FR2479993A1 - Mechanically operated switch for fibre=optic transmission lines - has movable plate on which are mounted intermediate fibres which connect incident and outgoing fibres between two positions - Google Patents

Abstract

The switch selectively connects at least one incident optical fibre (f(e)) with one of a pair of outgoing fibres (f(S1),f(S2)). The device has two support elements (2,3) which are fixed to each other by a plane surface on which are disposed the incident optical fibre and the pair of outgoing fibres. A movable plate (4) is placed between the two support elements (2,3), and has two predetermined stable positions (I,II). The plate supports intermediate fibres (f(i1),f(i2)) of which one (f(i1)) provides the coupling between the incident and outgoing fibre when the plate is in one of its stable positions. In its other position, the plate connects the two fibres (f(S2),f(e)) via the other (f(i2)) of the intermediate fibres. The movable plate (4) may be displaced either manually or by electromagnetic devices.

Description

 The present invention relates to mechanically controlled optical switches used for fiber optic links.

A wide variety of optical switches employing electro-optical or magneto-optical effects are known. However, when the switching speed is not a paramount parameter, the mechanically operated switches, although having higher switching times (several milliseconds), have significant advantages:
- low insertion loss;
- low crosstalk between adjacent channels.

 The invention belongs to this category of switches. The term "switch" refers to devices that selectively couple an incident fiber to a fiber selected from two or more emerging fibers, or matrix switching of n × n fibers.

The terms "incident" and "emergent" are in no way limiting to a particular method of transmission and do not indicate, in particular, that the transmissions are unidirectional.

The switches of the prior art can be divided according to two approaches: the first approach involves the movement of fibers and the second approach involves the displacement of glass prisms with the use of intermediate optical elements. The first and second approaches can be illustrated, by way of non-limiting example, by the following articles:
- first approach: the article by HALE and KOMPFNER: "Mechanical
Optical Fiber Switch ", published in the US Journal: Electronics Letters, July 22, 1976, Vol. 12, No. 5, page 388
- second approach: - the communication of AOYAMA: "High Reliability Optical Switch for Optical Fiber Transmission", published in the accounts of: "Fourth European Conference on Optical Communication", Geneva, 12-15 September 1978, pages 383-391, an approach illustrated in particular by FIGS. 4 and 8 and the communication of FUJII and the optical switching network for optimal fiber communication; published in the reports of: "Fifth European
Conference on optical communication ", Amsterdam, 17-19 September 1979, pages 13.3-1 to 13.3-4.-
With regard to the second approach, in addition to increasing the complexity due to the use of intermediate optical elements, this introduction of optical elements also increases the insertion losses. The first approach, for its part, leads to simpler design devices, but requires a rigorous positioning of the fibers relative to each other and in particular the fiber made mobile ensuring switching. In the opposite case, there is appearance of coupling loss similar to the losses due to the misalignment of two fibers to be connected with respect to one another. In the context of optical links, the switches being blessed in large numbers along a link, these losses quickly become prohibitive.

 The invention on the contrary proposes a switch of simple design, allowing a rigorous positioning of the fibers relative to each other and not requiring the introduction of intermediate optical elements other than fibers. It allows simple switching of one fiber to two fibers or matrix switching from a 2 x 2 basic cell.

 An interesting application of the invention is the production of reliable optoelectronic devices that are self-repairing by switching redundant elements by means of a switch of the invention.

 The subject of the invention is therefore a mechanically controlled optical switch for establishing selective optical couplings between at least one incident optical fiber and one of the fibers of an emerging pair of optical fibers, a switch mainly characterized in that it comprises first and second support elements, fixed relative to each other, each having at least one plane surface on which the incident optical fiber and the pair of emerging optical fibers are respectively arranged, the surfaces being coplanar, and a movable plate, disposed between the first and second fixed support members, capable of taking first and second predetermined positions and supporting intermediate optical fibers, a first intermediate optical fiber providing an optical coupling between the incident optical fiber and one of the two emerging optical fibers when the movable platen is in the first predetermined position, and a seco an intermediate fiber providing optical coupling between the incident optical fiber and the other emerging optical fiber when the movable platen is in the second predetermined position.

 The invention also relates to a self-repairing optoelectronic device implementing such a switch.

The invention will be better understood and other advantages will appear on reading the description which follows, with reference to the appended figures:
FIG. 1 illustrates an optical switch according to a first approach of the prior art;
- Figures 2 to 5 illustrate optical switches according to a second approach of the prior art;
FIGS. 6 and 7 schematically illustrate an exemplary embodiment of an optical switch of the invention according to a first approach;
FIG. 8 illustrates a second variant embodiment of this optical switch;
- Figures 9 and 10 illustrate - details of implementation of these optical switches;
FIG. 11 illustrates a first concrete embodiment of an optical switch according to the invention;
FIG. 12 illustrates a third embodiment of an optical switch according to the invention;
FIG. 13 illustrates an optoelectronic device implementing a switch of the invention;
- Figures 14 and 15 illustrate a dual optical switch made according to the invention and its application to optical links.

FIG. 1 illustrates an embodiment of an optical switch according to a first approach of the prior art. The switch comprises a body or casing, consisting of a glass tube 1 of square internal section 10, so that output or emerging optical fibers fs1 and fs2 are positioned along two edges, 11 and 12 respectively, and that an input or incident optical fiber fe can move between two positions
I and 11 so as to switch between the fibers f51 and fs2 The fibers fs f51 and f52 are threaded into protective sheaths, 13, 14 and 15 respectively, and fixed to the ends of the tube 1 with the aid of a cement or sealing glue 16, for example. For convenience of explanation, the fiber fe is said input or incident fiber and the fibers i and fs2 outgoing or emerging fibers. This does not imply that the links are unidirectional. The device is rigorously symmetrical as to the direction of propagation of the light in the fibers.

 To ensure switching, the fiber f e is provided with an additional sheath 17, nickel for example. If this sheath 17 is subjected to a magnetic field produced by an electromagnet (not shown), it is possible, depending on the direction of the applied field, to draw the fiber fe towards the edge II (position I) or towards the edge 12 ( position II). It follows that the transmissions take place between the fibers fe and fs1 (position I) or between the fibers f e and f52 (position II).

 Being entirely floating inside the tube, it seems difficult to very rigorously position the fiber f l -long respective edges 11 and 12, depending on whether it is in position I or II.

 An example of a switch according to this approach is described in the article by HALE and KOMPFNER supra.

 A switch according to a second known art approach is illustrated in FIG. 2. According to this approach, the switches between the positions I and II are obtained by the use of an intermediate prism P. It is then necessary to implement additional optical elements constituted by additional lenses Le, L51 and Ls2, respectively collimating and focusing to ensure the interface between the prism Pi and an input fiber fe and with the same prism Pi and the output fibers fs1 and f52. . Depending on whether the prism Pi is in position I or II, the emerging beam of the lens Le is directed on the lenses Lsl or Ls2.

 This approach allows the realization of matrix type optical switches. An example of a 4x4 matrix optical switch is illustrated in FIG. 3. This switch allows matrix-type selective links between four input fibers f1-f and four output fibers fs1-f54. It comprises as previously collimating lenses Lel to Le4 and focusing lenses Lsl to L54. Between these two sets of lenses are arranged switching elements C1 to C6 each comprising a prism P'i similar to the prism Pi of Figure 2, and can move between two positions I and II, as shown in Figures 4 and 5 respectively. These prisms have a hexagonal shape and cause a cross transmission of two parallel beams respectively a and b to c and d in position i (Figure 4) or direct transmissions a towards c, b to d, in position II (Figure 5).

 This second approach of the known art leads to more complex optical design switches and, in addition, increases the insertion losses of intermediate optical elements.

Such an approach is described for example in the two aforementioned communications: the communication of AOYAMA on the one hand and the communication of FUJII et al on the other hand
The invention, on the contrary, allows the realization of optical switches which do not require the use of intermediate optical elements but in which the optical transmission losses are limited as much as possible by a rigorous positioning of the fibers relative to each other. other.

 Figures 6 and 7 schematically illustrate a first example of optical switch execution according to the invention. The switch of the invention, constituting in the specific example of FIGS. 6 and 7 a 2 x 2 switching cell, comprises three parts: two fixed parts or supports 2 and 3 and a moving part or plate 4. The fixed parts 2 and 3 are integral respectively the two input fibers fel and fe2 and the two output fibers f51 and fs2> being understood that the optical switch is fully symmetrical, these names being adopted only for the convenience of the description Of the movable platen forming a drawer in a reference plane XY, four intermediate fibers are attached to each other. The movable plate 4 can move between two positions, the position I illustrated by FIG. 6 providing via the fibers yarn and fi2 direct connections between the input fiber pair fel and f and the pair of output fibers f51 and fus2 ; and the position II, illustrated in FIG. 7, according to which the transmissions between these two pairs of incoming and outgoing fibers are crossed, via intermediate fibers fi3 and fui4.

All the fibers are positioned in the same plane, for example the XY reference plane. The input fibers + and fe2 and the output fibers f51 and fs2 are fixed and parallel to each other, the respective axes of symmetry of the pairs of fibers fel and sl fe2 and f52 being merged. In reality, the two fixed substrates 2 and 3 can be merged into a single common substrate in which a channel has been dug for the movable tray 4 forming a drawer. It is therefore easy under these conditions to rigorously position these four fibers. On the other hand, the four intermediate fibers fi4 being also in the same plane as the input and output fibers and the plate moving in one direction (for example). example
X) in the reference plane XY, the only difficulty to be overcome in order to obtain a rigorous positioning of the intermediate fibers with respect to the input fibers and the output fibers is to define precisely the two extreme positions of the moving plate 4. This can be achieved by using stops integral fixed parts 2-3 of the preferably adjustable switch and shown in Figures 6 and 7 by screws 5 and 6.

 Index-type optical fibers of the "telecommunication" type have a standardized outer diameter of the order of 122 μm. The theroric displacement to be performed by the mobile plate 4 is therefore 125 .mu.m. In practice, except in an exemplary embodiment which will be described later in connection with FIG. 10, the fibers can not be joined. It follows that the displacement to be expected is greater than or equal to about 1 millimeter. To reduce the amplitude of the mechanical displacement, and therefore the switching time, the configuration illustrated in FIG. 8 can be adopted. In this figure, only the mobile plate 4 has been illustrated, the remaining portion of the optical switch constituted by the fixed supports 2 and 3, being identical to that of FIGS. 6 and 7. As illustrated in FIG. 8, the two pairs of intermediate fibers fil-fi2 and fi3-fi4 are interlaced.

the amplitude of the displacement is thus reduced by half.

 The very precise positioning of the fibers in the XY reference plane can be very advantageously achieved by using, as support material, monocrystalline silicon in which V-shaped positioning grooves are etched by chemical etching.

 Another possible positioning support is a copper base in which are also formed V-shaped grooves. In order to obtain the flanks of these planar grooves and having a very precise inclination, it is advantageous to use the method described in the application. patent filed May 18, 19782 as No. 78 14 763 and published in No. 2,426,271.

According to this method, the grooves reproduce in negative a dihedral -forsigned by etching monocrystalline silicon, the attack being effected preferably along the axes of the crystal.

 On the other hand, to achieve the crossing of the fibers A.3 and fi, it is necessary to provide a recessed impression in the plate 4. To do this, the method described in the French patent application filed on March 18 Mal 1978, under A3 78 14 762 and published under No. 2,426,347.

An imprint, as well as the V-grooves required for the fiber housing can be obtained simultaneously in a semiconductor block by stamping.

 Figure 9 illustrates a fragment of the tray highlighting the grooves 41 to 44 for the positioning of intermediate fibers fi3 and fi4 and a footprint 40 in which these fibers intersect.

The positioning of the fibers can also be obtained by using U-shaped supports. FIG. 10 more particularly illustrates the mobile plate 4 constituted by a U-shaped support, ie a support in which a channel has been hollowed out. flat-bottomed and vertical-sided
The intermediate fibers fi4 are positioned on the bottom of the U. The distance I between the two edges of the support 4 must be equal to the sum of the outer diameters of the optical fibers. A cover 45 made of elastic material makes it possible to hold the optical fibers on the bottom of the U, the fibers fi 3 and fi 4 being able to cross between the two ends of the mobile plate 4, the extra thickness resulting from this crossing being absorbed by the elasticity of the material forming the cover 45. The fixed portions of the optical switch are also constituted by U-shaped supports. According to this configuration, the turns being contiguous, a movement of less amplitude is necessary to move from the position I to the position II.

 The displacement can be carried out manually or preferably, using conventional electromechanical means.

 FIG. 11 illustrates an example of a concrete embodiment of an optical switch according to the invention controlled by an electromagnet 7 and a return spring 8. In this figure the elements common to the preceding figures bear the same references and will not be described. again. The electromagnet 7 acts by means of a flexible lever 70 pivoting about an axis 71 on the movable plate 4. This plate, as previously described, moves between two stops, (not shown in Fig. 11) defining the positions l-and II. In Figure 11, the plate is shown in position II. The return to position I is ensured by means of the return spring 8. The active parts of the optical switch of the invention are enclosed in a protective box 9 in which partial cuts have been made to highlight the non-transparent elements. visible. The housing assembly 9 and the control electromagnet 7 are integral with a common support S, for example a printed circuit board supporting several 2 × 2 matrix optical switches identical to that shown in FIG. 11, for the embodiment for example, 2n x 2n higher order switching matrix, unrepresented metallized strips capable of carrying the electrical signals for controlling the electromagnets associated with each of the optical switches.

 The invention is not limited to configurations of moving plate moving parallel to the XY plane parallel to the plane of the fibers. FIG. 12 illustrates an embodiment of an optical switch according to the invention in which the mobile part 4 carrying the fibers intermediate moves in a Z direction, orthogonal to the XY plane. On the upper part 45 of the movable plate 4 are arranged the pairs of intermediate fibers, respectively for the direct transmissions fil-fi2 and for cross transmissions fi3-fi4, on the underside and top faces of the upper part 45 of the plateau. 4, the movable plate 4 is provided with at least two stops, upper 46 and lower 47, possibly associated with adjustment means to define the two positions I and II.

To illustrate the invention, a 2 x 2 matrix switching cell has been realized and its main characteristics and performances are summarized in the table below:
- insertion losses: <0.5 dB;
- repeatability of the couplings: <0.05 dB
- switching time: <20 ms
(control by electromagnet)
- inter-channel crosstalk: <-60 dB adjacent.

 A first application of the invention is the realization of 2n x 2n switching matrices, simply by association of 2 x 2 basic cells, as described in relation with one of FIGS. 5 to 12. The configuration of a such matrix can be deduced simply from the architecture of the matrix ss x 4 described in relation with FIG. 3 and can be generalized to any dimension 2n x 2n.

 Another interesting application of the invention is the production of self-repairing optoelectronic devices for high reliability optical fiber links. A first example of a device is described with reference to FIG. 13.

 The device of FIG. 13 is a laser source with redundant elements. It finds its use, for example, in the context of underwater transmissions installed permanently in places by definition inaccessible. The proposed source that can be implemented in a submarine repeater, essentially comprises a switch according to the invention in which there are the fixed supports 2 and 3 and the movable plate 4. The fixed support 3 comprises two laser sources Lal and La2, respectively disposed at the bottom of cavities 31 and 32. This part further comprises two optical fibers f51 and f52 each provided with an optical coupling with the semiconductor laser sources, respectively L1 and L2.

The mobile plate 4 comprises two intermediate optical fibers fil and fi2, the latter being disposed in the bottom of the cavity 40. The fixed part 2, which supports an incident optical fiber fes is coupled to the laser Lal, via the intermediate optical fibers wire and the optical fiber a In case of failure of the laser source Lal, an automatic switching system (not shown), for example an electromagnet of a type similar to that described in relation to FIG. 11, switches the moving plate 4 from position I to position II. It follows that the optical fiber fe is then coupled via the second intermediate optical fiber f12 and the optical fiber fs2 to the reserve laser source La2. The defective operation detection of the laser Lal and the switching control can be performed by means of local or centralized detection and control means. In the second case, it must be possible to convey control signals via auxiliary link circuits or to be able to send service messages via the usual links. These techniques are known to those skilled in the art.

 The optical couplings between the laser sources Lal and La2 and the optical fibers 51 and 52 can be produced according to the method described in the patent application No. 79 22 886, filed on September 13, 1979. The device described in this patent application comprises a plankonvex lens adhered to the input face of an optical fiber consisting of a glass of refractive index greater than that of the optical fiber and obtained by contacting the end of the optical fiber with a drop of glass in fusion. A spherical cap whose parameters are controllable is then obtained. In the context of the present invention, the optical fiber in question constitutes one of the two optical fibers + or f52 and the spherical cap constitutes one of the two lenses L1 and L2.

Another example of application of the invention is described with reference to FIGS. 14 and 15. In the context of data transmission along a loop, these transmissions being provided by optical fibers to which terminals of Tx.The connection to the loop comprising the optical fibers fox 1 and f is provided via the input optical fiber f, the receiver Rx, the optical input coupling fiber f ex 9 on the one hand, and the optical coupling fiber fsxS output the Ex transmitter and the output optical fiber f 'on the other hand. When the terminal Tx is in faulty operation, a double switch K can be isolated by shortcircuiting it purely and simply using an optical fiber provided for this purpose fcc. Since a large number of terminals can be arranged along the loop and consequently the same number of optical switches K being arranged in series on this loop, it is particularly important that these optical switches introduce only very low losses as well low cross-talk between transmission paths. The optical switch of the invention, whose configuration for this particular application is shown in FIG. 13, responds particularly well to these requirements. The switch control of the position 1, or normal regime in which the terminal Tx is coupled to the transmission loop, at position II or faulty operating regime in which the terminal
Tx is isolated can be done by a remote control or by local fault detection.

 The invention is not limited to the exemplary embodiments which have been described by way of illustration, in particular to the only applications described with reference to FIGS. 13 to 15. In particular without changing anything in the architecture of FIG. 13, the device can be used to couple an optical fiber selectively and at will, to one of two Lal or La2 laser sources of different wavelengths.

Claims (10)

 1. Mechanically controlled optical switch for establishing selective optical couplings between at least one incident optical fiber and one of the optical fibers of a pair of emerging fibers (f51, fs2), characterized in that it comprises first (2) and second (3) support members, fixed relative to each other, each having at least one planar surface on which the incident optical fiber (fe) and the emerging pair of optical fibers (f51, fs2) respectively are arranged , the surfaces being coplanar, and a movable platen (4), disposed between the first and second fixed support members (2, 3), which can take predetermined first and second stable positions (I, II) and supporting intermediate optical fibers (fi1fi2 ) a first intermediate optical fiber (wire) providing optical coupling between the incident optical fiber and one of the two emerging optical fibers (fs1) when the movable platen (4) is in the first pos predetermined ion (I) and a second intermediate fiber (fi2) providing optical coupling between the incident optical fiber (fe) and the other emerging optical fiber (fs2) when the movable platen (4) is in the second predetermined position (II). ).  2. Switch according to claim 1, characterized in that it comprises first and second incident optical fibers (fe1 fe2) and first and second emerging optical fibers (f51, f52) and in that the movable plate (4) comprises first and second pairs of intermediate optical fibers <e14e2 'fe3-fe4), the first pair of intermediate optical fibers (fil-fi2) establishing direct optical couplings respectively between the first incident optical fiber (fel) and the first emerging optical fiber (fs1)> and between the second incident optical fiber (fe2) and the second emergent optical fiber (fol) when the movable platen (4) is in the first predetermined position (I) and the second pair of intermediate optical fibers (fi3- fi4) establishing optical couplings crossed between these same fibers when the movable platen (4) is in the second predetermined position (II).  3. Switch according to any one of claims 1 or 2, characterized in that the movable plate (4) moves in a direction parallel to the plane (XY) of the optical fibers.  4. Switch according to any one of claims 1 to 3, characterized in that the first and second fixed support members (2, 3) and the movable plate (4) are made of cold-malleable metal in which grooves are made straight lines 41-44 in V resulting from the permanent deformation by crushing of the metal, grooves in which the incident optical fibers (fel fe2)> the emergent optical fibers (fs1, f52) and the intermediate optical fibers (fil-fi4) are arranged respectively  5. Switch according to any one of claims 1 to 3, characterized in that the first and second fixed support members (2, 3) and the movable platen (4) are made of semiconductor material in which rectilinear grooves are made. 41-44 in V, grooves in which the incident optical fibers are respectively arranged (irons e2 'emerging optical fibers (fols2 fs2) and intermediate optical fibers (fi3-f4)  6.Switch according to any one of claims 1 to 5, characterized in that the movable plate (4) further comprises at least one cavity-shaped bowl (4û) in which can intersect at least two intermediate optical fibers (fi3 , fi4).  7. Switch according to any one of claims 1 to 3, characterized in that the first and second fixed support members (2, 3) and the movable plate (4) each consist of a piece of rigid material in which has been formed a flat-bottomed channel with vertical walls on the bottom of which adjacent optical fibers (fe1, fe2), emerging optical fibers (f1 'fs2) and intermediate optical fibers- (fil-fi4) are respectively disposed adjacent to each other; ); the distance (1) separating the two vertical walls of a channel being equal to the sum of the diameters of the optical fibers arranged on the bottom of the channel and a cover (45) of elastic material holding these optical fibers on the bottom of the channel.  8. Switch according to any one of claims 1 to 7, characterized in that it comprises electromechanical drive means of the movable plate and in that these means comprise an electromagnet (7) exerting a thrust by the intermediate a lever (70) on the movable plate (4) when it is excited so as to impose a translation of the movable plate (4) of the first (1) and the second (II) predetermined stable position and a spring biasing (8) causing the reverse translation when the electromagnet (7) ceases to be excited; mechanical stops (5, 6) further defining the first and second predetermined positions (1, II) of the movable platen (4).  A self-repairing electro-optical device comprising at least one incident optical fiber (fe) for coupling the device to an optical fiber data transmission system; characterized in that it comprises a first optoelectronic element (La1) a second optoelectronic element (La2) each associated with an emerging optical fiber (f51, fs2) and at least one optical switch according to any one of claims 1 to 8 , associated with means for detecting the correct or defective operation of the first optoelectronic element (Lal) coupled to mechanical control means acting on the mobile plate (4) so as to couple the first optoelectronic element (Lal) to the incident optical fiber (fe) on detecting a faulty operation of the first optoelectronic element, the movable platen (4) supporting a first and a second intermediate optical fiber (thread fi2).  10. Device according to claim 9, characterized in that the first and second optoelectronic elements are semiconductor laser sources (Lal, La2).