Optical subassembly with port configuration

The present invention relates to optical subassemblies.

FIG. 1 illustrates an exemplary conventional optical subassembly. The device 100 comprises a glass slab 102 , a plurality of filters 104 a - 104 d coupled to the glass slab 102 , an input/output port 106 optically coupled to the glass slab 102 , and a plurality of ports 108 a - 108 d , each optically coupled to a filter 104 a - 104 d . Each filter 104 a - 104 d transmits a particular wavelength or range of wavelengths, i.e., a channel, while reflecting the remaining channels. An optical fiber can be positioned at the ports 106 and 108 a - 108 d.

For example, as a demultiplexer, the input/output port 106 transmits a beam comprising channels λ 1 -λ 4 through the glass slab 102 to the filter 104 a . Filter 104 a transmits λ 1 to port 108 a and reflects λ 2 -λ 4 through the glass slab 102 to filter 104 b . Filter 104 b transmits λ 2 to port 108 b and reflects λ 3 -λ 4 through the glass slab 102 to filter 104 c . Filter 104 c transmits λ 3 to port 108 c and reflects λ 4 through the glass slab 102 to filter 104 d . Filter 104 d transmit λ 4 to port 108 d . In this manner, the device 100 functions as a demultiplexer. As would be understood by one of ordinary skill in the art, the beam paths are approximately the same when the device 100 functions as a multiplexer except the channels are traveling in the opposite direction.

However, because the ports 108 a and 108 c are positioned at a different side of the device 100 than ports 106 , 108 b , and 108 d , the device 100 can be difficult to fit onto a board. One or more of the optical fibers at the ports 106 and 108 a - 108 d may need to be bent in order to accomplish the desired fit. This is especially difficult at the corners of the board.

Accordingly, there exists a need for a multiplexer/demultiplexer device with an improved port configuration. The improved port configuration should have all beams entering and exiting at the same side of the device, making the device easier to fit onto a board. The present invention addresses such a need.

The present invention provides an optical subassembly with an improved port configuration. The improved port configuration has all beams entering and exiting at the same side of the device.

In one aspect of the present invention, the device utilizes a core comprising a first and a second face, where the first and second faces are coupled, non-parallel, and non-co-planar for changing a path of a beam.

In another aspect of the present invention, filters are coupled to a third face of the core. An axis of the core is defined by the intersection of the first and second faces of the core. The light path traverses between the first or second face and each filter, such that, at each filter, no portion of the light path interferes with any other portion of the light path. In addition, the light path traverses the core in a direction along this axis.

In another aspect of the present invention, the device further comprises a common port and a plurality of ports residing at the same side of the device.

In another aspect of the present invention, the filters are optically coupled to the plurality of ports.

In another aspect of the present invention, the positions of the ports are adjustable to facilitate alignment.

In an exemplary embodiment, the positions of the ports are adjusted to facilitate alignment by first aligning the common port to a target at one of the plurality of ports, then aligning the plurality of ports to the common port.

In an exemplary embodiment, an optical device is provided, which comprises: a core, comprising a plurality of reflective faces; a plurality of filters, wherein each filter corresponds to one of a plurality of channels, wherein each filter is optically coupled to at least one of the plurality of reflective surfaces; a common port optically coupled to the core for transmitting a beam comprising the plurality of channels; and a plurality of ports, wherein each port is optically coupled to one of the plurality of filters, wherein each port transmits one of the plurality of channels, wherein the core and the plurality of filters multiplex the plurality of channels or demultiplex the beam, wherein the common port and the plurality of ports reside at a same side of the device.

In another exemplary embodiment, a multiplexing device is provided, which comprises: a core, comprising a plurality of reflective faces; a plurality of ports at a side of the device, wherein each port transmits one of a plurality of channels; a plurality of filters coupled to the core, wherein each filter corresponds to one of the plurality of channels, wherein each filter is optically coupled to the port that transmits its corresponding channel, wherein each of the plurality of filters transmits its corresponding channel to one of the plurality of reflective surfaces, wherein each channel is reflected to a common port such that the plurality of channels is multiplexed; and the common port at the side of the device, wherein the common port is optically coupled to the core for receiving the multiplexed plurality of channels.

In another exemplary embodiment, a demultiplexing device is provided, which comprises: a core, comprising a plurality of reflective faces; a common port optically coupled to the core at a side of the core for transmitting a beam comprising a plurality of channels to one of the plurality of reflective faces; a plurality of filters coupled to the core, wherein each filter corresponds to one of the plurality of channels, wherein each of the plurality of filters receives its corresponding channel from one of the plurality of reflective surfaces, wherein each filter transmits its corresponding channel and reflects any other channel; and a plurality of ports at the side of the device, wherein each port is optically coupled to one of the plurality of filters, wherein each port receives the corresponding channel transmitted by the filter.

Exemplary embodiments of the present invention may use one or more of the aspects described above, alone, or in combination.

FIG. 1 illustrates an exemplary conventional optical subassembly.

FIGS. 2A-2E illustrate an exemplary embodiment of an optical subassembly.

FIG. 3 is a flowchart illustrating an exemplary embodiment of a method for aligning the ports of the device.

FIG. 4 illustrates another exemplary embodiment of a packaging for the optics of the device.

FIG. 5 illustrates an exemplary embodiment of cascading multiplexing/demultiplexing devices.

The present invention provides an optical subassembly with an improved port configuration. The following description is presented to enable one of ordinary skill in the art to make and use the invention and is provided in the context of a patent application and its requirements. Various modifications to the preferred embodiment will be readily apparent to those skilled in the art and the generic principles herein may be applied to other embodiments. Thus, the present invention is not intended to be limited to the embodiment shown but is to be accorded the widest scope consistent with the principles and features described herein.

To more particularly describe the features of the present invention, please refer to FIGS. 2A through 5 in conjunction with the discussion below.

FIGS. 2A-2E illustrate an exemplary embodiment of an optical subassembly 200 . FIG. 2A illustrates an exploded isometric view; FIG. 2B illustrates a front cross-sectional view; FIG. 2C illustrates a side cross-sectional view; FIG. 2D illustrates a top cross-sectional view; and FIG. 2E illustrates an assembled isometric view of the device 200 .

The optics comprising the device 200 includes a common port 202 , a plurality of ports 204 a - 204 d , a plurality of filters 208 a - 208 d , and a core 210 . Each port 202 and 204 a - 204 d comprises a collimator coupled to an optical fiber, waveguide, or other light guiding medium. The position of each collimator is adjustable for alignment purposes. The core 210 comprises a first face 212 b and a second face 212 c . The first and second faces 212 b - 212 c are coupled, non-parallel, and non-co-planar for changing a path of a beam. The core 210 further comprises a third face 212 a , to which the filters 208 a - 208 d are coupled. The intersection of the first and second faces 212 b - 212 c defines an axis of the core 210 . In this embodiment, the axis is parallel to the y-axis. A light path traverses between the first 212 b or second 212 c face and each filter 208 a - 208 d , such that, at each filter 208 a - 208 d , no portion of the light path interferes with any other portion of the light path. Also, the light path traverses the core 210 in a direction along the y-axis. Thus, in this embodiment, a “zig-zag” light path results, as illustrated in FIG. 2 D. To facilitate this light path, it traverses an external surface of a filter at an angle. For example, the angle can be approximately 10° relative to the external surface's normal axis.

In the exemplary embodiment, the core 210 is a glass prism with a base face as the third face 212 a and side faces as the first and second faces 212 b - 212 c . Other shapes may be used for the core 210 . The side faces 212 b - 212 c are either coated with a reflective coating or uses total internal reflection (TIR) to traverse light through the core 210 , as described further below. The core 210 can also be doped to obtain desired characteristics, such as amplification or attenuation.

The filters 208 a - 208 d are coupled to the base face 212 a of the core 210 using any variety of methods, such as epoxy, heat fusion, and solder. Each filter 208 a - 208 d transmits a particular channel in a beam and reflects the remaining channels. Each filter 208 a - 208 d comprises a coating either on the surface coupled to the core 210 or on the external surface away from the core 210 . In addition, either surface of each filter 208 a - 208 d may be coated with an antireflective coating. The base face 212 a of the core 210 may also be coated with the anti-reflective coating.

For example, as a demultiplexer, a beam comprising channels λ 1 -λ 4 enters the device 200 through the common port 202 at an angle. In the exemplary embodiment, the angle is such that the beam enters the core 210 at approximately 10° relative to the z-axis. The beam comprising λ 1 -λ 4 enters the core 210 from the port 202 and reflects from the side face 212 c . Since the side face 212 c is coated, the beam is reflected to the side face 212 b . Since the side face 212 b is also coated, the beam is further reflected to the filter 208 a.

The filter 208 a transmits λ 1 to the port 204 a . Because the beam comprising λ 1 -λ 4 enters the device 200 at an angle, the filter 208 a transmits λ 1 at an angle. Thus, the port 204 a is also positioned at an angle relative to the z-axis such that the port 204 a is properly aligned to receive λ 1 . Ports 204 b - 204 d are also positioned for alignment to receive their respective channels. The filter 208 a reflects λ 2 -λ 4 to the side face 212 b . The side face 212 b reflects λ 2 -λ 4 to the side face 212 c , which in turn reflects these channels to the filter 208 b . (See FIG. 2B.)

The filter 208 b transmits λ 2 to the port 204 b and reflects λ 3 -λ 4 to the side face 212 c . The side face 212 c reflects λ 3 -λ 4 to the filter 208 c . The filter 208 c transmits λ 3 to the port 204 c and reflects λ 4 to the side face 212 b . The side face 212 b reflects λ 4 to the filter 208 d . The filter 208 d transmits λ 4 to the collimator 204 d.

Because λ 1 -λ 4 enter the device 200 at an angle, the channels travel down the core 210 along the y-axis (see FIG. 2 D), and exits the core 210 at an angle. Each of the ports 204 a - 204 d is positioned at an angle for alignment to properly receive their respective channels. The ports 202 and 204 a - 204 d are each adjustable by rotation of the collimator's “tail” (the end distal to the filters 208 a - 208 d ) about the θy- and θx-axes and by translation along the x- and y-axes.

As is understandable to one of ordinary skill in the art, the device 200 can function as multiplexer with λ 1 -λ 4 having similar paths through the device 200 but in the opposite direction. Channels λ 1 , λ 2 , λ 3 , and λ 4 are input into the device 200 through ports 204 a , 204 b , 204 c , and 204 d , respectively. The channel λ 4 is transmitted by filter 208 d to side face 212 c . Side face 212 c reflects λ 4 to the side face 212 b . Side face 212 b reflects λ 4 to filter 208 c.

Filter 208 c transmits λ 3 and reflects λ 4 to side face 212 b . Side face 212 b reflects λ 3 -λ 4 to side face 212 c . Side face 212 c reflects λ 3 -λ 4 to filter 208 b . Filter 208 b transmits λ 2 and reflects λ 3 -λ 4 to side face 212 c . Side face 212 c reflects λ 2 -λ 4 to side face 212 b . Side face 212 b reflects λ 2 -λ 4 to filter 208 a . Filter 208 a transmits λ 1 and reflects λ 2 -λ 4 to side face 212 b . Side face 212 b reflects λ 1 -λ 4 to side face 212 c . Side face 212 c reflects λ 1 -λ 4 to the common port 202 .

In this manner, channels in a beam is demultiplexed or multiplexed with the ports 202 and 204 a - 204 d . In this exemplary embodiment, the ports 202 and 204 a - 204 d are on the same side of the device 200 . This allows the device 200 to fit more easily onto a board. For example, the device 200 can be placed at a corner of the board without requiring any of the optical fibers at the ports 202 and 204 a - 204 d to be bent. Alternatively, a retroreflector or mirror may be used in place of the common port 202 . The common port 202 is repositioned to a side of the device 200 opposite to the other ports 204 a - 204 d . The retroreflector redirects a light beam to or from the newly placed common port 202 .

FIGS. 2A-2E also illustrate an exemplary embodiment of a packaging for the device 200 . The packaging comprises a chassis 206 , bars 214 , and a chassis base 216 . The chassis 206 comprises metal with bores 218 and 220 a-d . The “head” (the end proximate to the filters 208 a - 208 d ) of the collimator of the common port 202 is coupled to the chassis 206 within the bore 218 . The heads of the collimators of the ports 204 a - 204 d are each coupled to the chassis 206 within one of the bores 220 a - 220 d . The coupling can be accomplished through the wicking of thermally cured adhesive, the use of ultraviolet light cured adhesive, use of glue around the edges of the ports 204 a - 204 d so that there is no glue in the optical path, solder, epoxy, welds, or any other coupling means. In the exemplary embodiment, the bores 218 and 220 a - 220 d are tapered, with the top of the bore being wider than the bottom of the bore, to allow for the rotational alignment of the ports 202 and 204 a - 204 d . The width of the bores 218 and 220 a - 220 d are such that translational alignment of the ports 202 and 204 a - 204 d are possible. Other configurations of the bores 218 and 220 a - 220 d are possible. The chassis 206 also comprises an opening (not shown) at the bottom of the chassis 206 that lead to a cavity 222 with the chassis 206 (see FIGS. 2 B- 2 C).

The packaging further comprises bars 214 and a chassis base 216 . The bars 214 reside within the chassis base 216 , so that the side faces 212 b - 212 c of the core 210 abut against the bars 214 and is properly supported. The chassis base 216 , bars 214 , core 210 , and filters 208 a - 208 d are then placed within the cavity 222 through the opening. An assembled and packaged device 200 is illustrated in FIG. 2 E.

FIG. 3 is a flowchart illustrating an exemplary embodiment of a method for aligning the ports of the device 200 . The alignment of the device 200 is accomplished in two parts, with an alignment of the common port 202 and alignments of each of the other ports 204 a - 204 d . First, the device 200 is mounted to an alignment station, via step 302 . Next, the common port 202 is aligned, via step 304 .

In aligning the common port 202 , a target is first placed in one of the bores, such as bore 220 d , via step 304 a . The target can be a standard collimator that is glued to a sleeve which fits the target in the bore 220 d in a nominal position. Next, the target is connected to a laser source, via step 304 b , and the common port 202 is mounted to a chuck of the alignment station, via step 304 c . The common port 202 is then aligned using an alignment software, via step 304 d . The alignment software first positions the common port 202 in the best known nominal position above the chassis 206 (but not in the chassis 206 ), via step 304 d (i), i.e., along the Z-axis. Next, the software performed algorithms that aligns the common port 202 in the U, V, X, and Y-axes, via step 304 d (ii). The U-axis rotates about the X-axis, and the V-axis rotates about the Y-axis. The alignment algorithms may include a spiral search algorithm to find an initial start position for alignment with power above a certain threshold and a hill climb algorithm for finding a position with optimized lighting. A surface fitting approach, raster scan or other algorithm may also be provided.

The alignment algorithms may be used to incrementally step through different positions along an axis of motion. The axis is expected to have the greatest impact on alignment may be used first, followed by movement along less significant axes. The process may be iterated until a desired alignment has been achieved.

In the following, the algorithm to align the XYZ position of an optical component is described. The UVZ position may be similarly adjusted. First, a spiral scan or raster scan in the U and V axes may be used to find a power reading above some threshold. After the threshold is reached, a 2-dimension auto alignment algorithm is run to align XY position to the maximum power. The 2-dimension auto alignment algorithm may include: Spiral Scan, Raster Scan, XYZ Hill Climb, and YXY Hill Climb.

For the Hill Climb algorithm, the following three parameters are specified: initial step size, number of check points, and number of iterations. The “step size” is a parameter that determines the magnitude of motion along each axis. “Check points” is the parameter that specifies the number of steps the algorithm takes past each maximum point in order to check to see if the hill would begin to rise again or not. After locating the absolute peak of the hill, the Hill Climber reduces the size of its step size by a factor of two and goes climbing in the reverse direction. The Hill Climber repeats the process and passes over the hill as many times as indicated by a parameter referred to as “iterations”. This is done in order to fine tune the alignment. Upon arrival to the top of the peak the next time, the Hill Climber no longer crosses over it and rests at the top. Once the Hill Climb process is completed along one axis, it is repeated along the other axis and after that once again along the first axis.

After the above Hill Climb method is performed, a fine alignment may be performed. While the Hill Climb finds a position with optimal power, the power may drop off much more rapidly by movement in one direction along the axis rather than the other direction. In such cases, it may be desirable to center the alignment point in between points along the axis where the power starts to drop below a desired threshold (such as 99% of the optimum power found using the Hill Climb method). The Hill Climb method is first completed to determine an initial optimum alignment position. The component is then moved in a positive direction along the first axis. A check is made to see if the power is still above 99% of the initial optimum. If so, the position is recorded and another step is taken along the axis is taken. This continues until the power drops below 99% of optimum. Once the power drops below 99%, the method moves back to the last position that was above 99%. This is recorded as an axis point (X1). This marks the last position before the power drops below the 99% threshold due to movements in the positive direction along the first axis.

Then, the algorithm steps along the axis in the negative direction. A check is made to see if the power remains above 99% of optimum. If so, the position is recorded and another step is taken along the axis in the negative direction. This is repeated until the power falls below 99% of optimum. Once the power drops below 99%, the method moves back to the last position that was above 99%. This is recorded as an axis point (X2). This marks the last position before the power drops below the 99% threshold due to movements in the negative direction along the first axis. The mid point between X1 and X2 is calculated (i.e., the mid point between the positions along the axis where the power falls below 99% of optimum). This allows for the same alignment error in either direction before the power drops below the threshold. The fine alignment is then repeated for the other axes. The fine alignment repeats in the same manner as the Hill Climb algorithm. The first axis and second axis are aligned and then the first axis is aligned again. For a three axis alignment, the third axis is then aligned and alignment of the first axis, second axis and first axis are then repeated again. For a four axis alignment, the third axis and fourth axis are aligned. Then the third axis is aligned again. Alignment of the first axis, second axis and first axis are then repeated again.

This alignment process is further described in the co-pending U.S. patent application titled “System and Method For Optical Multiplexing and/or Demultiplexing”, Ser. No. (2200CIP), filed on Jul. 26, 2002, assigned to the assignee of the present application. This co-pending application is hereby incorporate herein by reference.

Once the common port 202 is aligned along the U, V, X, and Y axes, the center wavelength of filter 204 d at the target is next adjusted until it is within the specifications, via step 304 d (iii). The center wavelength of the filter 208 d is a direct result of the angle of incidence of the laser beam from the target's laser source on the filter 208 d . The center wavelength of the filter 208 d can thus be adjusted by adjusting the angle of incidence. Next, the V angle of the common port 202 is adjusted until the center wavelength is also within the specification, via step 304 d (iv). Once the common port 202 is aligned with the center wavelength within the specification, the common port 202 is fixed in place to the core, via step 304 e . The center wavelength may be checked again to ensure that the alignment has been performed correctly.

Once the common port 202 is aligned, the target is removed from the chassis 206 . The device 200 is then relocated on the alignment station to positioned it for alignment of the rest of the ports 204 a - 204 d , via step 306 . Assume that port 204 d is being aligned. First, the common port 202 is connected to a laser source for providing the wavelength corresponding to the port 204 d , via step 306 a . The port 204 d is mounted into the chuck, via step 306 b . This port 204 d is then aligned using the alignment software, via step 306 c . The alignment software first positions the port 204 d in the best known nominal position above the chassis 206 , via step 306 c (i). The alignment of the port 204 d for the U, V, X, and Y axes is then performed, via step 306 c (ii) in the same manner as with the common port 202 above. Once aligned, the port 204 d is fixed in place to the chassis 206 , via step 306 d . Steps 306 a - 306 d are repeated for each of the other ports 204 a - 204 c.

FIG. 4 illustrates another exemplary embodiment of a packaging for the optics of the device 200 . In this illustrated embodiment, the packaging comprises a chassis 404 with a bore 404 for the common port 202 and bores 406 a - 406 d for the ports 204 a - 204 d . However, in the chassis 404 , the opening 408 to the cavity 410 is at the side of the chassis 404 rather than the bottom. The core 210 and filters 208 a - 208 d are placed within the cavity 410 through the opening 408 without support from bars 214 or a chassis base 216 . A plate 412 can be used to seal the opening 408 using any one of a variety of mechanisms.

Although the exemplary embodiments are illustrated with the demultiplexing and multiplexing of four channels, any number of channels can be supported by the device without departing from the spirit and scope of the present invention.

For example, two or more of the devices 200 can be cascaded to demultiplex or multiplex eight channels. FIG. 5 illustrates an exemplary embodiment of cascading multiplexing/demultiplexing devices 200 - 1 and 200 - 2 . The first device 200 - 1 comprises the same optics as the device 200 except the common port 202 - 1 comprises a dual collimator and the addition of a fifth filter 402 coupled to the core 210 - 1 and optically coupled to the dual collimator 202 - 1 . The surface of the core 210 at which the fifth filter 402 is coupled can be angled to maintain the proper angle for the beam traversing between the core 210 and the dual collimator 202 - 1 .

A beam with eight channels λ 1 -λ 8 is input into the first device 200 - 1 from the input of the dual collimator 202 - 1 . The fifth filter 402 transmits channels λ 1 , λ 2 , λ 3 , and λ 4 , which are demultiplexed and exit the first device 200 - 1 through ports 204 - 1 a , 204 - 1 b , 204 - 1 c , and 204 - 1 d , respectively. The fifth filter 402 also reflects the remaining channels λ 5 -λ 8 to the output of the dual collimator 202 - 1 . The channels λ 5 -λ 8 are input into the second device 200 - 2 . The second device 200 - 2 comprises the same optics as the device 200 . The channels λ 5 , λ 6 , λ 7 , and λ 8 are demultiplexed and exit the second device 200 - 2 through ports 204 - 2 a , 204 - 2 b , 204 - 2 c , and 204 - 2 d , respectively.

When multiplexing, the channels λ 1 , λ 2 , λ 3 , and λ 4 are input into the first device 200 - 1 via the ports 204 - 1 a , 204 - 1 b , 204 - 1 c , and 204 - 1 d , respectively. These channels are multiplexed and transmitted to the fifth filter 402 . The channels λ 5 , λ 6 , λ 7 , and λ 8 are input into the second device 200 - 2 via the ports 204 - 2 a , 204 - 2 b , 204 - 2 c , and 204 - 2 d , respectively. These channels are multiplexed and transmitted to the fifth filter 402 of the first device 200 - 1 via port 202 - 2 and the dual collimator 202 - 1 . The channels λ 5 -λ 8 are reflected by the fifth filter 402 to the output of the dual collimator 202 - 1 , while channels λ 1 -λ 4 are transmitted through the fifth filter 402 to the output of the dual collimator 202 - 1 . A multiplexed beam comprising λ 1 -λ 8 results.

Although the exemplary embodiments are described above as operating as a multiplexer and a demultiplexer, the device 200 may be used to perform other functions, such as an optical add/drop module, without departing from the spirit and scope of the present invention.

An optical subassembly with an improved port configuration has been disclosed. The device utilizes a core comprising a first face and a second face, where the first and second faces are coupled, non-parallel, and non-co-planar for changing a path of a beam. The core further comprises a third face, to which are coupled a plurality of filters. An axis of the core is defined by the intersection of the first and second faces of the core. The light path traverses between the first or second face and each filter, such that, at each filter, no portion of the light path interferes with any other portion of the light path. Also, the light path traverses the core in a direction along this axis. This is facilitated by the light path traversing an external surface of a filter at an angle. In an exemplary embodiment, the core reflects beams such that channels to be multiplexed or demultiplexed enter and exit the device at the same side. In an exemplary embodiment, the core has a prismatic shape with a base face and coated side faces. Filters are coupled to the base face with adjustable ports comprising collimators optically coupled to the filters. The side faces reflect channels so that they travel down the core and exit or enter the device via the adjustable ports.

Although the present invention has been described in accordance with the embodiments shown, one of ordinary skill in the art will readily recognize that there could be variations to the embodiments and those variations would be within the spirit and scope of the present invention. Accordingly, many modifications may be made by one of ordinary skill in the art without departing from the spirit and scope of the appended claims.