JPS6239402B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an optical crossbar switch that efficiently switches connections between a plurality of fibers used in an optical exchange communication path with low loss.

Conventionally, as this type of device, for example,
-The one seen in No. 29223 is known. Its configuration is as shown in FIG. 1, with input optical fibers 1 ₁ , 1 _{2 .} . . and output optical fibers 2 ₁ , 2 ₂ .
are arranged on the same plane in a direction in which the respective central axes 6 intersect. The light rays from the input optical fibers 1 ₁ , 1 _{2 .} . . are converted into parallel rays by the focusing optical elements 3 ₁ , 3 ₂ . is reflected in the direction of the central axis 6 of the desired output optical fiber 2 ₁ , 2 ₂ . . . by any reflecting mirror ₅ on the central axis ₆ , and the output is Side optical fibers 2 ₁ , 2 ₂ ... light is focused on the end faces. in this case,
A light ray from a point light source can be converted into a parallel light beam by a converging optical element ₁ or the like, but the end face of an optical fiber is not a point light source, but has a certain radius (usually tens to hundreds of μm in multimode fibers). ]), it cannot be completely converted into parallel rays. In addition, in the case where the focusing optical elements used on the input side and those used on the output side are not confocal arranged along the optical path, the input side optical fibers 1 ₁ , 1 ₂ . . . output from the end face. The angle of the light beam with respect to the central axis 6 of the output optical fiber 2 ₁ ,
2 ₂ It can be seen from a simple drawing that the light is not conserved when it is focused on... Therefore, the following drawbacks arise.

(b) Since the light is not completely parallel, the light beam from the input optical fiber will spread out, and unless a large-diameter focusing optical element is used, a portion of the light beam will be blocked by vignetting and reach the end face of the output fiber. Light is not focused and losses occur. However, the use of a large-diameter focusing optical element increases the volume of the switch, which is undesirable.

(b) The amount of the loss depends on the distance between the end faces of the focusing optical element, so in the configuration shown in Figure 1, the distance varies depending on the connection path, and the non-uniformity of the connection loss occurs. arise.

(c) The distance between the end faces of the above-mentioned focusing optical element, that is, the optical path length within the switch, increases in proportion to the number of input and output optical fibers arranged, but the loss also increases for the above reasons. , it is difficult to construct a large-scale switch that can accommodate many optical fibers.

(d) Since the angles of the light rays with respect to the central axis are not conserved at the end faces of the input and output optical fibers, the output optical fiber is excited with a distribution different from that in the input optical fiber, resulting in so-called mode conversion. will occur.

In order to solve these drawbacks, the present invention arranges the optical fiber end faces so that all optical path lengths between arbitrary input and output optical fiber end faces are equal, and uses a confocal arrangement for the optical system in the middle. This system connects input and output optical fibers using a complete imaging relationship that includes distribution preservation, and also enables the construction of large-scale optical switches with low connection loss.

FIG. 2 shows the relationship between the incident light and the outgoing light in the confocal lens system used in the present invention, and the lens will be explained as a thin lens for convenience.

In the figure, two lenses with the same focal length F
L ₁ and L ₂ are arranged with their optical axes aligned at an interval twice the focal length, resulting in a so-called confocal arrangement. In this optical system, 2F from the front of one lens
A light ray whose light source E is located at a distance h from the optical axis within the range forms a same-size inverted image behind the other lens.

In this case, the light ray emitted from the light source E at an angle to the optical axis also has an angle to the optical axis at the imaging point, and it can be seen that the angle of the light ray is preserved. Therefore, by arranging the optical fiber end faces in an imaging relationship before and after this system, coupling without mode conversion is possible. However, with this system as it is, it is not possible to increase the ratio between the lens spacing, that is, half the distance F between lenses _L1 and _L2 , and the maximum diameter of the luminous flux, making it unsuitable for use in a matrix-like multi-input/output switch. . For example, assume that the numerical aperture of the optical fiber is α, the core radius is a, and the end face of the optical fiber is placed at a position where the maximum diameter of the light beam can be minimized, that is, at the front focal plane of the lens _L1 . At this time, the maximum diameter of the luminous flux 2Rmax
is 2Rmax=2(a+Fα), and the ratio with F is Therefore, for example, when using a general optical fiber and setting a = 0.03 mm and α = 0.2, F/2Rmax is approximately 2.5 regardless of the focal length F, which is difficult for a matrix switch larger than 2 x 2. Indicates that configuration is difficult.

Next, an embodiment of the present invention will be described in which this system is combined with another optical system to use an optical system that can suppress the maximum diameter of the luminous flux even if the lens interval 2F is increased.

FIG. 3 shows only one optical path in an embodiment in which four focusing optical elements are placed on one optical path in the present invention.

In the figure, lenses L ₁ and L ₁ ′ have a focal length F ₁ , and lenses L ₂ and L ₂ ′ have a focal length F ₂ , and will be described as thin lenses for convenience.

F ₁ , F ₁ + F ₂ , in order from the input optical fiber end face Q ₁
2F ₂ , F ₁ +F ₂ , Lens L ₁ with an interval of F ₁ ,
L ₂ , L' ₂ , L' ₁ , and the output optical fiber end face Q ₂ are arranged with their optical axes aligned with each other. In this optical system, the relationship between the 1st space and the 1st space in the figure is the same as in Figure 2, and the relationship between the 1st space and the 1st space is the same as the relationship between the 1st space and the 1st space, so the input optical fiber end face Q The light rays emitted from ₁ at an angle to the optical axis are incident on the exit side optical fiber end face at the same angle, and a space is formed between both end faces F ₁ /F ₁
Since there is an imaging relationship of equal magnification via a real image of double magnification, combination without mode conversion is possible. Here, the distance between the input optical fiber end surface Q ₁ and the lens L ₁ and the distance between the exit optical fiber end surface Q ₂ and the lens L' ₁ do not have to be F ₁ , but the arrangement shown in FIG. Sometimes the maximum diameter of the luminous flux becomes the minimum and becomes =''. moreover,
When the numerical aperture of the input optical fiber is α (=tan) and the core radius is a, the maximum diameter of the luminous flux is 2 = 2 {(F ₁ + F ₂ )tanθ−(a−F ₁ α)}, Here, since tanθ=a/F ₁ , substitute it into the formula and calculate the ratio with F ₂ , it becomes 2/F ₂ = 2 {a/F ₁ + F ₁ / (F ₂ α)}, and F ₂ F ₁ which minimizes the expression while holding constant
When we find _{the relationship between F 2 and} growing. This means that in the configuration of the optical crossbar switch shown below, when F ₂ is constant and F ₁ is used that satisfies the formula, the maximum number of optical fibers can be installed in the structure without causing vignetting. It shows.

FIG. 4 shows an embodiment of the present invention in the case of 4 inputs and 4 outputs using the optical system shown in FIG.

In the figure, optical fibers 1 and 2 having convergent light transmitters 7 and 8 at their tips are arranged so that their respective optical central axes 6 form a grid on the same plane. The convergent light transmitters 7 and 8 are cylindrical lenses having a distribution in which the refractive index of the medium decreases in proportion to the square of the distance from the optical central axis in a plane perpendicular to the light beam transmission direction. .

The central axes 6 of the four mutually parallel input optical fibers 1 and the four mutually parallel output optical fibers 2 intersect at a certain angle θ, and the bisector of the angle θ at the intersection point l ₁ two straight lines parallel to and sandwiching the intersection points
The end face of the input optical fiber 1 is arranged on one of the straight lines l ₂ and l ₃ , and the end face of the output optical fiber 2 is arranged on the other straight line. The reflection device 5' is a device whose reflection and transmission can be controlled mechanically or electrically, and its function is the same as that shown in FIG. Assuming that the length of the convergent light transmission body 7 is Z and the focal length of the convergent optical element 3 is _F2 , the optical axis of the principal plane of each convergent optical element 3 coincides with the position at a distance of Z + _F2 from the end face of the optical fiber. Place it like this.

Here, the length Z of the focusing transmitter 7 is 1/4 of its natural period length, and the focal point is located on its end surface.

In the arrangement shown in FIG. 4, the optical path length from the end face of any input optical fiber 1 to the end face of any output optical fiber 2 can be made equal, so that the two straight lines l ₁
By appropriately determining the positional relationship between and _l2 , the optical system between the end face of any input optical fiber 1 and the end face of any output optical fiber 2 can be made in the same state as shown in FIG. Therefore, the end face of any incoming optical fiber 1 and the end face of any outgoing optical fiber 2 can be optically completely coupled in the same state as shown in FIG. 3, allowing connection switching with low loss and no mode conversion. can. In this case, the ratio of the width of the area where the central axes 6 of the input and output optical fibers intersect (equal to F ₂ ) and the maximum diameter of the luminous flux, that is, the reciprocal of the equation F ₂ /2, is F as explained in FIG. Since it can be optimized (maximized) by changing _{F 1} and is an increasing function of F ₂ , it is possible to construct a multi-input/output switch equipped with many optical fibers. For example, if you use the same optical fiber as in the calculation example in Figure 2, set numerical aperture α = 0.2, core radius = 0.03 mm, and use a lens with F ₂ = 30 mm, F ₁ = √4.5.
The optimized F ₂ /2 at ≒2.1mm is 17.7, making it possible to configure switches up to 17×17.
Increasing F ₂ allows even larger switch configurations.

In addition, since the image created in the middle of the optical path with this switch is an enlarged image with a magnification of F ₂ /F ₁ (F ₁ < F ₂ ),
Tolerances for axis deviations caused by alignment errors in parallel movement of the converging optical element and the reflecting device and distance variations between lenses can be made larger than in the case of directly connecting the optical fiber end faces.

FIG. 5 shows only one optical path in an embodiment in which six focusing optical elements are placed on one optical path in the present invention.

In the figure, the focal length of the lens is
L ₀ , L′ ₀ is F ₀ , lens L ₁ , L′ ₁ is F ₁ , lens L ₂ ,
L' ₂ is F ₂ , and for convenience, it will be described as a thin lens.

F ₀ , F ₀ +F ₁ , in order from the input optical fiber end face Q ₁
F ₁ +F ₂ , 2F ₂ , F ₁ +F ₂ , F ₀ +F ₁ , Lenses L ₀ , L ₁ , L ₂ , L' ₂ , L' ₁ , L' ₀ , Output light at intervals of F ₀ The fiber end faces _Q2 are arranged with their optical axes aligned. In this optical system, the relationship between the 1st space and the 1st space in the figure is the same as in Figure 2,
Since the relationship between the 1st space and the 1st space is the same as the relationship between the 1st space and the 1st space, the input optical fiber end face
The light rays emitted from Q ₁ at an angle to the optical axis are incident on the exit optical fiber end face Q ₂ at the same angle, and
Since there is a same-magnification imaging relationship between the two end faces via the F ₁ /F ₀ times real image formed in the ′th space, coupling without mode conversion is possible. Here, the distance between the input optical fiber end surface Q ₁ and the lens L ₀ and the distance between the exit optical fiber end surface Q ₂ and the lens L' ₀ do not have to be F ₀ , and the distance between the lens L ₀ and the lens L ₁ is also possible. Although it is possible that the distance and the distance between lenses L' ₀ and L' ₁ are not F ₀ + F ₁ , the maximum diameter of the luminous flux is the minimum in the arrangement shown in Fig. 5, and 2 = 2'' becomes. Furthermore, the numerical aperture of the input optical fiber is α (≒tan),
When the core radius is a, R ₀ = a + F ₀ α R ₁ = R ₀ − (F ₀ + F ₁ ) tan θ ₁ = (F ₁ + F ₂ ) tan θ ₂ − R ₁ , and tan θ ₁ = a/F ₀ tan θ ₂ =R ₀ -F ₀ tanθ ₁ /F ₁ =F ₀ /F ₁
Since α, the diameter of the luminous flux and
Calculating the ratio with F ₂ : 2/F ₂ = 2 {F ₀ /F ₁ α+F ₁ /F ₀ F ₂ a}
Then, holding F ₂ constant, the formula is minimized with F ₀ and
The relationship between F ₁ and F ₂ is found as follows: F ₁ /F ₀ = (α/aF ₂ ) 〓 If a constant F ₂ is used, then when F ₀ and F ₁ shown in the formula are used, the formula The reciprocal of F ₂ /2
becomes the largest.

This means that in the configuration of the optical crossbar switch shown below, lenses L ₀ , L _{′ 0 , L 1} and L′ ₁ , it is shown that the largest number of optical fibers can be attached structurally without vignetting.

FIG. 6 shows an embodiment of the present invention in which the optical system shown in FIG. 5 is used and has four inputs and four outputs.

The positional relationship between the optical fiber end face and the reflecting device 5' is the same as in FIG. 4. The function of the reflection device is also the same as in the case of FIG. Focusing light transmitter 7,
The length Z of 8 is 1/4 of its natural period length, the convergence point is located on its end surface, and its focal length is
F ₀ , the focal length of the first focusing optical elements 3 and 4 is
F ₁ , the focal length of the second focusing optical element 3', 4' is
_F2 are arranged along the optical fiber central axis 6 with their optical axes aligned so as to satisfy the relationship shown in FIG. As in the case of FIG. 3, with this arrangement method, the optical path length from any input optical fiber end face to any output optical fiber can be made equal, so the positional relationship between the two straight lines _l1 and _l2 can be adjusted appropriately. By defining the entire optical system between the end face of any input optical fiber and the end face of any output optical fiber,
The same state as in Fig. 5 can be achieved. Therefore, optical fibers can be completely coupled between the end faces of arbitrary optical fibers in the same state as shown in FIG. 5, so that connection switching can be performed with low loss and without mode conversion. In this case, the ratio of the width of the area where the central axes 6 of the input and output optical fibers intersect (equal to F ₂ ) and the maximum diameter of the luminous flux, that is, the reciprocal of Eq.
As explained in Figure 5, F ₂ /2 can be optimized (maximized) by changing F ₀ and F ₁ , and is an increasing function of F ₂ , so it can be used for multiple inputs with many optical fibers attached. Output switches can be configured. For example, using the same optical fiber as in the calculation example in Figure 2, numerical aperture α=
0.2, core radius = 0.03mm, and using a lens with F ₂ = 30mm, the optimized F ₂ /2 is 17.7 when F ₁ /F ₀ = √200≒14.1mm, and in the case of the switch in Figure 4. Similarly, switches up to 17x17 can be configured, and by increasing _F2 , even larger switches can be configured. In addition, in this switch, the light rays in the middle of the optical path are close to parallel, and the light beam is thicker than the core diameter of the optical fiber. , the tolerance for distance variation between the focusing optical elements can be made larger than when the optical fiber end faces are directly connected.

Next, a method for reducing the deviation of the optical axis caused by a setting error of the reflecting surface of the reflecting device 5' will be explained with reference to FIG.

FIG. 7 is an enlarged view of the vicinity of the reflection device in the optical path between the pair of input and output optical fibers in FIGS. 4 and 6.

The central axes 6 of the input and output optical fibers intersect so that the angle of reflection of the light beam is more acute than 90 degrees. If the reflection angle is θ, and if the setting error of the reflection surface is Δm in the direction perpendicular to the reflection surface, then the axis deviation amount Δb of the reflected light optical axis 6' is Δb=2Δmsinθ/2, and it is necessary to reduce θ. Accordingly, the allowable value of Δm can be increased. For example, compared to θ=90°, θ=
In the case of 60°, the allowable value can be multiplied by √2.

It is clear that in the converging optical elements used in the explanation so far, a switch having the same effect can be constructed by using a converging optical transmitter or a normal convex lens.

The effects of the present invention are listed below.

1. Since the optical path lengths from the input optical fiber end face to the output optical fiber end face are all equal, variations in connection loss due to optical path differences do not occur.

2 The confocal arrangement of the lens preserves the angle of the light rays at the input and output end faces, and uses a perfect imaging relationship that focuses all incident light on the exit fiber end face, resulting in low loss and no mode conversion. Switch can be realized.

3 By appropriately selecting the focal lengths of the two types of lenses, the ratio of the distance between the two inner lenses along the optical path to the maximum beam diameter can be increased, so the switch size can be increased without sacrificing connection characteristics. It is possible to scale up, that is, configure a switch with a large number of input/output terminals.

4. By converting the light from the optical fiber into a thick beam before switching, and by setting the reflection angle of the light beam at an acute angle, it is possible to suppress the increase in splice loss caused by optical system misalignment and setting errors. .

[Brief explanation of the drawing]

1 is an explanatory diagram of a conventional example, FIG. 2 is a diagram illustrating the optical relationship of a confocal arrangement lens system, FIG. 3 is a diagram illustrating the optical relationship of an embodiment of the present invention, and FIG. The figure is an explanatory diagram of one embodiment of the present invention, FIG. 5 is a diagram explaining the optical relationship of another embodiment of the present invention, and FIG.
The figure is an explanatory diagram of another embodiment of the present invention, and FIG. 7 is an explanatory diagram regarding the arrangement of the reflecting device. In the figure, 1 is an input optical fiber, 2 is an output optical fiber, 3 and 4 are convergent optical elements, 5' is a reflection device, and 7 and 8 are convergent optical transmission bodies.

Claims (1)

[Claims] 1 M and N are positive integers of 2 or more, and the centers of N mutually parallel input optical fibers and M mutually parallel output optical fibers on the same plane The axes intersect at a certain angle θ, and the reflection angle θ of the ray at the intersection point is 2
The end face of the incoming optical fiber is placed on one side of two straight lines that are parallel to the equal dividing line and sandwich the group of intersection points, and the end face of the outgoing optical fiber is placed on the other side, and from each intersection point position to the central axis. The sum of the distances to the end faces of the incoming optical fiber and the outgoing optical fiber measured along the central axis is made equal at all intersection points, and the light from the incoming optical fiber is transmitted at each intersection point of the central axis. Alternatively, a reflection device is provided to reflect the light in the direction of the central axis of the output optical fiber, and an optical system that forms an image of the input optical fiber end face onto an arbitrary output optical fiber end face is aligned with the central axis of each optical fiber. An N-input, M-output optical crossbar switch characterized by having optical axes aligned with each other. 2. In the N-input M-output optical crossbar switch according to claim 1, the optical system includes:
Focal lengths F ₁ , F ₂ , F ₂ , F ₁ in order along the optical path between the end faces of any input optical fiber and output optical fiber.
The four lenses are mutually confocal arranged with adjacent lenses along the optical axis, and the intersection point group has a focal length
1. An N-input, M-output optical crossbar switch located between lenses of F ₂ and characterized in that the end face of the optical fiber is placed on the focal plane of the lens of focal length F ₁ . 3. In the N-input M-output optical crossbar switch according to claim 1, the optical system includes focal lengths F ₀ and F ₁ in order along the optical path between the end faces of arbitrary input optical fibers and output optical fibers. ,F ₂ ,F ₂ ,
Six lenses F ₁ and F ₀ are placed confocal with adjacent lenses along the optical axis, the intersection point group is located between the lenses with focal length F ₂ , and the end face of the optical fiber is in focus. An N-input M-output optical crossbar switch, characterized in that it is placed on the focal plane of a lens with a distance _F1 . 4. In the N-input M-output optical crossbar switch according to claim 2, where the core radius of the input optical fiber and the output optical fiber is a, and the numerical aperture is α, the focal lengths of the lenses F ₁ and F ₂ An N-input, M-output optical crossbar switch is characterized in that F ₁ =(a/αF ₂ )〓 is satisfied. 5. In the N-input M-output optical crossbar switch according to claim 3, where the core radius of the input optical fiber and the output optical fiber is a, and the numerical aperture is α, the focal lengths of the lenses F ₀ , F ₁ , F ₂ satisfy the following condition: F ₁ /F ₀ =(α/aF ₂ )〓. 6. In the N-input M-output optical crossbar switch according to claim 1, 2, 3, 4, or 5, the end face of any incoming optical fiber and any outgoing optical fiber. When the centers of are P and Q, and the intersection of the central axes extending from P and Q is X, the end faces of the input optical fiber and the output optical fiber are arranged at positions where ∠PXQ is smaller than 90 degrees. N characterized by
Input M output optical crossbar switch.