed
toward O. This can be achieved using the equation (Eq. 1.c), where
the angle .phi.' is calculated from the points I and O. Calculate
also the angle .phi. using (Eq. 1.b), the angle .gamma. using (Eq.
1.a), and from this, calculate t.sub.I=(-cos .gamma., sin .gamma.),
which is the macroscopic vector tangent to S.sub.2 at I. [0039] d)
Find the first section of S.sub.3 above O with the condition that
the rays proceeding from I are refracted on that portion toward the
receiver with angle of incidence .beta.. The solution to this
problem is given by the constancy of optical path from point I up
to a flat wavefront sloped with the angle .beta., and is an
ellipse. This constitutes a particular case of the so-called
Cartesian ovals. The tangent to S.sub.3 at these points can be
found, once these have been calculated, by (inverse) application of
Snell's Law as in step a). The last point of this portion is marked
by the ray that, after refraction, passes through R'. [0040] e)
Find the following section of S.sub.3 with the condition that the
rays proceeding from I are refracted on that portion toward point
R'. Once again, the solution is given by the constancy of optical
path between the two points, and constitutes a particular case of
Cartesian ovals, and the tangent to S.sub.3 at these points is
found by (inverse) application of Snell's Law. The last point of
this section, which will be called H.sub.0 and its tangent
t.sub.H0, is that for whose calculation the ray i(-) that comes
from I has been used. [0041] f) Rename I, t.sub.I, O and to as
F.sub.0, t.sub.F0, G.sub.0 and t.sub.G0, respectively. From the
sections of S.sub.3 calculated in d) and f) select a number M of
uniformly-distributed points (for example, M=500) and name them
from F.sub.1 to F.sub.M, with tangents t.sub.F1 to t.sub.FM. Note
that H.sub.0.ident.F.sub.M (y t.sub.H0.ident.t.sub.FM). [0042] g)
Find the following macroscopic point G.sub.1 of the surface S.sub.2
as the point of intersection between the straight line that passes
through G.sub.0 with direction vector t.sub.G0 and the trajectory
of the ray refracted at F.sub.1 proceeding from R (traced in the
reverse direction). This ray is the ray i(+) associated with the
infinitesimal tooth at G.sub.1, so that it also gives the angle
.phi.' at that point. With equations (Eq. 1.c), (Eq. 1.b) and (Eq.
1.a) we can calculate, respectively, the angles .delta., .phi. and
.gamma., and from the last of these, t.sub.G1=(-cos .gamma., sin
.gamma.), which is the macroscopic vector tangent to S.sub.2 at
G.sub.1. [0043] h) Calculate the following point H.sub.1 of the
surface S.sub.3 as the point of intersection between the straight
line that passes through H.sub.0 with direction vector t.sub.H0 and
the ray i(-) associated with the infinitesimal tooth of G.sub.1.
The tangent t.sub.H1 to S.sub.3 at H.sub.1 can once more be found
by (inverse) application of Snell's Law. Identify
H.sub.1.ident.F.sub.M+1 (and t.sub.H1=t.sub.FM+1). [0044] i) Repeat
steps g) and h), increasing the subindices by one unit, until the
abscissa of a point G.sub.n is greater than the abscissa of point P
(selected as entry parameter). Since the precision on the abscissa
of point P chosen is not important (and that, this precision being
determined by the value of the chosen parameter M in step f), it
can be improved through choice), it will be considered for what
follows that P.ident.G.sub.n.
[0045] The profile of the central region of S.sub.2 (between P and
P') will be calculated (together with the remaining portion of
S.sub.3), once again according to the edge-ray theorem, so that it
directs the rays e(+) toward R' and the rays e(-) toward R (note
that this assignation is the opposite of what was carried out in
steps g) and h) for the exterior portion of S.sub.2). Given that
the surfaces are continuous, this implies that the optical path
from the wavefront associated with rays e(+) up to R' will be
constant, as will that from the wavefront associated with rays e(-)
up to R. So that the surfaces S.sub.2 and S.sub.3 do not have
discontinuities in their respective vertices, the symmetry of the
design obliges the two optical paths (measured with respect to
symmetrical wavefronts), moreover, to be equal. This condition will
allow evaluation of the initial choice of the abscissa of point I.
[0046] j) Find the tangent to S.sub.2 at P so that the impinging
ray e(-) is transformed after refraction into the ray i(+)
calculated at point P in step i). Calculate the ray e(+) after the
refraction at P. If the angle it forms with the horizontal is
superior to the angle .phi. calculated at point P in step i),
return to the beginning choosing a lower value for the abscissa of
point P. [0047] k) Calculate a new section of S.sub.3 next to point
H.sub.n found in step i) with the condition that the rays coming
from P are refracted on that portion toward point R'. Once again,
the solution is given by the constancy of optical path between the
two points, and the tangent to S.sub.3 at these points is found by
(inverse) application of Snell's Law. The last point of this
section is that for whose calculation the ray e(+) after refraction
at P has been used. Choose a number M' of uniformly-distributed
points (for example, M'=50) and name them in a way correlative to
the previous ones, that is, from H.sub.n+1 to H.sub.n+M' (and from
F.sub.M+n+1 to F.sub.M+n+M'). [0048] l) Calculate the optical paths
C(+) and C(-) associated with the rays e(+) up to R' and the rays
e(-) up to R, respectively. [0049] m) Repeat the steps from a) to
l) iterating on the value of the abscissa of point I until it is
achieved that |1-C(+)/C(-)|<.epsilon., with .epsilon. being a
pre-fixed margin of error (e.g., 0.0001). [0050] n) Calculate the
following point G.sub.n+1 of S.sub.2 with the condition that the
trajectory of the ray refracted at F.sub.n+1 coming from R (traced
in the reverse direction) is transformed after refraction at the
desired point into a ray e(-). Once again, the solution is
calculated because the optical path C(-) is known, and the tangent
to S.sub.2 at G.sub.n+1 is found by (inverse) application of
Snell's Law. [0051] o) Calculate the following point H.sub.n+M'+1
of S.sub.3 with the condition that the trajectory of the ray e(+)
refracted at G.sub.n+1 is directed, after refraction at the desired
point toward R'. Again, the solution is calculated because the
optical path C(+) is known, and the tangent to S.sub.3 at
H.sub.n+M'+1 is found by (inverse) application of Snell's Law.
[0052] p) Repeat steps n) and o) until the symmetry axis is
reached, that is, until the abscissas of points G and H calculated
are negative.
[0053] Finally, to conclude the basic design there remains only the
third phase, which involves the generation of the teeth of S.sub.2
with finite size (as they will be manufactured in practice) and
faces with rectilinear profile on the basis of the macroscopic
surface and the infinitesimal teeth calculated in the previous
phase. The procedure moves from the edge toward the center of the
lens observing the following steps: [0054] a) Select, for example,
size D of the horizontal projection of the finite teeth. This size
should be such that that the subsequent ray-tracing shows no
important degradation in the functioning of the device with respect
to that obtained with size D/2. [0055] b) Take as central points of
the finite teeth those points G.sub.i of the macroscopic surface
between P and I whose abscissa differ less from point I by an odd
number of times D/2. [0056] c) Define the slope of the face T of
the finite tooth to which G.sub.i belongs as the slope of the face
T defined at G.sub.i by the infinitesimal tooth. The face T of the
finite tooth is extended symmetrically with respect to the point.
[0057] d) The faces V are thus situated at abscissas that differ
from point I by a whole number of times D.
[0058] The basic concentrator design is complete. In this last
phase another criterion can be taken for the generation of finite
teeth, such as that the distance between the upper and lower
evolvent of the teeth takes the value D. The generation procedure
is similar to that described, and the adjustment of the central
points G.sub.i of each tooth can carried out in an iterative
way.
[0059] The standard mode differs from the basic mode in the third
phase, where the faces T of the finite teeth have an arc of
circumference as a profile. The design procedure of this mode is
similar to that of the basic one. In the second phase, although the
resulting design is identical, the standard mode adds the
calculation of the curvature of the faces T of the infinitesimal
teeth (for its later use in the third phase), which constitutes a
higher order of precision than that employed in the basic mode. In
order to make this calculation the following equation is used,
which relates the radii of curvature of a surface and those of the
incident and refracted/reflected wavefronts: (n.sub.i cos.sup.2
.theta..sub.i)/.rho..sub.i+(n.sub.r cos.sup.2
.theta..sub.r)/.rho..sub.r=(n.sub.i cos .theta..sub.i-n.sub.r cos
.theta..sub.r)/.rho..sub.s (Ec. 2)
[0060] where the sub indices i, r and s refer to the incident
wavefronts, refracted/reflected wavefronts and the surface,
respectively, n denotes the refractive index, .theta. the angle of
the ray with respect to the normal and .rho. the radius of
curvature. Equation (Eq. 2) is applied to the reflection, making
.theta..sub.r=.theta..sub.i and n.sub.r=n.sub.i.
[0061] In order to calculate the radius of curvature .rho..sub.sT
of the face T of the infinitesimal teeth it is necessary to find
first the radius of curvature of S.sub.3 at points F.sub.1 to
F.sub.M during their calculation in steps d) and e) of the second
phase. For this the expression (Eq. 2) is applied to the refraction
at these points of the rays coming from I. In this case, for each
point F.sub.k and denoting by AB the length of the segment of edges
A and B, we have .rho..sub.i= (IF.sub.k), .rho..sub.r=.infin. in
step d) and .rho..sub.r= R'F.sub.k in step e).
[0062] It is in step f), in which the points G.sub.k are calculated
on the basis of the points F.sub.k, where the desired values of
.rho..sub.sT should be calculated. The calculation involves the use
of the expression (Eq. 2) for the three successive incidences
undergone by the ray that goes (in the reverse direction) from R
toward F.sub.k. Given that in step f) the points and the normals to
the surfaces are calculated, the angles of incidence and of
refraction/reflection, like the refractive indices, are known
parameters in the three incidences. In the first, at F.sub.k, as
the radius of curvature .rho..sub.s is already known and
.rho..sub.i= RF.sub.K, from (Eq. 2) we obtain the radius of
curvature of the refracted wavefront .rho..sub.r1. For the second
incidence, which occurs on the face V of the tooth calculated at
G.sub.k, the radius of curvature of the incident wavefront is
.rho..sub.i= G.sub.kF.sub.k-.rho..sub.i and the radius of curvature
of the surface is known (.rho..sub.sv=.infin.), so that from (Eq.
2) we obtain the radius of curvature of the refracted wavefront
.rho..sub.r2. Finally, for the third incidence, which occurs on the
face T of the tooth, it is known that .rho..sub.i=.rho..sub.r2 and
.rho..sub.r3=.infin., so that (Eq. 2) can be solved with the radius
of curvature .rho..sub.sT as an unknown, which was the desired
value.
[0063] Given that in step g) new points F.sub.j, are calculated,
initially called H.sub.k, and which will be used again in step f)
on repeating it as h) indicates, it is also necessary to calculate
the radius of curvature of S.sub.3 at these points. For this, the
procedure is analogous to that of the calculation of .rho..sub.sT
previously indicated, using the trajectory of the ray used to
calculate H.sub.k, which is the ray e(+) impinging at G.sub.k, and
taking advantage of the fact that .rho..sub.sT is already
known.
[0064] The third phase of the standard mode, which concerns the
generation of the teeth of finite size, differs from the basic mode
in that the T faces, instead of being rectilinear, are generated as
arcs of circumference. The procedure of generating the teeth is
analogous to that seen for the basic mode, the only difference
being that the face T of the finite tooth to which the central
point G.sub.i of a finite tooth belongs is the arc of circumference
that passes through that point, with the slope and radius of
curvature associated with the infinitesimal tooth, and that extends
symmetrically with respect to the point. This concludes the
standard design mode.
[0065] Lastly, the advanced design mode is characterized by the
faces T of the teeth having an aspheric profile. The calculation of
these profiles is made from the finished basic design (with finite
teeth), observing the following steps: [0066] a) Trace in the
reverse direction the uniparametric ray bundles that leave from R
and R' and are refracted on S.sub.3 and on the faces V of the
finite teeth. [0067] b) For each tooth, whose central point is
G.sub.i, calculate the aspheric profile of the face T that passes
through G.sub.i and whose points Q are such that the ray that
impinges vertically is reflected in accordance with the direction
bisecting of the rays of the uniparametric bundles that pass
through Q calculated in a). This problem, which can be expressed in
the form of a first-order differential equation, has a single
solution when one ray--and only one of each bundle--passes through
each point Q.
[0068] The advanced design mode is finished. FIG. 11 shows an
example of an advanced design. As mentioned above, the provision of
aspheric profiles (54) for the facets allows them to be made larger
than in the basic and standard modes while maintaining excellent
performance, even close to the thermodynamic limit.
[0069] The description of the design procedures for the three modes
(basic, standard and advanced) is finished.
[0070] The design is essentially similar in the case that the
profiles of the faces V are not vertical lines, but have a
pre-fixed sloping rectilinear, circular or aspheric profile. In
fact, an aspect not considered in the descriptions of the designs
concerns the fact that the manufacture of teeth with totally
vertical faces V may be impractical (in the case of lens
manufacture by plastic injection, removal of the part from the
mould is difficult). It is possible to correct this aspect, for
example, by considering design of the faces V with inclinations of
a certain angle (in the range of 0.5 degree to 1 degree may be
sufficient), which entails appropriate modification of the
equations (Eq. 1). This inclination is also useful to avoid the
undesired effects produced in practice by the rounding of the
vertices of the teeth. As a negative consequence, a sloping face V
means that it is not parallel to a flow line of the incident
bundle, so that the reflection on that face will modify (slightly)
the geometry of the bundle. This means that the characteristic of
angular transmission will be degraded (i.e., it will be less
stepped) with respect to that corresponding to vertical faces V.
Meanwhile, the profiles of the faces V can be produced as arcs of
circumference or pre-fixed aspheric curves to facilitate their
manufacture even more (at the cost of making the production of the
mould more difficult), by decreasing, for example, the curvature
necessary for the profiles of the faces T.
[0071] Another aspect not dealt with up to now concerns the fact
that the condition imposed on the design of the infinitesimal teeth
in concentration that obliges the edge ray e(-) impinging at point
(34) to be transformed into the ray i(-) that passes through point
(40) may be relaxed (i.e., allowing it to pass slightly above or
below that point) without producing a serious degradation in
functioning.
[0072] Taking into account all of these considerations, we can
affirm the usefulness of the possibility of the profiles of the
faces V or T having at each point a slope modified by an angle of
less than 2 degrees.
[0073] The device described for concentrating radiation on a
receiver may be axisymmetrical or cylindrical, and is characterized
by transforming the edge rays of an input extended ray bundle into
edge rays of an output extended ray bundle that illuminates a
receiver, both bundles being defined in the plane of a
cross-section (which contains the symmetry axis in the
axisymmetrical case, or is perpendicular to the direction of
symmetry in the cylindrical case), by means of: (a) a lens L.sub.1
composed on one side of a refractive aspheric surface, S.sub.1, on
which the input bundle impinges, and on the other side, S.sub.2, of
another refractive aspheric surface in its central region and with
a discontinuous-slope structure in its external region, whose
cross-section is formed of teeth with two aspheric faces such that
one of them, V, is parallel to the flow lines of the bundle
transmitted by the dioptric S.sub.1, and the other face, T,
reflects the bundle by total internal reflection toward the face V
where it is refracted so that no ray intercepts the adjacent tooth
and that the nearest edge ray to do so is tangent to that tooth;
and (b) a second lens L.sub.2 that surrounds the receiver composed
of an refractive aspheric surface on which the bundle transmitted
by the lens L.sub.1 impinges.
[0074] On the other hand, the device used for collimating the
radiation generated by an emitter may be axisymmetrical or
cylindrical, and is characterized by transforming the edge rays of
an input extended ray bundle generated by an emitter into edge rays
of an output extended ray bundle, both bundles being defined in the
plane of a cross section, by means of: (a) a lens L.sub.2 that
surrounds the emitter composed of a refractive aspheric surface on
which the input bundle impinges; and (b) a second lens L.sub.1
composed on one side of a refractive aspheric surface, S.sub.1,
from which the output bundle leaves, and on the other side,
S.sub.2, of another refractive aspheric surface in its central
region and with a discontinuous-slope structure in its external
region, whose cross-section is formed of teeth with two aspheric
faces such that on one of them, V, the bundle transmitted by the
dioptric S.sub.3 is refracted so that all the rays are reflected by
total internal reflection on the other face, T, that the edge ray
nearest to not being reflected is tangent to the profile of the
tooth, and that the face V is parallel to the flow lines of the
bundle transmitted toward S.sub.1.
[0075] U.S. Pat. No. 5,577,493, November/1996 describes an
axisymmetric device which is qualitatively similar to this
invention, and it is used to collimate the radiation generated by
an emitter and in which the bundles of rays are chosen to provide
uniform irradiance in three dimensions at the exit aperture.
However, due to the restrictions of that design method, the device
disclosed will provide such performance only when the angular
spread of the ray bundles on passing through any of the optical
surfaces is small (<10 degrees). Moreover, the design conditions
of the infinitesimal teeth used in this invention (Equations 1.a,
1.b, and 1.c), which guarantee the optimal performance of the
teeth, are not used in the '493 patent, which leads to the optical
surfaces of their invention being substantially different and their
resulting optical performance being noticeably inferior.
[0076] The procedures described for the three design modes are
equally applicable to the situation shown in FIG. 5, in which the
lens L.sub.2 consists of two different dielectric materials
separated by a spherical or aspheric refractive surface (43), with
pre-fixed profile, just considering the refraction of the edge rays
on this surface during the process.
[0077] A variant of the configuration described up to now consists
in substituting the refractive surface S.sub.1 by a
discontinuous-slope Fresnel structure (44), as shown for example in
FIG. 6 for the case of the flat and horizontal S.sub.1. It is thus
possible to use less dielectric material, which reduces its weight
and absorption. The two surfaces, discontinuous and continuous,
work in the same way. In fact, the profiles of the remaining
optical surfaces are identical in the two designs. The only
difference with respect to the trajectories of the rays is that
these can now impinge on the vertical face of the steps, the face
that coincides with the flow lines of the incident bundle. This
implies, once again, that if these faces were mirrors the
reflection of the rays on them would not modify the geometry of the
transmitted bundle. Although when the incidence takes place from
the interior face of the dielectric material this interface indeed
behaves like a mirror due to the phenomenon of total internal
reflection, this is not the case with incidence from the air, which
leads to some losses. Nevertheless, for small acceptance angles
.alpha.(<5.degree.) these losses are negligible due to the
combination of two effects: the reflectivity of this interface,
although not 100%, is very high for large angles of incidence (and
in the present case they will be superior to 90.degree.-.alpha.),
and the fraction of rays that impinge on the vertical faces from
the air is also small if the acceptance angle is moderate.
[0078] The surface S.sub.1 as a discontinuous-slope structure can
also have another use, as FIG. 7 shows. In this case the flat
dioptric of FIG. 2 has been substituted by a discontinuous-slope
structure with saw-toothed profile (45) that diverts the input
bundle to modify the direction of the flow lines (46). The lens is
attached to a dielectric plate 45' by means of an adhesive 45''
with a refractive index slightly different from that of the lens.
This structure refracts the rays of the incident bundle so that
they progress toward S.sub.2 with a slight divergent direction.
This means that the face V of the teeth of the external region of
S.sub.2, if designed with a non-null tilt angle to facilitate
manufacture, will produce a lesser degradation of the optical
performance on being closer to (or even coinciding with) the
divergent flow line. As the vertical face of the S.sub.1 in turn
causes a degradation (by blocking the ray trajectories), for each
inclination of the faces V there is an optimum angle of divergence
of the bundle, for which degradation is minimum.
[0079] Another possibility (which can also be combined with any of
the previous ones) consists in making the central portion of
S.sub.2 as a discontinuous-slope Fresnel structure (47), as shown
in FIG. 8.
[0080] Another possible configuration consists in designing the
lens with the surfaces S.sub.1 and S.sub.2 interchanged, so that
the teeth appear inverted (48), as shown in FIG. 9. In the case of
rotation symmetry, for its production by molding, it is necessary
for either the mould or the lens to be flexible, so that the lens
can be extracted from the mould. In the case of translation
symmetry this would be unnecessary, since manufacture could be by
extrusion. The design procedure of the optical surfaces is common
to all the configurations indicated.
[0081] In the proposed invention, used for concentrating radiation
on a receiver, this could be optoelectronic, such as a photodiode,
phototransistor or solar cell. On the other hand, if it is used for
collimating the radiation produced by an emitter, this could also
be optoelectronic, such as an LED, an IRED or a laser.
[0082] The manufacturing of the concentrator of this invention can
be carried out using a diamond-tip lathe with numerical control
(CNC) on plastic material, such as acrylic (PMMA). Another
possibility worthy of mention is that of the injection of the PMMA
in a suitable mould, which allows a production also covered by this
application, and which is shown in FIG. 10: the device can be
manufactured with an optically inactive portion (49) that joins the
two lenses, and in such a way that they constitute a single piece
that includes an interior space (53). The joining can be carried
out by contact before solidification of the final injected part or
by means of subsequent gluing. As it is a single piece, the space
between the lenses is protected from dust and humidity. This space
can be evacuated or filled, if desired, with an inert gas. The
adhesion of the receiver or the emitter to the secondary can be
carried out by means of the casting of a transparent epoxy
resin.
[0083] The improvements and differences this invention introduces
with respect to the mentioned state of the art can be summarized as
follows: [0084] (a) The designed surfaces and the faces of the
teeth are such that the device couples in two dimensions the edge
rays of two extended ray bundles, while those of the mentioned
inventions couple only the central ray of the bundles or its
first-order environment. [0085] (b) In the case of the axisymmetric
device to collimate the radiation generated by an emitter, the
design ray bundles include as a particular case the one that
produces uniform irradiance in the exit aperture, the case
considered in U.S. Pat. No. 5,577,493, November, 1996, but in said
patent the design that is described there is only adequate when the
angular spread of the ray bundles on passing through all of the
optical surfaces is small (<10.degree.). [0086] (c) The
conditions for the design of the infinitesimal teeth facets used in
this invention (given by the Equations 1.a, 1.b, & 1.c, which
provide that the faces of the teeth are such that one guides the
bundle as a flow line, they produce maximum uniformity of
irradiance at the exit of the tooth, and they avoid undesired
incidence on the adjacent tooth) are not used in the previous state
of the art, which lead to the optical surfaces of the invention
being substantially different, and also their resulting optical
performance. [0087] (d) Its use as a concentrator on a receiver.
[0088] (e) The cylindrical symmetry, where appropriate. [0089] (f)
Its possible manufacture as a single part with an interior space.
[0090] (g) The grouping of a set of devices fixed to a dielectric
plate.
[0091] Differences (a) and (c) confer upon this invention an
optical performance noticeably superior to that of the previous
inventions, especially when the angular spread of the ray bundles
as they pass through any of the optical surfaces is large (>10
degrees).
[0092] The presented invention has direct applications in diverse
fields, such as that of radiation sensors, illumination systems
with LEDs, wireless optical communications or photovoltaic solar
energy.
[0093] In the field of sensors, the proposed invention allows the
achievement of high sensitivities, close to the thermodynamic
limit, without affecting the simplicity and compactness of the
device. Also, in the field of illumination with LED, this invention
provides an optimally-collimated bundle with geometry readily
compatible with current production techniques.
[0094] In wireless optical communications, the control of angular
response of the emitting and receiving devices and the use of
almost all possible directions of emission/reception in the design
permits connections whose signal-noise relationship is close to the
maximum possible. Employed in reception, the proposed invention
would use an optoelectronic sensor as receiver (e.g., a photodiode
or phototransistor); in transmission the invention would use an
optoelectronic emitter (LED, IRED or laser).
[0095] Finally, in photovoltaic applications this invention
constitutes an appropriate device for high-concentration solar
cells. Its performance close to the theoretical limit means that
for a given concentration factor, the angular acceptance of the
device is close to the maximum possible, which is useful for
permitting high tolerances in the manufacture of the device itself
and in the alignment of several of them to form a module (which can
be made simply by gluing them to a flat piece of glass), a light
support structure for modules with low sun-tracking accuracy.
* * * * *