JPH03215811A - Matrix optical switch - Google Patents

Abstract

PURPOSE:To offer a low-loss, small-sized, and high-density multi-terminal matrix optical switch which can hold itself by holding a chip made of a mirror which reflects in-waveguide propagated light with the attractive force of an electromagnet and moving it in a waveguide which is formed in a lattice shape. CONSTITUTION:A three-axial positioning device 22 positions a positioning head 19 on the chip 14. Then the electromagnet generates a traction magnetic force larger than a magnetic force (self-holding force) operating between a permanent magnet 9 and the magnetic metal 17 of the chip 14 to grip the chip 14 by the positioning head 19. Then the chip 14 is moved by the three-axial positioning device 22 to a waveguide intersection part as a connection and switching destination and the mirror 4 formed on the chip 14 is inserted into a mirror insertion space gap 12 at the movement destination. At this time, in-core propagated light 10-1 at the movement destination after straight traveling is reflected by the mirror 4 to bend the optical path and thus switched. Similarly, the magnetic force operates between the permanent magnet 9 and the magnetic metal 17 of the chip 14 and even if the magnetic force of the electromagnet 20 is ceased, the chip 14 holds itself on a matrix plate 6.

Description

[Detailed Description of the Invention] (Field of Industrial Application) The present invention reflects light with mirrors installed in the middle of each optical path arranged in a matrix and switches the optical path to freely achieve low loss. This invention relates to a multi-terminal matrix optical switch that can be connected and switched, and has a self-holding function.

(Prior art) Figure 3 shows the document rNONBLOcKING 8 x 8
0PTICAL MATRIX SWITCH FOR
FIBER-OPTIC COMMUNICATION
N, ELIECTRONICS LETTERS, V
OL. 16, NO. 11, pp. 422-423 (1
980) is a plan view showing the basic configuration of a conventional matrix optical switch of this kind, such as that seen in 980) J. The optical switch shown in FIG. 3 fixes the relative position of the end of the optical fiber 1-1.1-2 and the lens l-1.1-2 to form beam collimators 3-1 and 3-2. The lens l
-1. 22 facing each other, and by moving the mirror 4 downward perpendicularly to the plane of the paper using a solenoid actuator, the light beam 5 is reflected/passed, and the light beam 5 is connected to an arbitrary optical fiber 1-. This is a matrix optical switch configured to connect and switch to 1,]-2.

By the way, optical MDP (Main Distribution)
When the switching frequency of an optical switch is as low as several times a year, such as when switching on a frame (on frame), it is desirable to be able to self-maintain the switched state without consuming power. Literature [Optical switch construction method, NTT Research Practical Application Report, Vol. 30, No. 11, p. There is a mirror self-holding mechanism as seen in 2789, 1981J. This self-holding mechanism consists of a solenoid actuator and a permanent magnet, and is provided for each switching/connection point by the product of the number of rows and the number of columns (3).

(Problems to be Solved by the Invention) However, the matrix optical switch having the configuration shown in FIG. 3 has the following problems, and it is difficult to reduce loss, increase the size and density, increase the number of terminals, and achieve self-retention.

In other words, (1) loss is caused by the axial deviation and angular deviation between the respective axes of the optical fiber 1-1 and the lens 2-1 that constitute the beam collimator 3-1, and the beam collimator 3-1 and the beam collimator 3-2.
It is determined by the axial deviation and angular deviation between the mirror 4 and the mirror 4, the manufacturing accuracy (angle, etc.) of the mirror 4 itself, the installation accuracy (angle, etc.) of the mirror 4, etc. With the current manufacturing technology for mechanisms and parts, even if a pentagonal prism that can absorb a considerable amount of the error in the angle of incidence of the light beam is used, there is a problem in that the loss becomes large. For example, the optical fiber 1-1 that constitutes the beam collimator 3-1.3-2
(1.-2) and lens 2-1 (2-2) Axis deviation between each axis 2 μm 1 Lens 2-1. Angular deviation between 2-2 10-
3 radians, and angular error of mirror 4 2 x 1.0
- Mechanical and component accuracy of 4 radians, a single mode fiber with a spot size of 5 μm and a focal length of 1
．． When constructing a parallel beam system using a 5 mm lens, the loss between the optical fibers 1-1, 1-2 is the diffraction loss Od
When calculated under the condition of B, there are cases where it becomes as large as about 10 dB. In order to reduce loss, it is extremely difficult to manually optimally adjust the axis misalignment and angular misalignment for all rows or columns, and as a result, there are many cases where the loss becomes large.

(2) The maximum optical path length resulting in diffraction loss OdB is determined by the focal length of the lens, and the light beam 5 expands three-dimensionally with the optical path length due to diffraction once it exceeds the beam waist. The optical path length is limited by the diameter of the light receiving lens and loss, and there is a limit to the number of terminals. For example, in the case of calculation example (1) above, using a spherical lens as the lens,
If the pitch between rows or columns is determined by the lens diameter, there is a problem that even if a loss of approximately ]OdH is allowed, the number of terminals of the matrix optical switch is limited to IOXIO.

(3) There is a limit to the ability to manufacture optical components and mechanical components 5 (self-holding mechanisms, etc.) that are arranged in a matrix in a small size and with high precision, and there is a problem that it is not possible to miniaturize and high-density optical switches. Particularly, when manufacturing mechanical components to reduce deterioration due to friction in order to ensure long-term loss reproducibility, there is a problem in that the dimensions become large. In addition, when using a solenoid actuator at each switching/connection point to move the mirror 4, if the pitch between rows or columns is reduced and the solenoid actuators are arranged in high density,
There was a problem in that magnetism leaked to the adjacent solenoid actuator, causing malfunctions in connection and switching, making it impossible to simultaneously achieve miniaturization and high density of the matrix optical switch and self-holding function.

(4) The required number of solenoid actuators and mirrors 4 are both the product of the number of rows and the number of columns. For example, even for a XIOO matrix optical switch with 100 terminals, the number of required parts is 10,000 each, and processing accuracy and assembly accuracy Problems include light loss due to variations in the amount of light and economic efficiency in terms of quantity. On the other hand, if we try to save money by introducing a mirror 1-axis moving mechanism that moves the mirror row-by-row or column-by-column, the dimensions of the mirror 1-axis moving mechanism will increase, and as a result, the pitch between rows or columns will increase. becomes large, making it impossible to achieve high density, and the mirror setting position accuracy varies, resulting in large variations in loss.
There is a problem in that the switching time becomes longer by at least the mirror movement time.

In view of the above-mentioned problems, an object of the present invention is to provide a low-loss, compact, high-density, multi-terminal, and self-holding multi-layer optical fiber that can be switched and connected between arbitrary optical fibers in a large-scale matrix by mirror reflection. 1) To provide an optical switch.

(Means for Solving the Problems) The matrix optical switch of the present invention includes waveguides formed in a lattice shape, gaps for mirror insertion spaces formed at the intersections of the waveguides, and positioning grooves or protrusions formed on the surface. , a matrix plate in which magnetic metals or permanent magnets are arranged in a matrix, positioning protrusions or grooves provided on both sides facing the grooves or protrusions, and an attractive force acts on the magnetic metals or permanent magnets of the matrix plate. A chip consisting of a permanent magnet or magnetic metal, a mirror that reflects the light propagating in the waveguide, and an electromagnet with sufficient traction force to counteract the attractive force and pull the chip away from the matrix plate, facing a positioning groove or protrusion on one side of the chip. A positioning head consisting of a positioning protrusion or groove for positioning, matching oil filled between core end faces facing each other at a waveguide intersection and a mirror part, and a device for positioning the positioning head.

(Function) Light changes from spatial propagation to waveguide propagation, and is always confined within the waveguide except for the cap for the mirror insertion space, which greatly eases the restriction on the optical path length and allows the optical switch to have multiple terminals. Furthermore, due to the matching oil, there is no Fresnel loss, and there is no axial or angular misalignment at waveguide intersections, resulting in low loss. In addition, by using semiconductor process technology, it is possible to form positioning grooves or protrusions between the matrix plate, the chip, and the positioning head with high precision.As a result, the chip, in other words, the mirror, is It can be set accurately and there is little variation in loss due to connection switching. Further, the number of chips only needs to be the smaller of the number of rows and the number of columns, while the number of electromagnets only needs to be one. In addition, the chip is attached to the matrix plate by magnetic force without any power consumption except when connecting or switching. On the other hand, at the time of connection/switching, there are no adjacent electromagnets, so there is little leakage of magnetism to adjacent chips, and only the chip in question can be attached without causing the adjacent chips to separate.

(Example) FIG. 1(a) is a cross-sectional view of an embodiment of the switching part, positioning head, and positioning device of the matrix optical switch of the present invention, and FIG. 1(b) is A of FIG. 1(a). -A'
FIG. FIG. 2 is a plan view in which the switching sections shown in FIG. 1 are arranged in a matrix. 1st
The figure is a cross-sectional view taken along line BB' in FIG. 2 plus a cross-sectional view of the positioning head and the positioning device. In FIG. 1, the matrix plate 6 includes a Si substrate 7, a Si substrate 7,
It consists of a waveguide plate 8 made on top and a permanent magnet 9 provided at the bottom of the Si substrate 7. As shown in FIG. 2, the matrix plate 6 has a core 10-1 in the center. 10-
The waveguide 11 is formed in a lattice shape with 2 and a mirror insertion space gap 12 (shown in FIG.
For example, a width of 10 μm) and a V-groove 13 are formed. However, core 10-1. 10-2 is formed using semiconductor process technology, and there is no axis shift or angle shift at the intersection of the waveguides 11. In addition, in FIG. 1, the chip 14 has a positioning protrusion l5 formed on one side thereof so as to face the groove l3 provided in the matrix plate 6.
, the chip l so that the attractive force from the groove 16 formed on the other side and the permanent magnet 9 provided on the matrix plate 6 acts.
The core 10-1 includes a magnetic metal 17 provided on both sides of the core 10-1, and a mirror 4 (with a width of 4 μm, for example) that reflects the light propagating within the core 10-1 at the mirror insertion space gap 12 at the waveguide intersection. At least the mirror insertion space gap l2 is filled with matching oil 18 as shown in FIG. In addition, in FIG. 1, the positioning head 19 consists of a permanent magnet 9 provided on the matrix plate 6 and a chip 1.
An electromagnet 20 having a traction force sufficient to separate the chip 14 from the matrix plate 6 against the shearing attraction force 10 and the magnetic metal 17 provided on both sides of the chip 14.
It consists of a positioning protrusion 21 that faces a positioning V-groove l6 provided on one side of the holder. Reference numeral 22 denotes a three-axis positioning device, to which a positioning head 19 is fixedly attached via an elastic body 23 that reduces positioning accuracy by about 10 μm. In FIG. 2, the stippled portion of the chip 14 is a magnetic metal l7 provided on one side and receives an attractive force from the permanent magnet 9. At this time, the groove 13 provided in the matrix plate 6 is formed in an X-shape, and the chip l4, that is, the mirror 4, can be accurately positioned three-dimensionally with respect to the matrix plate 6, that is, the waveguide l1. It has a structure.

On the other hand, the V-groove 16 provided on one side of the chip 14 is also formed in an X-shape with a cut in the center, and similarly, as shown in FIG. 1, the chip 14 and the positioning head l9 are accurately positioned. The structure is such that

Next, the switching operation of the light propagating in the waveguide with the above configuration will be explained. Assume that the initial state is, as shown in FIG. 2, in which the chip 14 is self-held by the permanent magnet 9 on the matrix board 6 without consuming power. At this time, the light propagating within the core lO-1 is transmitted to the mirror 4
This is a switching state in which the optical path is bent and the light enters the core IO-2. First, as shown in FIG. 1, the three-axis positioning device 22 positions the positioning rod 19 on the chip l4. Next, the electromagnet 20 generates a traction magnetic force that overcomes the magnetic force (self-holding force) acting between the permanent magnet 9 and the magnetic metal 17 of the chip 14, and
4 is gripped by the positioning head l9. At this time, the chip 4 and the positioning head 19 are placed in the V-groove l for positioning.
6 and the protrusion 2l, as if they were an integrated rigid body, and as a result, the positioning accuracy of the three-axis positioning device 22 becomes approximately equal to the positioning accuracy of the mirror 4. Next, the chip 14 is connected and connected using the 3-axis positioning device 22.
The mirror 4 formed on the chip 14 is moved to the waveguide intersection at the switching destination and inserted into the mirror insertion space gap 12 at the destination. At this time, the intra-core propagating light 10-1 that was traveling straight is reflected by the mirror 4, its optical path is bent, and the light is switched. At the same time, =12 A magnetic force (self-holding force) acts between the permanent magnet 9 and the magnetic metal 17 of the chip 14, and even if the magnetic force of the electromagnet 20 is cut off, the chip 14 is self-held by the matrix plate 6. .

In the above explanation, the case where the permanent magnets 9 are arranged in a matrix at each intersection of the waveguide 11 has been explained.
A single plate-shaped permanent magnet may be used below the Si substrate 7. Further, the permanent magnet 9 and the magnetic metal 17 may be replaced.

Next, consider the standard of magnetic force of permanent magnets and electromagnets.

As a guideline, the weight of a chip with a volume of 1 mm x 1 mm x 1 mm and made of Si is considered as a waveguide pitch of 1 + nm. However, since magnetic metal is also attached to the chip, the chip now has Si (specific weight 2.
3g/cm3) heavier magnetic metals (specific weight 8g/cm3)
), the approximate weight of the chip is 8 mg.

As mentioned above, the pitch of the waveguide is 1 mm or less, and the number of terminals is 10.
A matrix optical switch with a switching time of about 1 second can be constructed using about 0 XIOO. Additionally, due to advances in waveguide formation technology, the loss in waveguides has been reduced, and the loss of an optical switch with 100 XIOO terminals is approximately 3. It becomes 5dB or less.

(Effects of the Invention) As explained above, in the matrix optical switch of the present invention, light is confined within the waveguide, the restriction on the optical path length is reduced, and it is possible to have multiple terminals. Furthermore, there is no Fresnel loss due to the machining oil, and there is no axis shift or angular shift at the waveguide intersection, resulting in small loss. Furthermore, by using semiconductor process technology, it is possible to process parts with high precision, and no matter how many times the mirror is attached and detached, the mirror can be positioned three-dimensionally with high precision, and there is little variation in loss due to connection and switching. In addition, the mirror can be self-held to the matrix board by permanent magnets without consuming power except when connecting or switching. On the other hand, at the time of connection/switching, there are no adjacent electromagnets, and only the chip in question can be attached or detached without detaching the adjacent chip. As described above, the matrix optical switch of the present invention is capable of multi-terminal design, low loss, small size and high density, and self-retention, and also has high productivity and therefore can be made economical. .

[Brief explanation of drawings]

FIG. 1 is a cross-sectional view of an embodiment of the switching unit, positioning head and positioning device of the matrix optical switch of the present invention, FIG. 1(b) is a cross-sectional view at AA' in FIG. 1(a), This figure is a plan view showing the switching sections shown in FIG. 1 arranged in a matrix, and FIG. 3 is a plan view showing the basic configuration of a conventional matrix optical switch of this type. 4... Mirror 6... Matrix plate 7... Si substrate 8... Waveguide plate 9.
...Permanent magnet 10-1. 10-2
... Core 11 ... Waveguide l2 ... Gap for mirror insertion space 13. 16...V groove 14...chip 15. 21... Protrusion l7... Magnetic metal 18... Matching oil l9... Positioning head 20... Electromagnet 22... Positioning device 23... Elastic body 15-

Claims (1)

[Claims] 1. A self-maintaining matrix optical switch that switches the optical path by reflecting light with a mirror placed in the middle of each optical path arranged in a matrix, which comprises a waveguide formed in a lattice shape, a guide A gap for mirror insertion space formed at the intersection of the wave paths, a positioning groove or protrusion formed on one side, a matrix plate with magnetic metal or a permanent magnet on the other side, and the groove or the groove formed on both sides. A positioning protrusion or groove facing the protrusion, a permanent magnet or magnetic metal that exerts an attractive force on the magnetic metal or permanent magnet of the matrix plate,
A chip consisting of a mirror that reflects light propagating within the waveguide;
An electromagnet with sufficient traction force to pull the chip away from the matrix plate against the attractive force, a positioning head consisting of a positioning groove or protrusion on one side of the chip and an opposing positioning protrusion or groove facing each other at the waveguide intersection. A matrix optical switch characterized by having a matching oil filled between the core end faces and a mirror part, and a device for positioning a positioning head.