Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
Case 2351 CA 02328213 2000-i2-i4
OPTICAL COUPLER FOR COUPLING LIGHT BETWEEN ONE AND A
PLURALITY OF LIGHT PORTS
Field of the Invention
The present invention relates to optical couplers for coupling light between
one
and a plurality of light ports, and more particularly to such couplers in
which the etendue,
or brightness, of the light is preserved to a high extent.
Background of the Invention
1o A common problem in designing fiber-optic lighting systems is to minimize
the
size (i.e., diameter) and number of light guides required to deliver
sufficient light. The
smaller the fibers required, the lower the cost and the easier to install.
Smaller fibers are
more flexible and more easily concealed. If multiple outputs are required, the
least
expensive and neatest method is to run one large fiber from a light source to
an
intermediate position and then split the light into a number of smaller
fibers. If the large
fiber is transporting light at angles up to its acceptance angle, then the
coupler should
make the split without substantially increasing the angular distribution of
light, to reduce
light loss. This is possible in principle, if the total areas at input and
output are the same,
which is often not practical due to design issues, owing to the law of
conservation of
etendue (or brightness or sparkle of light). The law only states that it is
possible, but does
not require it.
A primary goal in designing a coupler is to avoid light loss in the coupler.
In an
ideal case, the aggregate area of the outputs can be the same as the area of
the input where
the numerical apertures of the input and outputs are the same. Since this is
not always
practical due to fabrication issues, it is often necessary to allow the
aggregate area of the
outputs to increase to approximately a factor of two over the input area.
A further goal is to avoid light loss in coupling light to, for instance,
output light
guides. For this it is necessary that the angular distribution of light at the
output not
CA 02328213 2006-O1-23
exceed the acceptance angle of the output light guides. Where the numerical
aperture of
the input and output fibers are the same, this is accomplished by not
substantially
increasing the angular distribution of light during splitting.
A yet further goal of some embodiments is to split the light between the
output
ports in a controlled manner. The most typical example is splitting the light
evenly
between the different ports. It is most often desired, if not a requirement of
a fiber
system, that the fraction of light and the color of the light be the same for
all the ports so
that post installation testing can be minimized.
Another goal is to provide output ports which are spatially separate from each
other. Ln the case of typical large core plastic optical fibers, this
eliminates the need to
strip the cladding and jacket off of the fiber. In the case where the coupler
is used as a
combiner, this allows space for mechanical packaging of the input sources.
A yet further goal is to minimize the area of the output, since smaller-sized
fibers
are more economical, etc. It is also desirable to preserve the etendue of the
coupled light
to a high degree. Preserving the etendue is especially important since it
enables the size
of the optics that are attached to the fibers, such as automotive headlamps,
to be
minimized.
Summary of the Invention
2o A preferred embodiment of the invention provides an optical coupler for
coupling
light along an axis between a light port on one side of the coupler and a
plurality of light
ports on another side of the coupler. The coupler comprises a one-side stage
with a light
port, the one-side stage being free of segmentation along the axis; and a many-
side stage
with a plurality of arms situated about the axis, each arm having a light
port. A midput
region separates the one-side and many-side stages and is situated along the
axis where
the plurality of arms at least initially starts to split from each other in a
direction towards
the many-side light ports along the axis. Cross sections of each of the
respective initial
portions of the arms along the direction are arranged about the same distance
from the
axis. The cross sectional areas of the arms along the axis are larger (and
preferably
3o substantially larger) at a break point region at which the arms fully
separate from each
2
CA 02328213 2006-O1-23
other along the axis than at the many-side light ports. At least a pair of the
arms each
start to split along the direction in a substantially symmetrical manner about
the axis.
In a further aspect, there is provided an optical coupler for coupling light
along
an axis between a light port on one side of the coupler and a plurality of
light ports on
another side of the coupler, comprising: a) a one-side stage with a light
port; the one-side
stage being free of segmentation along the axis; b) a many-side stage with
more than two
arms situated about the axis, each arm having a light port with a non-
polygonal cross
section along the axis; c) a midput region separating the one-side and many-
side stages
and being situated along the axis where the plurality of arms at least
initially starts to split
from each other in a direction towards the many-side light ports along the
axis; a break
point region being defined along the axis where the arms fully separate from
each other;
d) cross sections of each of the respective initial portions of the arms along
the direction
being arranged about the same distance from the axis; e) at least a pair of
the arms each
starting to split along the direction in a substantially symmetrical manner
about the axis;
t 5 and f) sequential cross sectional areas of the arms substantially changing
shape between
the break point region and the many-side light ports.
In another aspect, there is provided an optical coupler for coupling light
along an
axis between a light port on one side of the coupler and a plurality of light
ports on
another side of the coupler, comprising: a) a one-side stage with a light
port; the one-side
2o stage being free of segmentation along the axis; b) a many-side stage with
more than two
arms situated about the axis, each arm having a light port with a non-
polygonal cross
section along the axis; c) a midput region separating the one-side and many-
side stages
and being situated along the axis where the plurality of arms at least
initially starts to split
from each other in a direction towards the many-side light ports along the
axis; a break
25 point region being defined along the axis where the arms fully separate
from each other;
d) cross sections of each of the respective initial portions of the arms along
the direction
being arranged about the same distance from the axis; e) at least a pair of
the arms each
starting to split along the direction in a substantially symmetrical manner
about the axis;
and fJ cross sectional areas of the output arms along the axis being
substantially larger
3o at the break point region than at the many-side ports.
2a
CA 02328213 2006-O1-23
In yet another aspect, there is provided an optical coupler for coupling light
along
an axis between a light port on one side of the coupler and a plurality of
light ports on
another side of the coupler, comprising: a) a one-side stage with a light port
positioned
at a first point along the axis; the one-side stage being free of segmentation
along the
axis; b) a many-side stage with a plurality of arms situated about the axis,
each arm
having a light port, with at least two of the arms which are generally
parallel to the axis
having their associated light ports positioned at a second point along the
axis; c) a
midput region separating the one-side and many-side stages and being situated
along the
axis where the plurality of arms at least initially starts to split from each
other in a
to direction towards the many-side light ports along the axis; cross sections
of each of the
respective, initial portions of the arms along the direction being arranged
about the same
distance from the axis; d) a break point region being defined along the axis
where the
arms fully separate from each; e) the coupler having a substantially final
configuration
including an angle-to-area shape along the axis and defined by the conceptual
steps of:
i) designing a conceptual, single angle-to-area converter for placement
between the first
and second points along the axis; the axial end of the converter at the first
point defining
the one-side light port; ii) designing a plurality of conceptual angle-to-area
sub-converters
for positioning between the first and second points along the axis; the axial
ends of each
of the sub-converters at the second point defining the many-side light ports;
adjacent
2o sub-converters being positioned at the first point in a manner for defining
a first area at
least covering the one-side light port of the single converter; iii) unioning
the overlapped
sub-converters together; iv) the single converter being designed with an end
at the second
point that surrounds all of the many-side light ports; and v) interpositioning
the converter
and the unioned sub-converters along the axis, each between the first and
second points,
and intersecting them to form a substantially final configuration of the
coupler.
The coupler according to the foregoing embodiment beneficially has low light
loss, while minimizing the area of the output. It beneficially can be made
compact in
length and diameter, and may have at least a pair of output arms substantially
parallel to
each other to facilitate coupling to output light guides. Alternatively, the
coupler can be
3o designed to combine light from a plurality of light sources.
2b
CA 02328213 2006-O1-23
Description of the Drawings
Fig. 1 is a perspective view of a coupler according to one embodiment of the
W vention.
Fig. 2 is a perspective view of a coupler that is alternative to that of Fig.
1.
Fig. 3 is a perspective view of another coupler according to the invention.
Fig. 4 is a perspective view of a further coupler according to the invention.
Figs. SA-SH show sequential, equally spaced cross sections of coupler 10 of
Fig.
1.
Figs. 6A-6G show sequential, equally spaced cross sections of coupler 30 of
Fig.
3.
Figs. 7A-7H show sequential, equally spaced cross sections of coupler 50 of
Fig.
4.
Fig. 8 is a detail view showing portions of output arms of a coupler and
corresponding light guides connected to the output arms.
Fig. 9A is a side view of an angle-to-area converter used to produce a
preferred
embodiment of the inventive coupler in accordance with a design example.
Fig. 9B is an end view of the converter of Fig. 9A taken along line 9B--9B in
Fig.
9A.
3
Case2351 CA 02328213 2000-i2-i4
Fig. 9C is an end view of the converter of Fig. 9A taken along line 9C-9C in
Fig.
9A.
Figs. l0A is a side view of a pair of angle-to-area sub-converters used to
produce
a preferred embodiment of the inventive coupler in accordance with the
mentioned design
example.
Fig. lOB is an end view of the sub-converters of Fig. l0A taken along line lOB-
l OB in Fig. 10A.
Fig. lOC is an end view of the sub-converters of Fig. l0A taken along line lOC-
l OC in Fig. 10A.
1o Fig. lOD is a side view of the sub-converters of Fig. l0A after undergoing
a union
operation.
Fig. 11A is a side view of the converter of Fig. 9A and sub-converters of Fig.
l OD
interpositioned with respect 'to each other, in accordance with the mentioned
design
example.
Fig. 11B is an end view of the converters of Fig. 1 lA taken along line 11B-
11B
in Fig. 11 A.
Fig. 11C is an end view of the converters of Fig. 11A taken along line 11C-11C
in Fig. 11 A.
Fig. 12A is a side view of a coupler produced according to the mentioned
design
2o example.
Fig. 12B is an end view of the coupler of Fig. 12A taken along line 12B-12B in
Fig. 12A.
Fig. 12C is an end view of the coupler of Fig. 12A taken along line 12C-12C in
Fig. 12A.
Detailed Description of the Invention
Fig. 1 shows an optical coupler 10 in accordance with one embodiment of the
invention. Coupler 10 includes an input stage 12 having an input port 14, and
an output
4
Case2351 CA 02328213 2000-i2-i4
stage 16 having four output arms 18a, 18b, 18c and 18d, each with a respective
output
port 20a, 20b, 20c or 20d. (In an alternative embodiment, the input and output
ports are
interchanged, so that light entering ports 20a-20d becomes combined and
directed out of
the coupler through port 14.)
A midput region 22 is defined along a central axis 24 of the coupler, where
the
output arms at least first start to separate from each other along the input-
to-output
direction of light propagation. Midput region 22 coincides with a break point
region 25,
at which output arms 18a-18d fully separate from each other. The arms separate
from
each other in an "X" shape 26, with the "X" being located in a plane. Midput
region 22 is
to shown with zero length along axis 24, although it could be of non-zero
length, e.g., a
cylinder.
In a preferred construction of coupler 10, the diameters of input port 14 and
output ports 20a-20b are 7 mm and 5 mm, respectively; the axial distance from
input port
14 to midput region 22 is 50 mm, and the axial distance from the midput region
to the
output ports is 40 mm.
Coupler 10 is preferably formed as a unitary piece of acrylic, for example,
relying
on total internal reflection as the primary means of light propagation.
Alternatively,
coupler 10 (and the other couplers described herein) could comprise a hollow
coupler
with reflective walls, such as coupler 28 in Fig. 2. Another option is to form
the device
2o using a material with a non-uniform index of refraction, often called a
Gradient Index
material.
Although optical coupler 10 is shown with four output arms, it could
alternatively
have other numbers of output arms, such as two, three, five or six. Moreover,
the cross
sections of either the input port or the output ports could be polygonal
(e.g., rectangular),
rather than circular as shown. The shape of coupler 10 will be more fully
described
below.
Fig. 3 shows a preferred coupler 30 in accordance with the invention. The
configuration of coupler 30 is selected to more closely preserve the etendue
of light
5
Case2351 CA 02328213 2000-i2-i4
transmission than coupler 10 (Fig. 1). Like reference numerals as between
different
embodiments refer to like parts. As with coupler 10 (Fig. 1), coupler 30
includes an input
port 14 and output arms 18a, 18b, 18c and 18d with respective output ports 20a-
20d. Its
input stage 32 extends along axis 24 from input port 14 to midput region 34 at
which the
output arms first start to separate from each other along the input-to-output
direction of
light propagation. Input stage 32 may be shorter than as shown. This
separation is
imparted by a void 36 centered about axis 24 and which starts at midput region
34 and
continues for the length of output stage 38 of the coupler. Input stage 32 may
be
essentially of zero length where void 36 starts at input port 14.
1o Coupler 30 also includes a break point region 40 at which the output arms
are
fully separated from each other along the direction of light propagation.
Peripheral voids
42 extend along the outer periphery of output stage 38 in elongated manner
along the
direction of light propagation. Voids 42 face outwardly with respect to axis
24 and are
smoothly integrated with a void 44 separating output arms 18a-18d immediately
downstream of break point region 40 (i.e., further along the path of light
transmission).
By "smooth" is meant that the cross section at any point along the axial
length transitions
to the next point without any substantial discontinuities.
Fig. 4 shows a further coupler 50 according to the invention. Coupler 50 is
generally similar to coupler 10 (Fig. 1), but its output arms split from each
other in an
2o "X" shape 26, as viewed along axis 24, that is non-planar so as to reduce
light loss owing
to the splitting between the arms. Non-planar "X" shape 26 will be further
described
below.
Figs. SA-SH show sequential cross sections of coupler 10 of Fig. 1, starting
with
input port 14 in Fig. SA, including an "X" shape 58 in Fig. SE where the
output arms
separate from each other, and ending with output ports 20a-20d in Fig. SH.
Figs. 6A-6G show sequential cross sections of coupler 30 of Fig. 3, starting
with
input port 14 and ending with the output ports in Fig. 6G. Figs. 6B-6E show
cross
sections of central void 36, with a periphery defining four arc segments 36a-
36d (Fig. 6D)
6
Case 2351 CA 02328213 2000-i2-i4
corresponding in shape to the inner peripheries 60 (Fig. 6E) of the of the
output arms
downstream of the point where the arms are fully separated from each other.
Peripheral
voids 42 are shown in Fig. 6D. Central and peripheral voids 36 and 42 assist
in
preserving the etendue of the transmitted light.
Figs. 7A-7G show sequential cross sections of coupler SO of Fig. 4, starting
with
input port 14 and ending with output ports 20a-20d. Splitting pattern 26, in
the form of a
non-planar "X", is shown through progressive cross sections, indicating that
the center of
the "X", in Fig. 7E, is downstream of the peripheral regions of the "X" in
Figs. 7B-7D.
With each of the three couplers 10, 30 and SO described so far, the outer
1o peripheries of the output arms are substantially parallel to each other. By
"substantially
parallel" is meant in the specification and claims less than 30% of the half
angle of output
light. (In other words, the output arms may be tilted, with respect to the
main object of
the sputter by 30% of the half angle of the output light). Further, the output
arms are
spatially separated from each other. As such, as shown in Fig. 8, output arms
18a and
18b can be beneficially connected to light guides 68a and 68b without removing
cladding
and outer jacket 70a and 70b from the light guides. A 4 mm spacing between
arms 18a
and 18b accommodates typical cladding and outer jacket of adjacent output
light guides
68a and 68b. Such spacing is preferably at least about 2 mm, and more
preferably at least
about 3.2 mm.
2o Desi ng_Examnle
Case 2351 CA 02328213 2000-i2-i4
Preferred steps for designing a coupler 70 (Fig. 12A) are illustrated in
connection
with Figs. 9A-12C. Coupler 70 has only two output arms, but is otherwise most
similar in
to coupler 30 of Fig. 3. Establishing rotational symmetry about central axis
24 (Fig. 12A)
helps ensure that the light is split equally into each output arm. Such
symmetry can also
be stated as rotational symmetry about axis 24, or as n-fold symmetry about
axis 24,
where "n" is the number of output arms. If desired, however, a coupler need
not employ
rotational or n-fold symmetry, whereby one arm may be larger than an adjacent
arm.
However, it is usually desired that each arm be spaced about the same distance
from axis
24.
to Diameters of the input and output ports 14 and 20a-20b (Fig. 12A) and the
spacing between output arms 18a and 18b are determined. It is preferred that
the area of
the input port substantially equals the combined area of the output ports. It
is also
preferred that the numerical aperture of the input port be substantially the
same as that at
the output ports. For ease of explanation, specific dimensions are described.
Assuming
i5 the diameter of the input port is 10 mm, the diameter of each of the two
output ports is
10/(2"~) mm, or about 7 mm. A center-to-center spacing between output ports of
11 mm
provides a 4 mm gap between ports, which accommodates typical cladding and
jacket on
downstream light guides as described above with respect to Fig. 8.
In Fig. 9A, a forward converter 72 is designed to start with the desired input
port
2o size and to end with a shape surrounding all of the output ports. The shape
of converter
72 is chosen to provide an appropriate angle-to-area conversion. While a
straight taper is
shown for converter 72, other tapers will be obvious to those of ordinary
skill in the art,
as now discussed.
As reported in X. Ning, R. Winston, and J. O'Gallagher, Appl. Optics, vol. 26,
no.
25 2 (Jan. 1987), pp. 300-305, a dielectric totally internally reflecting
concentrator can
transform a port of area A1 with maximum angle Theta 1 to an area A2 with
maximum
angle Theta 2, where A1*sin(Theta 1)z = A2*sin(Theta 2) Z. Such an angle-area
transformer is also called a Theta 1/Theta 2 converter by the foregoing Ning
et al.
8
Case2351 CA 02328213 2000-i2-i4
Reference.
One particular embodiment of a Theta 1/Theta 2 converter is a dielectric
compound parabolic concentrator (DCPC) which are described in W.T. Welford and
R.
Winston, High Collection Nonimaging Optics, New York: Academic Press, Inc.
(1989),
chapter 4 (pp. 53-76, 82-84). In some cases, a DCPC can be replaced with a
tapered cone
or two tapered cones attached to one another, except that the DCPC is
typically shorter in
length. Another method to implement an angle-area-converter is to combine a
gradient
index (GRID rod with a tapered cone, as discussed in A. Cutolo, et. al., Appl.
Optics,
vol. 29, no. 9 (March 20, 1990), pp. 1353-1363.
l0 Typically, angle-to-area converters have the same shape at the input and
output
ports. If the shapes are different, the skew invariant, such as discussed in
the foregoing
Welford et al. reference at pp. 228-230, may limit the performance of a Theta
1/Theta 2
converter. When the input and output areas are different, a slow taper from
one shape to
the other provides reasonable performance for a Theta 1/Theta 2 converter, as
suggested
by Garwin, R.L. in "The design of Liquid Scintillation Cells," Rev. Sci.
Instruments, vol.
23 (1952), pp. 755-757. Adjusting the cross-section along the taper may
provide
improved performance in some cases, especially if the aspect ratio of the
output ports is
much different than the input port, and the length of the coupler is to be
minimized.
Some investigations of nonrotationally symmetric, non-imaging optic devices
have been
2o performed recently. See "Nonrotationally Symmetric Reflectors for Efficient
and
Uniform Illumination of Rectangular Apertures', by Shatz, et. al., SPIE vol.
3428, pp.
176-183. An extension of the ideas presented by the foregoing Shatz article is
to create
an angle-to-area device which transitions from one shape to another where the
intermediate cross-sections have star-like cross-sections.
The shorter the forward converter 72 , the shorter the overall coupler
produced. A
longer coupler will ensure more thorough light mixing and make the illuminance
distribution more uniform at the break point (not shown). This can make
dividing of light
among the output arms more uniform. On the other hand, a shorter coupler is
desirable
9
Case2351 CA 02328213 2000-i2-i4
for compactness.
Fig. 9B shows input port 14 with a diameter of 10 mm. Fig. 9C superimposes
output ports 20a and 20b, shown in dashed lines, on face 74 of converter 72
(Fig. 9A).
With each output port having a 7 mm diameter and with a 4 mm spacing between
ports,
as assumed above, the minimum diameter of face 74 is 18 mm.
According to Fig. 10A, two angle-to-area converters 76a and 76b are designed,
and are referred to herein as sub-backward converters (or sub-converters). The
sub-
backward converters preferably each overlap the other in the vicinity of
numeral 79 so
that the axial shape of their left side is the same as the axial shape of
their right side. (In
1o contrast, coupler 10 of Fig. 1 and Figs. SA-SH was designed with the left-
hand ends of
the sub-converters being pie- or wedge-shaped, like the 90-degree portion of
coupler 10
as shown in Fig. SE, so that adjacent sub-converters abut each other.) Sub-
converters 76a
and 76b, as shown, are parallel to each other such that their axial center-to-
center spacing
(not shown) of 11 mm at the left-shown end is the same as the center-to-center
spacing
(not shown) at the right-hand end. The two sub-converters need not be parallel
to each
other, but if they are not parallel, then they may not be axisymmetric, that
is, rotationally
symmetric about the main axis (e.g., axis 24, Fig. 1).
Fig. lOC shows output ports 20a and 20b dimensioned as noted with respect to
Fig. 9C. Fig. lOB shows end faces 78a and 78b of converters 76a and 76b
dimensioned
2o so that, when such converters overlap, input port 14 is preferably at least
covered to
prevent light loss. Thus, the triangle at 80 shows that for input port 14 of
10 mm
diameter (5 mm radius), and with a center-to-center spacing between faces 78a
and 78b of
11 mm (or half spacing of 5.5 mm), the minimum diameter for faces 78a and 78b
is 16
mm each (radius of 8 mm). The 8 mm radius in triangle 80 is approximately the
square
root of the sum of 5.5 mmZ and S mm2.
Fig. lOD shows both converters 76a and 76b unioned together according to the
logic of the Boolean "OR" function, which, incidentally, eliminates any
overlap (e.g. at
79 in Fig. l0A)..
to
Case2351 CA 02328213 2000-i2-i4
Fig. 11A shows the unioned sub-backward converters 76a and 76b overlapping
forward converter 72, shown in dashed lines for clarity. Only a part of
converter 72, at
73, is directly visible. Fig. 11B shows the left-hand side of the overlapped
converters of
Fig. 11A, with input port 14 shown in dashed lines for clarity. Fig. 11C shows
the right-
s hand side of the overlapped converters, with converter 72 shown in dashed
lines for
clarity.
As shown in Fig. 12A, the overlapped converters are then intersected according
to
the Boolean "AND" function, so that only the portions of each which overlap
each other
remain in the intersected product, shown as coupler 70. Fig. 12B shows the
left-hand
t o side of coupler 82 with input port 14, and Fig. 12C shows its right-hand
side with output
ports 20a and 20b.
According to the foregoing design example, the sub-backward converters (or sub-
converters), each of which may be substantially axisymmetrical, are unioned
together and
then intersected with the forward converter. Alternatively, each sub-converter
can be first
15 intersected with the forward converter, producing a non-axisymmetrical
shape, before
being unioned to the other sub-converters. Both approaches can result in the
same shape,
so that they are mathematical equivalents of each other.
Coupler 10 (Figs. 1 and SA-SH) constructed according to the preferred
dimensions mentioned above achieves a light loss of only about 11 %, which is
the result
20 of 8% Fresnel reflection losses at the input and output ports and 3% from
losses at the
"X". This loss can be minimized through the use of index matching or anti-
reflection
coatings. Analysis of coupler SO (Figs. 4 and 7A-7H) also indicates a reduced
light loss
of only about 8% from Fresnel reflections, where the 3% loss at the "X" is
removed
because the "X" is no longer perpendicular to the main axis of light
propagation. The
25 combined area of output ports for the various couplers described is
preferably limited to
no more than approximately twice the area of the input port, and preferably is
less than
the mathematical product of 0.95, the number of output ports, and the area of
the input
port (e.g., 1.9 times the area of the input ports for the case of two output
ports). The
n
Case 2351 CA 02328213 2000-i2-i4
maximum cross sectional dimension of the output ports along the described axis
is
preferably less than the maximum cross sectional dimension of the input port
along the
axis.
The couplers described herein may be formed, for instance, from a clear
plastic
material such as acrylic, by molding, casting, machining, or related
fabrication methods.
The surfaces should be smooth to avoid scattering light passing through the
coupler by
total internal reflection. The couplers preferably each comprise an
integrated, single,
preformed unit, into which light guides, etc., can be coupled or integrated if
desired. In
some cases, it may be desirable to make the input to breakpoint portion of the
coupler
1o from one piece of material, and make the breakpoint to output portions
separately. In the
vicinity of each light port, a short cylindrical section can be added to
facilitate coupling to
light guides, etc. The couplers are preferably axially surrounded by a
protective cover
(not shown) having a lower index of refraction than the coupler.
While the invention has been described with respect to specific embodiments by
way of illustration, many modifications and changes will occur to those
skilled in the art.
It is, therefore, to be understood that the appended claims are intended to
cover all such
modifications and changes as fall within the true scope and spirit of the
invention.
~2
