Optical transceiver with a pluggable function

[0001] The present application claims the benefit of the provisional application Ser. No. 60/615,587, filed Oct. 5, 2004; the provisional application Ser. No. 60/620,325, filed Oct. 21, 2004; the provisional application Ser. No. 60/632,937, filed Dec. 6, 2004; and the provisional application Ser. No. 60/652,846, filed Feb. 15, 2005.

[0002] 1. Field of the Invention

[0003] The present invention relates to a structure of a pluggable optical transceiver, and a method for manufacturing the same.

[0004] 2. Related Prior Art

[0005] The pluggable transceiver 1 (hereinafter denotes as transceiver) shown in FIG. 1A is used by being inserted into the cage 2 provided on the circuit board of the host system. The transceiver 1 inserted into the cage 2 is shown in FIG. 1B. On the bottom of the cage 2 is formed with stud pins 2c, and the cage 2 is installed on the circuit board of the host system by soldering these stud pins 2c to the circuit board so as to expose the opened end 2a from the face panel of the host system. In the inner end of the cage 2 is provided with an electrical connector 2b such that the each electrode of the connector 2b is in contact to, or soldered to the wiring pattern on the circuit board. By inserting the transceiver 1 into this cage 2, the electrical plug configured in the rear end of the transceiver 1, mates with this connector 2b within the cage to enable the transceiver 1 to communicate with the host system. In FIG. 1B, the host system, the circuit board and the face panel thereof, are omitted. Moreover, to configure the electrode of the plug makes the transceiver 1 so-called hot-pluggable, by which the transceiver 1 is able to insert into/remove from the cage 2 as the host system is powered on.

[0006] The cage 2, as shown in FIG. 1B, is a metal box with one end 2a being opened. Although a tip of the tab plate protrudes from the cover 23 of the transceiver 1, when the transceiver 1 is set within the cage 2, this tab plate 14 is also set within the cage 2 as the tip thereof coming in contact to the inside of the cage 2. Moreover, a plurality of fins 2d extends outward in the front portion thereof to surround the opened end 2a. When the cage 2 is installed on the circuit board of the host system, these fins 2a come in electrically contact to the face panel to shield the transceiver 1 and the host system.

[0007] Some multi-source agreements (MSA) rule the outer dimension and the electrical specification for such pluggable transceiver. The normal operation can be ensured as long as the transceiver follows the agreements, even when such transceiver is provided from the extraordinary manufacturer. Further, since the agreement rules the outer dimension, the insertion/extraction of the transceiver is also ensured.

[0008] However, the cage itself is made of a thin metal plate to deform in ease by an extra force, even by the insertion/extraction of the transceiver to the cage. The tab plate is necessary to come in reliable in contact to such deformed cage. The tab plate is provided to secure the EMI (Electro-Magnetic Interference) shield function for the transceiver 1. Recently, the operational speed, i.e., the clock speed of the transceiver reaches and exceeds 1 GHz. In such high speed operation, the high frequency signal is firmly prevented from leaking from the equipment. The tab plate is necessary to come in contact to the physically deformed cage 2.

[0009] Another subject under such high operational speed is that, the active devices, such as ICs and transistors, installed within the transceiver becomes necessary to be provided with large current to operate in a high frequency band, which results in the further heat being generated in the devices and the request to provide a further effective mechanism for the heat dissipation. However, the transceiver is unable to touch the case thereof in tight to the cage because the pluggable transceiver is inherently assumed to be inserted into/extracted from the cage 2. Tightly and firmly touching to the cage disturbs the smooth insertion/extraction of the transceiver.

[0010] Further, the transceiver comprises of three parts, namely, subassemblies to covert an optical signal to an electrical signal, or an electrical signal to an optical signal, a body portion installing the electronic circuit and an optical receptacle, and an mechanism to engage the transceiver with the cage. The U.S. Pat. No. 6,349,918 has disclosed the engaging mechanism using a bail and an actuator. Rotating the bail in front of the optical receptacle, one end of the bail pushes up the actuator, accordingly, the other end of the actuator is pressed down to disengage with the cage because the actuator moves in the seesaw motion. The other United States patent published as 2003/0142917A has also disclosed a similar mechanism.

[0011] Recently, one application has been proposed that, a plurality of transceivers is arranged in dense to configure a hub system for the optical communication. For example, to arrange the transceiver, the cross section of which is about 1 centimeter square, by 16 pieces in horizontal and two to four in vertical forms the hub system providing 32 to 64 optical channels. In such system, when the target transceiver is to be removed from the cage as the other cages surrounding the target transceiver are set with transceivers each providing an optical cable, the bail of the target transceiver is hard to rotate because the neighbor transceivers and their optical cables prevent from touching thereto, which prohibit the target transceiver from being extracted from the cage.

[0012] Therefore, the present invention is to provide a new structure of the optical pluggable transceiver based on the MSA and a method for manufacturing the same. The transceiver according to the present invention, not only satisfies the MSA standard, but also ensures the reliable shielding by stably touching to the cage, provides the effective heat dissipation mechanism from the devices to the outside thereof, and provides a disengaging mechanism with the cage even a plurality of transceivers with the optical cable is densely arranged.

[0013] One aspect of the present invention relates to a structure of a pluggable optical transceiver that is inserted into or extracted from the cage of the host system. The cage exposes one end portion thereof from the face panel of the host system to receive the pluggable transceiver, and provides an electrical connector in the other end. The connector is mounted on the circuit board of the host system.

[0014] The optical transceiver of the present invention comprises an optical assembly unit and a body unit. The former unit includes a receptacle member, a tab plate and an optical sub assembly. On the other hand, the latter unit includes a substrate, a base, a heat conducting plate, and a cover. The receptacle member may be made of resin and configured with an optical receptacle for receiving an optical connector with an optical fiber. The tab plate is made of metal and engaged with the receptacle member. The optical subassembly, which may include a transmitting optical subassembly (TOSA) and a receiving optical subassembly (ROSA) when the transceiver performs a full duplex communication, includes a semiconductor optical device that is optically coupled with the optical fiber within the optical connector. This optical subassembly is installed on the receptacle member.

[0015] The substrate mounts an electronic circuit and mates with the electronic connector within the cage. The base, made of metal, installs the substrate and is engaged with the optical assembly unit. The heat conducting plate, also made of metal, is in contact to the electronic circuit mounted on the substrate. This heat conducting plate is engaged with the base. The cover, also made of metal, puts the base, the heat conducting plate, the optical subassembly, and the base therein.

[0016] One feature of the transceiver according to the present invention is that these metal tab plate, the metal base, the metal heat conducting plate, and the metal cover are, although the materials are different to each other, all made from a metal plate by cutting, bending, and tapping without welding and gluing. Due to this configuration, not only the material cost of the transceiver may be reduced but also the manufacturing process thereof may be simplified, thereby reducing the cost thereof.

[0017] The receptacle member of the invention may include first to third portions. The first portion is configured to form the optical receptacle. The second portion forms a first projection in an outer side thereof. The third portion forms a second projection in an outer side thereof. On the other hand, the tab plate may include first and second portions. The first portion of the tab plate forms a latch tab engaged with the first projection of the receptacle member. The second portion, folded twice to form a U-shaped cross section, is received within an opening of this U-shape of the second portion of the tab plate.

[0018] Thus, the tab plate may be assembled with the receptacle member only by inserting the third portion of the receptacle member in to the second portion of the tab plate and by engaging the latch tab of the tab plate with the first projection of the receptacle member. Therefore, the manufacturing cost of the transceiver may be reduced.

[0019] The base may include a bottom, a pair of sides extending from the base, and a center post also extending from an end of the bottom. On the other hand, the receptacle member may include a pair of sides forming a second projection and a center partition for forming a cavity to install the optical subassembly therein. The receptacle member may further include an opening behind the center partition. This opening receives the center post of the base.

[0020] The side of the base may include a support with an opening extending from the side. This opening engages with the second projection of the receptacle member. Thus the base and the receptacle member of the present invention may be assembled to each other by inserting the center post of the base into the opening of the receptacle member and by engaging the second projection of the receptacle member with the opening of the base, without any screwing. Therefore, the manufacturing cost of the transceiver may be reduced. Moreover, the engagement of the base with the receptacle member may be carried out in compensate, that is, the former engagement of the center post to the opening is between the convex in the base and the concave in the receptacle member, while, the latter engagement between the projection in the receptacle member and the opening in the base is between the concave in the base and the convex in the receptacle member, which may strengthen the engagement.

[0021] The base of the present invent may include a first portion for mounting the substrate and a second portion for engaging with the heat conducting plate. The second portion may include a pair of coupling surface with a slit therebetween. Each coupling surface has firs and second fitting surface bent downward along a longitudinal direction of the transceiver. On the other hand, the heat conducting plate may include first to fourth portions. The first portion is in contact to the electronic circuit on the substrate. The second portion is in contact to the coupling surfaces of the base. The third and fourth portions of the heat conducting plate each includes a surface fitted to the first and second fitting surface of the coupling portion.

[0022] The heat conducting plate may further include, in the second portion thereof, a pair of fitting tabs and a spacer tab both bent downward. The spacer tab has a width slightly larger than that of the slit between the coupling portions of the base. On the other hand, the base may further include, in the coupling surface thereof, a pair of tabs each bent downward. These tabs are in contact to the fitting tab of the heat conducting plate. Thus, the coupling surface of the base is put between the spacer tab and the fitting tab of the heat conducting plate. That is, the spacer tab presses the coupling surface of the base outward, while, the fitting tabs abut against the tab of the coupling surface to press inward. Thus, the base and the heat conducting plate of the present invention is assembled only by fitting to each other without no screwing, which may simplify the manufacturing process and reduce the cost thereof.

[0023] The optical transceiver of the present invention may further include a supplementary substrate to mount a supplementary circuit. The supplementary substrate, electrically connected to the substrate with a flexible printed circuit board, is mounted on the heat conducting plate so as to fold the flexible printed circuit board. The flexible printed circuit board provides a projection that is set behind the coupling portion of the base. Since this configuration hides the edge of the flexible printed circuit board by the base, the cover does not hook the board when the cover is set to the body unit.

[0024] The optical transceiver may further include a supporting plate to position the supplementary substrate. That is, the heat conducting plate may further include, in the first portion thereof, a supporting post bent upward. On the other hand, the supporting plate may include first to fourth portions. The first portion is inserted under the heat conducting plate and has a pair of legs each having a step inserted into the notch of the supplementary substrate. The second portion includes a surface abutted against the side of the supporting post. The fourth portion has a pair of legs each having a step engaged with the step formed in the and of the supplementary substrate. The third portion that connects the second portion with the fourth portion is configured with an arched shape protruding outward. When the cover is set to body unit, the cover presses the third portion of the supporting plate, accordingly, the supporting plate presses down the supplementary substrate. Thus, the supporting plate of the present invention may be assembled with the supplementary substrate and the heat conducting plate only by fitting without any screwing, which may also reduce the manufacturing cost.

[0025] The optical transceiver of the invention may further include a bail and an actuator for providing the releasing mechanism of the transceiver from the cage. The optical receptacle member may further provide structures to assemble these bail and actuator. That is, the receptacle member may provide third projections in both outer sides thereof, and a center projection and a pair of beaked hooks in the bottom surface thereof. The bail is configured with a pair of legs each having an opening for receiving the third projection and a bridge portion connecting these legs. The actuator has a hollow for receiving the center projection, an axis that is fitted into the beaked hook, and an engaged projection in a tip thereof. To rotate the bail with the third projection as a center of the rotation causes a seesaw motion of the actuator around the axis thereof, which draws the engaged projection inward to release the optical transceiver from the case. Thus, the present transceiver assembles with the bail and the actuator only by fitting, which reduces the manufacturing cost.

[0026] Moreover, the actuator of the present transceiver includes a pressing surface in the end opposite to the side where the engaged projection is formed in the tip thereof. To press the pressing surface of the actuator causes a rotational moment to the bail, which enables to release the optical transceiver from the cage.

[0027] Still further, the bridge portion of the bail may provide a color code to identify an emission wavelength of the transceiver. When the transceiver is applied to the wavelength division multiplex (WDM) system, the emission wavelength is necessary to identify in precise. However, when the transceiver is set within the cage, the information of the emission wavelength is unable to distinguish even such information is recorded in the label stuck in the top and side surfaces of the transceiver. In the configuration that the emission wavelength is identified by the color code marked in the bridge portion of the bail, the information may be easily distinguished.

[0028] The base of the present invention, in the bottom thereof, may include a rib extending along a longitudinal direction and hollowed inward. On the other hand, the cover inevitably includes a slit extending along the longitudinal direction because the cover is made from a metal plate. In the present invention, a pair of edges forming the slit is each set within the rib. Therefore, the strength of the base and the cover may be enhanced by the rib and the edge within the rib, respectively.

[0044] Next, a configuration of an optical transceiver and a method for assembling the transceiver according to the present invention will be described as referring to accompanying drawings. FIG. 2 is an exploded view of the optical transceiver of the present invention. The pluggable transceiver 1 shown in FIG. 2 comprises a body unit and an optical subassembly unit. The body unit includes a primary substrate 17, a supplementary substrate 18, a supporting mechanism of these substrates, and a cover 23. The supporting mechanism comprises a base 20, a heat-conducting plate 21 and a supporting plate 22. The optical sub-assembly unit includes transmitting and receiving optical subassemblies, 15 and 16, respectively, and members, 11 to 14, to show the optical coupling function as described below.

[0045] Optical Sub-Assembly Unit

[0046] The optical sub-assembly unit includes, in addition to two optical subassemblies (OSAs), 15 and 16, a receptacle member 13 made of resin, an actuator 12, a bail 11 and a tab 26 plate 14. The bail 11 may be made of resin or metal. The bail 11 and the actuator 12 perform a mechanism to disengage this transceiver 1 with the cage 2. As described later in this specification, to rotate the bail 11, with a projection formed insides of the receptacle member 13 as the center of the rotation, moves the front end of the actuator 12 downward. The actuator 12 is fixed to the receptacle member 13 in the center thereof. Accordingly, due to the seesaw motion as this fixed point being the center of the rotation, the rear end of the actuator 12 moves upward, which pulls the projection formed in the tip of the rear end of the actuator 12 inward to release the engagement of the projection with the opening of the cage 2.

[0047] For the releasing mechanism by the combination of the bail 11 and the actuator 12, it is known that the actuator 12 is slid backward as the rotation of the bail 11. In this mechanism, the projection to be engaged with the cage is fixedly formed in the bottom of the receptacle member 13 not the tip of the actuator 12. Inserting the transceiver 1 into the cage 2, the projection in the bottom thereof engages with the opening of the cage as deforming the cage 2. When releasing the transceiver 1 from the cage 2, the rear end of the actuator 12 spreads the cage 2 outward as sliding the actuator in forward, thus disengaging the projection with the opening of the cage 2.

[0048] In both mechanisms, the rotation of the bail 11 releases the engagement of the transceiver 1 with the cage 2. In other words, it is not restricted to those mechanisms mentioned above as long as the transceiver provides a mechanism to release from the cage and the mechanism is originally brought by the rotation of the bail 11.

[0049] An assembly of the receptacle member 13, the tab plate 14 and subassemblies, the TOSA 15 (Transmitting Optical Sub-Assembly) and the ROSA 16 (Receiving Optical Sub-Assembly), installed in the receptacle member 13, is called as an OSA unit 6 as a whole. From FIG. 3A to FIG. 4B show the assembly of the OSA unit 6.

[0050] The OSAs, 15 and 16, includes the TOSA 15 and the ROSA 16 when the transceiver 1 has a function of the full-duplex communication standard. Both OSAs, 15 and 16, has a coaxial package. The ROSA 16 installs a semiconductor light-receiving device, which is typically an avalanche photodiode, APD, or a PIN-PD, collectively called as the PD, and a preamplifier for amplifying an electrical signal generated by the PD. On the other hand, the TOSA 15 installs a laser diode (LD) and, in addition to the LD, a PD for monitoring the optical output from the LD. A thermo-electric device, called as a Peltier device, may be also installed within the TOSA 15. Moreover, when the driving frequency of the LD exceeds 10 GHz, the driver for the LD may be also installed within the package. The signals and power lines from the OSA to the board, or from the board to the OSA, are transferred through lead pins, 15b and 16b, provided in the rear end of the OSA, which is so-called the stem, 15a and 16a. A flexible printed circuit (FPC) board may be used in place of the lead pins, 15b and 16b.

[0051] The receptacle member 13, which is made of resin, includes first to third portions, 13a to 13c, as shown in FIG. 3A. The first portion 13a, positioned in the front end of the receptacle member 13, provides two cavity, 10a and 10b, into which the optical connector is to be inserted. These cavities, accordingly, called as the optical receptacle. The outer side of the first portion 13a forms a projection 13d that becomes the axis of the rotation of the bail 11. The bottom of the first portion 13a forms a projection 13e in the center and two beaked hooks 13f in the both sides of the projection 13e arranged along the boundary to the second portion 13b, as shown in FIG. 3A that look downs the bottom of the receptacle member 13. The center projection 13e mates with the hollow formed in the center of the actuator 12 and two beaked hooks receives the axis 12g, which enables the actuator 12 to operate in the seesaw motion around the axis 12g.

[0052] The second portion 13b includes side walls and a center partition 13i, which forms to two cavities 13k to receive the TOSA 15 and the ROSA 16, respectively. The bottom of the cavity 13k is opened to set the OSAs therein. U-shaped cuttings are formed in the boundary wall 13q to the third portion 13c. The U-shaped cutting provides a groove 13j along the exposed section, which receives the flange, 15c and 16c, of the OSAs to position respective OSAs in the cavity 13k. Thus setting the OSA within in the cavity 13k, sleeves, 15d and 16d, formed in the head of the OSA, protrude into the optical receptacles, 10a and 10b, of the first portion 13a. Accordingly, semiconductor optical devices such as the LD or the PD installed on the OSAs may optically couple with the optical fiber in the optical receptacles, 10a and 10b. The optical fiber provides, in the tip thereof, the optical connector to be inserted into the receptacle. Since the physical dimensions of the optical receptacles, 10a and 10b, and the position of the sleeves, 15d and 16d, with respect to the optical receptacles, 10a and 10b, are precisely ruled, the position of the OSAs, 15 and 16, in particular the sleeve thereof, must be fixed with respect to the receptacle member 13.

[0053] In a center of the side wall 13h forms a projection 13g. As described later in this specification, to mate this projection 13g with an opening in the tab plate 14 and an opening in the cover 23 assembles the tab plate 14 and the cover 23 with the receptacle member 13. The projection 13d has a gentle slope in the side where the tab plate 14 is inserted to mate the tab plate 14 in smooth, while, the other side thereof has a precipitous slope in order to prevent the tab plate 14 and the cover 23 engaged with the projection 13g from releasing in ease.

[0054] The third portion 13c includes an upper wall 13l continued from the upper wall in the second portion 13b and two projections formed in both ends thereof to mate with the base 20.

[0055] The tab plate 14 includes first and second portions, 14a and 14b, respectively. The first portion 14a forms a plurality of fine, 14c, 14d and 14f, which surround the receptacle member 13 and extend outwardly. These fins, 14c, 14d and 14f, as described later, are to come in contact to the inside of the cage 2 to function the electromagnetic interference (EMI) shield. The fin 14f in both sides and another fin 14f in the bottom have a configuration that the center portion 14 thereof protrudes to come in contact to the inside of the cage 2 in reliable. The tip of the fins, 14c, 14d and 14f, configures in wavy and the bottom portion of this wave fits within the groove formed between the first 13a and second 13b portion of the receptacle member 13. The fin 14d in both sides has an opening 14e within which a latch tab 14e is formed. A center hole of this latch tab 14e engages with the projection 13g of the receptacle member 13 to assemble the tab plate 14 with the member 13.

[0056] The second portion 14b of the tab plate is folded twice. That is, to bend a surface 14g extending form the fin 14f in almost right angle forms a surface 14h, and to bend again in almost a right angle forms another surface 14i. The surfaces, 14g and 14i, put the upper wall 13l of the receptacle member 13 therebetween. Thus, the tab plate 14 is securely assembled with the receptacle member 13.

[0057] The second portion 14b, bent in almost right angle to a direction opposite to the fin 14f, forms a surface 14l. This surface 14l provides two cuttings tracing the U-shaped cutting of the receptacle member 13. In the bottom of this cutting provides a latch tab 14k that fits to the groove 13n formed in the bottom of the U-shaped cutting of the receptacle member 13 to position the tab plate 14 with respect to the receptacle member 13. A center of this surface 14l, between two U-shaped cuttings, forms a spacer 14j by bending the tab plate 14. The width of the U-shaped cutting of the surface 14l is set slightly smaller than the diameter of the OSAs. Accordingly, the OSAs once set within the cutting of the tab plate 14, is hard to disassemble. A portion in the receptacle member 13 facing this spacer 14j, namely, between the end of the center partition 13i and the spacer 14j forms a hole 13r, into which the center post 20f of the base 20 is inserted in the assembly of the base 20 with the receptacle member 13. The tab plate 14 is made from a metal plate with a thickness of about 0.15 mm and may be made of stainless steel only by cutting, bending and tapping without welding, gluing, or screwing, which may not only reduce the cost of the component but also decrease the const of the manufacturing process.

[0058] FIG. 3B shows the assembly of the receptacle member 13 with the tab plate 14, FIG. 4A shows a step for setting OSAs into this assembly of the receptacle member 13 with the tab plate 14, and FIG. 4B shows that the OSAs, 15 and 16, are set within the cavities 13k of the receptacle member 13 with the flanges, 15c and 16c, aligned within the U-shaped cutting. Setting the OSAs, 15 and 16 within the receptacle 13 together with the tab plate 14, a respective pairs of the flanges, 15d and 16d, of the OSAs put the wall 13q behind the groove 13j forming the U-shaped cutting therebetween together with the surface 14l of the tab plate 14. Using the metal sleeve, 15d and 16d, and providing the latch 14k within the U-shaped cutting of the tab plate 14, the electrical conduction from the tab plate 14 to the metal sleeves, 15d and 16d, may be ensured. Since the tab plate 14 is connected to the cage 2 by the fins thereof, 14c, 14d, and 14f, the present transceiver 1 may firmly and securely ground the metal sleeves, 15d and 16d.

[0059] Body Unit

[0060] Next, the body unit of the transceiver 1 will be described. The body unit includes the base 20, the cover 23, the primary substrate 17, a supplementary substrate 1, the heat-conduction plate 21, and the supporting plate 22.

[0061] FIGS. 5A and 5B show the primary and supplementary substrates, 17 and 18, and the base 20. On the primary substrate 17 is installed with an electronic circuit including an integrated circuit (IC) 17c. A laser driver for driving the LD, a signal processor for extracting a clock component and a data component from an amplified signal sent from the ROSA 16, or, when the APD is used as a light-receiving circuit, a bias controller for generating and adjusting the reverse bias voltage provided to the APD, are known as the electronic circuit installed on the substrates, 17 and 18. Moreover, when the thermo-electric controller such as the Peltier device is installed within the TOSA 15, the Peltier driver is also installed within the substrates, 17 and 18. When the size of the circuit above mentioned becomes large, or an additional circuit such as a processor for controlling in inclusive those circuits and a memory are necessary to be installed within the transceiver 1, the supplementary substrate 18 may install circuits not processing a high speed signal, typically the controlling circuit for the APD and the Peltier driver.

[0062] The primary substrate 17 connects to the supplementary substrate 18 with a FPC board 18d. This FPC board 18d is soldered to the primary substrate 17 extending from the side of the supplementary substrate 18. As described later, the supplementary substrate 18 is overlaid on the primary substrate 17 by bending this FPC board 18d. The insulating film 19 is inserted between the supplementary substrate 18 and the heat conducting plate 21 to prevent the back side of the supplementary substrate 18 from electrically shorting to the heat-conducting plate 21.

[0063] On the rear end of the primary substrate 17 is provided with an electrical plug 17f. This plug 17f may mate with the electrical connector in the inner end of the cage 2, which enables the electrical communication of the electrical circuit installed within the transceiver 1 with that of the host system. To configure the electrode of this plug 17f with the predetermined pattern may realize the hot-pluggable function that the transceiver 1 may insert into or remove from the cage as the host system is powered on.

[0064] The electrical connection between the circuit installed on the primary substrate 17 with the OSAs, 15 and 16, are carried out by lead pins, 15b and 16b. The primary substrate 17 provides electrodes, 17a and 17b, each electrically connected to the lead pins, 15b and 16b, of the OSAs. The FPC board may serve as the lead pins. When the operational speed of the transceiver 1 enters the zone over 1 GHz, it is preferable to connect the OSAs, 15 and 16, to the primary substrate 17 with a means that secures the impedance-matching condition. Even the FPC board connects the OSAs, 15 and 16, to the primary substrate 17; the interconnection formed on the FPC board may be a micro-strip line or a co-planar line with the impedance-matched configuration. Moreover, when the connection is carried out under an impedance-mismatched condition, it is preferable to make the connection means as short as possible to escape from the deterioration in the signal transferred from/to the OSAs, 15 and 16 by the mismatching.

[0065] The primary substrate 17 forms a plurality of notches 17e, in the side thereof. As described later, to fit this notch 17e to the projection 20k formed in the base 20 and to abut the step 17d formed in the rear side of the substrate 17 against the side wall of the coupling portion 20h automatically determines the longitudinal position of the substrate 17 within the transceiver 1.

[0066] The base 20 covers the bottom side of the OSA and mounts the primary substrate 17 thereon. The base 20 is also made from a metal plate, such as iron or stainless steel, by cutting, bending and tapping without gluing, welding and screwing, which reduces the cost of the transceiver 1. That is, the base 20 includes a first portion 20a and a second portion 20b. The first portion 20a provides a space opened upward to install the primary substrate 17 thereon. The second portion 20b, formed in the rear of the first portion 20a, provides a space opened downward to install the heat-conducting plate 21.

[0067] The first portion 20a includes a bottom 20g, a pair of sides 20c extending from the bottom 20g, a support 20d that constitutes a portion of the sides 20c, and a front wall 20n extending from the bottom 20g at the front edge. On the center of the bottom 20g is formed with a rib 20q extending along the longitudinal direction of the transceiver 1 and depressed inward. The front end of the bottom 20g forms a center post 20f that is, as aforementioned, inserted into the hole 13r of the receptacle member 13 to assemble the base 20 with the receptacle member 13. The rib 20q not only positions the cover 23 when the base 20 is put into the cover 23 but also mechanically strengthens the base 20.

[0068] The bottom 20g further forms two openings 20l, through which the soldering between the lead pins, 15b and 16b, of the OSAs to the primary substrate 17 may be carried out. The substrate 17 is a double-sided wiring substrate to enhance the flexibility of the interconnection. Although FIG. 5B illustrates electrodes, 17a and 17b, only on the front surface of the substrate 17, electrodes similar to that of the front side are formed in the back side, and the lead pins, 15b and 16b, sandwiches the substrate 17 to connect to the electrodes in both sides thereof. Since to set the substrate 17 to the base 20 hides the electrodes in the back side, the openings 20l exposes the electrodes and the lead pins in the back side to enable the soldering thereto.

[0069] The side for the TOSA in the bottom 20q places the thermal sheet 27 thereon in advance to the assembly. The thermal sheet 27 comes in contact to the stem 15a of the TOSA 15 to dissipate heat generated within the TOSA 15 to the cover 23 via the base 20. Although the ROSA 16 generates heat greater than that from the TOSA 15 because the ROSA installs the pre-amplifier in addition to the PD, such circuit generally shows the dull temperature dependence compared to that of the LD. The LD shows the strong temperature dependence in its light-emitting characteristic. Therefore, the thermal sheet 27 is provided only for the TOSA 13. When the preamplifier installed within the ROSA 16 generates large heat, for example, in the operational speed greater than 10 GHz, it is preferable to provide the thermal sheet for the ROSA 16.

[0070] The side 20c forms a projection 20k in a center thereof and two steps 20m putting the projection 20k therebetween. The projection 20k, as previously described fits into the notch 17e formed in the side of the substrate 17 to position the substrate 17. The step 20m stabilizes the substrate 17 by overlaying the substrate 17 thereon. Without the step 20m, the substrate 17 comes in contact to the base 20 only by the section of the side 20c, which may be unstable for the substrate 17. The step 20m, which is formed by bending a portion of the side 20c inward, secures the stability of the substrate 17 with respect to the base 20.

[0071] The front portion of the side 20c forms the support 20d continuously extending upward. The support 20d provides the opening 20e in a center thereof that engages with the projection 13m of the receptacle member 13 to assemble the base 20 with the receptacle member 13. The front wall 20n configures a gentle U-shape that follows the outer shape of the OSAs, 15 and 16.

[0072] The rear portion of the side 20c forms the coupling portion 20h by extending thereof upward. The second portion 20b includes a coupling surface 20i by bending this coupling portion 20h inward. The coupling surface 20i includes a pair of tabs 20j by bending a portion thereof downward, to be put between the legs 21k of the heat-conducting plate 21. The rear end of the coupling surface 20i is partially bent downward to form the surface 20r and to protrude the rest portion 20s backward. As described later, the gap d₁ between the coupling surface 20i is set slightly narrower than the width of the spacer tab 21l. Assembling the heat conducting plate 21 with the base 20, the coupling surface 20i of the base 20 is to be spread outward by the spacer tab 21l, while the legs 21k of the heat conducting plate 21 presses the tabs 20j inward, which fits the heat conducting plate 21 to the base 20. The front and rear ends of the coupling surface 20i is bent downward to form surfaces, 20p and 20r, to abut against the heat-conducting plate 21, which perform the similar function to that between the tab 20j and the leg 21k. The second portion 21b of the heat conducting plate 21 assembles with the base 20 by fitting the surfaces, 20p and 20r, in the base 20 between the surfaces, 21i and 21s, and by fitting the tabs 20j between the legs 21k.

[0073] Thus, the base 20 of the present invention is made from a metal plate with cutting, bending, and tapping without welding, gluing, screwing or casting. The rib 20q in the bottom secures the mechanical strength of the base 20. Moreover, the assembly with the substrate 17, the receptacle member 13 and the heat conducting plate 21 may be carried out only by fitting, which not only reduces the cost of the material but also simplifies the assembling process.

[0074] Next, the heat-dissipating mechanism of the transceiver 1 will be described. It is inevitable to consume greater power by the devices used in the transceiver 1 as the operational speed thereof increases. Therefore, the heat-dissipating mechanism for these devices becomes necessary. However, since the pluggable transceiver 1 is used by inserting into/extracting from the cage 2, it is impossible to provide a mechanism such as the heat-dissipating fin, although quite popular in the electronic equipment, in the outer surface of the transceiver. These fins prevent the transceiver 1 from smoothly inserting into/extracting from the cage 2. Due to the same reason, it is prohibited that the outside of the transceiver 1 comes in contact to the inside of the cage 2.

[0075] The heat conducting plate 21 primarily performs the heat-dissipating mechanism of the present transceiver 1. From FIG. 7A to FIG. 8B illustrate the heat-conducting plate 21 to be assembled with the base 17, while FIGS. 9A and 9B illustrate the configuration of the heat-conducting plate 21 from the top (FIG. 9A) and from the bottom (FIG. 9B). The heat-conducting plate 21 is also made from a metal plate only by cutting and bending. The plate 21 includes first to fourth portions, 21a to 21d. The first portion 21a configures a substantially rectangle surface 21g, connected to the third portion 21c by bending a rear end thereof upward. This third portion 21c connects the second portion 21b by bending the other edge. The second portion 21b becomes substantially in parallel to the first portion 21a and covers a portion of the first portion 21a. The fourth portion 21d is bent by the other end of the second portion 21b downward to form a surface 21i.

[0076] The first portion 21a includes a surface 21g with a substantially rectangular, back surfaces, 21e and 21f. A plurality of legs 21j extends from a pair of sides thereof downward by bending, tips of which come in contact to the primary substrate 17 to press down the substrate 17. A pair of support posts 21m extends upward from the same sides, from which the legs 21j extends downward. The tips of the support are bent to form the abutting surface 21n. The back surface 21e of the first portion 21a comes in contact to the whole upper surface of the IC 17c installed on the substrate 17 to conduct heat from the IC 17c to the heat-conducting plate 21. The heat conducted to the plate 21 is transferred in the first portion 21a rearward through other surface 21f of the first portion 21a to the third portion 21c. It is preferable to put the thermal sheet 24 between the IC 17c and the back surface 21e of the plate 21 to enhance the efficiency of the heat conduction. The length of the leg 21j is determined to take the thickness of this thermal sheet 24 and that of the IC 17c into consideration. It is preferable to set the length of the leg 21j slightly smaller than the total thickness of the IC 17c and the sheet 24. To press this abutting surface 21n downward by put the cover 23 ensures the thermal contact between the plate 21 and the sheet 24 and between the sheet 24 and the IC 17c.

[0077] The third portion 21c extrudes rearward to come the end surface 21h in contact to the inner end of the cage 2, which enhances the heat transferred to the plate 21 from the IC 17c to dissipate to the cage 2. Even the end surface 21h comes in contact to the cage 2, the insertion into/extraction from the cage 2 of the transceiver 1 is not affected at all. Only the rear end surface 21h of the transceiver 1 provides the effective heat-dissipation path to the cage 2. The present transceiver 1 enables to dissipate heat generated from the IC 17c installed in the middle thereof to the cage 2 by the heat conducting plate 21. The performance of the heat conduction of the plate 21 depends on the cross section thereof. Since the width of the plate 21 is inherently restricted by the width of the transceiver 1, the performance of the heat dissipation primarily depends on the thickness of the plate 21. The present transceiver 1 uses copper based alloy with a thickness of 0.5 mm, which shows a quite effective heat-dissipating function.

[0078] The second portion 21b overlaps a rear half of the surface 21; of the first portion 21a. The second portion 21b also configures a substantially rectangular, an upper surface 21r thereof coming in contact with the inside of the cover 23 to assist the heat-dissipation from the plate 21 to the cover 23. Both sides of the second portion 21b have tabs 21k by bending downward to put the tab 20j formed in coupling surface 20i of the base 20 therebetween. Moreover, the rear of the second portion 21b is bent downward to form the spacer tab 21l. The width of this spacer tab 21l is set to be slightly wider than the gap d₁ between the coupling surfaces 20i of the base 20. Assembling the heat conducting plate 21 with the base 20, the spacer tab 21l and the tab 21k transversely put the coupling surface 20i therebetween, while the surfaces, 21i and 21s, longitudinally put the coupling surface 20i therebetween. Thus, the heat-conducting plate 21 is assembled with the base 20 only by fitting the coupling surface 20i of the base 20 into the second portion 21b of the plate 21 without any screw, welding, or gluing, which enables to simplify the assembling process of the transceiver 1.

[0079] The area of the first portion 21a not overlaid by the second portion 21b mounts the supplementary substrate 18 as described later. One side of the first portion 21a is cut to pass the FPC board connecting the supplementary substrate 18 to the primary substrate 17.

[0080] FIGS. 10A and 10B illustrate one modification of the base 20 and the heat conducting plate 21 shown in FIG. 5 and FIG. 9, respectively. The base 200 illustrated in FIG. 10A is configured in the coupling surface 200i to be wider than the surface 20i in the base 20 of FIG. 5A. The side cross section of the heat conducting plate 210 has a zigzag shape not the U-shaped configuration of the heat conducting plate 21. That is, the heat conducting plate 210 of this embodiment includes the first portion 210a coming in contact to the IC 17c, the fourth portion 210d bent upward at the rear end of the first portion 210a, the second portion 210b bent at the upper end of the fourth portion 210d, and the third portion bent downward at the rear end of the second portion 210b. The heat generated in the IC 17c is conducted to the first portion 210a via the thermal sheet 24, transferred in the heat conducting plate 210, and finally dissipated to the cage 2 at the rear end surface 210h of the third portion 210c.

[0081] Moreover, the third and fourth portions, 210c and 210d, configure to extend downward from the second portion 210b so as to put the coupling surface 200i therebetween. The whole coupling surface 200i of the base 200 comes in contact to the back surface of the second portion 210b. The first portion 210a forms legs 210j extending downward while supports 210m extending upward from the sides thereof. The tip of the support 210m becomes the abutting surface 210n to come in contact to the inside of the cover 23. The difference between two embodiments is that the present heat conducting plate 210 presupposes not to install the supplementary substrate 18. Although the first portion 210a has a space opened upward similar to the previous embodiment, this is for the direct contact of the heat conducting plate 210 to the IC 17c not to install the supplementary substrate 18.

[0082] Standards similar to those appeared in the previous embodiment may be applied for the selection of the material and its thickness for the heat conducting plate 210 of this embodiment. Further, the heat conducting plate 210 is also made from a metal plate only by cutting and bending without any welding and gluing. The plate 210 of this embodiment does not provide an overlaid portion, corresponding to the portion 21f in the previous embodiment, of the first portion 210a and the second portion 210b. Accordingly, the plate 210 may save the metal plate. The heat dissipating mechanism of this embodiment has the same efficiency with those in the previous embodiment.

[0083] FIGS. 11A and 11B illustrate that, after installing the supplementary substrate 18 onto the first portion 21a of the heat conducting plate 21, the supporting plate 22 is set to hold the supplementary substrate 18. As shown in FIG. 5B, the FPC board 18d connects the supplementary substrate 18 to the edge of the primary substrate 17. Installing the primary substrate 17 on the first portion 20a of the base 20, and setting the heat conducting plate 21 on the primary substrate 17, the supplementary substrate 18 is set on the first portion 21a of the heat conducting plate 21 so as to fold the FPC board 18d. At that time, the support post 21m of the heat conducting plate 21 passes through the opening 18e in the supplementary substrate 18. The rear end 18a of the supplementary substrate is configured to protrude from the primary portion thereof, as shown in FIG. 5B, which fits into the cutting portion 21q of the heat conducting plate 21. Since the supplementary substrate 18 is not fixed to the plate 21, the supporting plate 22 assists to fix the supplementary substrate 18 to the plate 21. Moreover, the side of the FPC board 18d forms a projection 18g. This projection is inserted behind the coupling portion 20h at the assembling of the supplementary substrate 18 onto the heat conducting plate 21. Accordingly, this projection may prevent the cover 23 from scratching the edge of the FPC board 18d when setting the cover 23 to the body unit.

[0084] The supporting plate 22, as shown in FIG. 11C, is also made from a metal plate by cutting, bending, and tapping, and includes first to fourth portions, 22a to 22d. The first portion 22a forms an opening 22f and a step. The opening 22f softens the resilient force of the tip portion 22j that comes in contact to the back surface 21s of the heat conducting plate 21. The legs 22l in the first portion 22a form a step 22m by cutting a portion of the tip thereof, which inserts into the notch portion 18f of the supplementary board 18. Moreover, the tabs 22i of the second portion 22b comes in contact to the side of the support post 21m of the heat conducting plate 21, and the legs 22e in the fourth portion 22d also forms a step 22m by cutting a portion of the tip thereof, which inserts into the cutting portion 18g formed in both front sides of the supplementary substrate 18. That is, the supporting plate 22 is assembled with the supplementary substrate 18 and the heat conducting plate 21 by three points, the steps, 22g and 22m, in the legs of the supporting plate and the tab 22i.

[0085] The third potion 22c that connects the second portion 22b to the fourth portion 22d configures in a beam to be bent outward by about 0.2 mm, which shows a resilient force. Putting the cover 23, the cover 23 in the center thereof comes in contact to and presses down this beam. Consequently, the fourth portion 22d of the supporting plate 22 is pressed downward to fix, by the tip 22m thereof, the supplementary substrate 18 to the heat conducting plate 21.

[0086] FIG. 12A illustrates the transceiver 1 including the OSA unit 6, the base 20, the primary and supplementary substrates, 17 and 18, the heat conducting plate 21 and the supporting plate 22, and completes the assembling thereof. The transceiver 1 shown in FIG. 12A further includes the actuator 12 in the front end thereof. Next, the actuator 12 and the configuration of the front side of the receptacle member 13 where the actuator 12 is assembled will be described as referring to FIGS. 12B to 12D.

[0087] FIG. 12B illustrates the actuator 12 from the bottom, while FIG. 12C shows the configuration of the upper surface of the actuator 12. The actuator 12 includes first to third portions, 12a to 12c, and is capable of rotating around the axis 12g formed in the third portion 12c, which makes a seesaw motion for the first and second portions, 12a and 12b, provided in both sides of the third portion 12c. The front end of the first portion 12a forms a pressing surface 12d, while both sides thereof configures inner and outer sliding surfaces, 12i and 12h, respectively. Rotating the bail 11, this rotational motion of the bail 11 may be converted to the up-and-down motion of the first portion 12a because two surfaces, 11f and 11h, of the bail 11 put these inner and outer sliding surfaces, 12h and 12i. That is, two surfaces, 11f and 11h, of the bail function as a cam. The second portion 12b provides an arm 12f and a projection 12e in the tip of this arm 12f. The projection 12e fixes the transceiver 1 to the cage 2 by engaging with the opening of the cage 2 and, by the seesaw motion of the actuator 12 due to the rotation of the bail 11, is brought inward to the transceiver 1, which enables to disengage the projection 12e with the opening and to release the transceiver 1 from the cage 2.

[0088] FIG. 12D illustrates the front configuration of the receptacle member 13. Two openings, the optical receptacles, 10a and 1b, receives the optical connector, and the TOSA 15 and the ROSA 16 installed on the other side of the receptacle member 13 may optically couple with the optical fiber attached to the optical connector. On the bottom of the receptacle member 13, in the center thereof, is configured with a projection 13e, whose section is semicircular, and, in both sides thereof, are configured with beaked hooks. On the other hand, the actuator 12 forms a hollow 12j in the center of the upper surface thereof to receive the center projection 13e of the receptacle member 13, and the axis 12g is fitted into the beaked hook 13f. The center projection 13e restricts the movement of the actuator 12 in the longitudinal direction, while the beaked hook 13f restricts the up-and-down motion thereof, accordingly, the actuator 12 may not be disassembled with the receptacle member 13. On the side walls of the receptacle member 13 are configured with the projection 13d, which becomes the axis for the rotation of the bail 11, as a portion 13p thereof is scooped.

[0089] FIG. 13A illustrates the bail 11 and the actuator 12, both assembled with the receptacle member 13, viewed from the front side, while FIG. 13B is a side view of these members. The bail 11 is a U-shaped member including a pair of legs 11b and a bridge 11a connecting these legs 11b. Although the following description primarily concentrates on the bail 11 made of metal by cutting and bending, the bail 11 may be made of resin. A center of the leg 11b forms an opening 11c to receive the projection 13d on the side wall of the receptacle member 13, a portion 11d of the edge thereof extrudes into the opening 11c and is set within the lacked portion 13p of the projection 13 to restrict the rotation of the bail 11.

[0090] Setting the bail in the neutral position, the initial position, the openings, 10a and 10b, of the optical receptacle wholly exposes not to be hidden by the bail 11. As shown in FIG. 13A, the sliding surface 11h and the cam surface 11f of the bail 11 put and come into contact to the inner and outer sliding surfaces, 12h and 12i. Rotating the bail 11, this rotational motion of the bail 11 operates to the actuator 12. That is, until the angle of the rotation θ, the bail 11 makes no operation to the actuator 12. The angle θ is determined when the cam surface 11f wholly comes in contact to the inner sliding surface 12i, as shown in FIG. 13B.

[0091] Further rotating the bail 11, the cam surface 11f presses down the inner sliding surface 12i, which continues until the outer surface 11i of the leg 11b comes in contact to the inner sliding surface 12i. Thus, by the first portion 12a of the actuator 12 pressing downward, the projection 12e in the second portion 12b is drawn up by the seesaw motion of the actuator 12, which may disengage the projection 12e with the opening of the cage 2.

[0092] During the rotation of the bail 11, the sliding surface 11h of the bail 11 supports the outer sliding surface 12h of the actuator 12, consequently, the actuator 12 is inevitably connected with the bail 11. In other words, the actuator 12 makes the seesaw motion as the inner and outer sliding surfaces, 12h and 12i, are put between the cam surface 11f and the sliding surface 11h of the bail 11. Moreover, the cover 23 supports the actuator 12 by the fin 23d thereof extending from the front edge of the cover 23 (FIG. 14A). The projection 11d within the opening 11c of the leg 11b slides only in the lacked portion 13p of the projection 13d. This combination of the projection 11d and the lacked portion 13p operates as the stopper for the rotation of the bail 11. When the bail 11 returns from the rotated position to the neutral position, the initial position in FIG. 13A, the operation mentioned above are performed in the opposite procedure.

[0093] In the description above, the bail 11 is rotated to disengage the transceiver 1 with the cage 2 by pulling the projection 12e inward. However, when the host system densely installs a plurality of cages 2 and the transceivers neighbor to the target transceiver receive the optical connector attached with the optical cable, a situation may occur that the bail can not rotate because the optical cables of the neighbor transceivers prevent from touching the bridge portion 11a of the bail 11. The transceiver of the present invention may exhibit the same function with the rotation of the bail 11 by pressing down the front end 12d of the actuator 12.

[0094] As shown in FIG. 13b, pressing the front end 12d of the actuator 12 downward, the outer sliding surface 12h pushes the sliding surface 11h of the bail 11 rearward, which influences the rotational moment on the bail 11. The combination of the conventional bail and the actuator may not convert the force downward for the actuator 12 into the rotational moment of the bail 11. The force applied to the actuator 12 may not convert to the rotational moment of the bail 11 because the bail itself operates as the stopper, thereby being unable to release the transceiver 1 from the cage 2. The mechanism provided in the present invention converts the force applied to the front end of the actuator 12 is adequately converted to the rotational motion of the bail 11, which enables the transceiver 1 to be released from the cage 2.

[0095] Recently, the pluggable transceiver 1 is proposed to be applied in the wavelength division multiplexed (WDM) system. In the WDM system using the pluggable transceiver, which provides functions of the optical transmission and the optical reception, the emission wavelength of the optical transmission function is set to be a preset value, and a plurality of optical signals each having the preset wavelength is transmitted by being multiplexed in the outside of the transceiver. In this case, each transceiver is inevitable to have an inherent emission wavelength. However, as shown in FIGS. 1 and 2, even the label attached in the upper or the side of the body records the information of the emission wavelength, it is unable to discriminate the information when the transceiver 1 is set within the cage 2. Accordingly, the present transceiver proposes a configuration that the bridge portion 11a of the bail 11 provides the colored identification, which enables to distinguish in visual the emission wavelength even when the transceiver 1 is in the cage 2.

[0096] According to the WDM communication standard in the wavelength band of 1.5 μm, the optical signals are divided into 64 channels by the interval of 0.8 nm, which corresponds to 100 GHz, and are multiplexed in the wavelength band from 1530 nm to 1650 nm. In order to distinguish respective wavelengths, the last two digits in the integer part and one digit after the decimal point, total three digits, may be marked in the color code. One example of color codes is shown in the following table:

[0097]

TABLE 1
Example of color code
	Number	Color
	0	Black
	1	Brown
	2	Red
	3	Orange
	4	Yellow
	5	Green
	6	Blue
	7	Magenta
	8	Gray
	9	While

By spacing between the first two digits, which corresponds to the integer part, and the last one digit that corresponds to the one after the decimal points, the position of the decimal point may be distinguished. Such colored codes can be recognized when the bail 11 is rotated in front of the optical receptacles, 10a and 10b, even when the transceiver 1 is in the cage 2. In stead of spacing between the digits, to mark the identification may be recognizable to denote the position of the decimal point and to distinguish the emission wavelength. Moreover, three digits for the integer part, the total of four digits, may be applicable for the identification.

[0098] The cover 23 is made from a metal plate with a thickness of about 0.3 mm by cutting and bending without any welding and gluing. The metal used for the cover 23 may be stainless, iron or copper alloy. The cover 23 is fit from the rear side of the transceiver 1 to cover the body unit thereof exposing only the optical receptacle 10. The bottom of the cover 23 forms a slit with two edges, 23a and 23b, forming this slit are fitted within the rib 20q formed in the bottom 20g of the base 20, as shown in FIG. 14B. This configuration of the slit and the rib 20q strengthens the transceiver 1 against the stress despite the cover 23 is made from a metal plate by bending.

[0099] The cover 23 forms a large opening 23e in the front bottom to protrude the projection 12e of the actuator 12 to engage with the opening in the cage 2. The upper surface of the cover 23 forms a plurality of structures, as shown in FIG. 1A, which includes two slits parallel to each other and a portion between these two slits is bent inward. This bent portion has a resilient function and comes in contact to the second portion 21b and to the abutting surface 21n of the heat conducting plate 21 when the cover 23 is fit to the body unit. Consequently, the heat conducting plate 21 not only securely comes in contact to the IC 17c but also ensures the heat dissipation path from the heat conducting plate 21 to the cover 23.

[0100] FIGS. 15A and 15B illustrate the front end of the transceiver 1 fitted with the cover 23. Fins, 14d and 14f, protrude from the front end of the cover 23. The upper of the transceiver 1 also forms the ground tab 23g that surrounds the fin 14f. In the midsection 14m of the fin 14f protrudes to securely come in contact to the inside of the cage 2. It may be applicable that this midsection 14m projects outward by forming two slits parallel to each other and surrounding the mid section 14m. That is, the fin 14f may be the double fin with the sub-fin 14m in the tip thereof. The double fin may have another slit connecting those two slits in the ends thereof. Moreover, another fin 14d may form the midsection protruding from the peripheral thereof, and the double fin structure 14m may be also applicable to this another fin 14d.

[0101] By applying such configuration to the fins, 14f and 14d, the transceiver 1 may securely come in contact to the cage 2. The multiple contacts of the fins, 14f and 14d, the double fin structure 14m, and the ground tab 23g of the cover 23 makes the contact of the transceiver 1 to the cage 2 secure. Since the cage 2 is made from a metal plate, the shape thereof is easily deformed. Even when the transceiver 1 is formed in firm, the secure contact between the transceiver 1 and the cage 2 may be performed if the cage 2 is deformed. The multiple contact of the present invention ensures the secure contact therebetween.

[0102] FIG. 15B illustrates the bottom of the front end of the transceiver 1. The fin 23d of the cover supports the actuator 12 and the projection 12e protrudes from the opening 23e. Moreover, the fin 14c protrudes from both sides of the fin 23d to set in the tip thereof within the groove at the boundary between the second and third portions, 13b and 13c, respectively, of the receptacle member 13. The side fin 14d of the tab plate 14 is also set within the groove at the boundary between second and third portions, 13b and 13c.

[0103] Next, an assembling process of the present optical transceiver will be described in sequence.

[0104] 1. Assembling of the Receptacle Member with the Tab Plate

[0105] First, as shown in FIG. 3A, the tab plate 14 is attached to the receptacle member 13. As described previously, the second portion 14b of the tab plate 14 is folded twice to configure a U-shape cross section. The opening of this U-shaped second portion 14b receives the third portion 13c of the receptacle member 13. At that time, the latch 14k, which comes in contact to the TOSA 15 and the ROSA 16, is set in the groove 13n provided in the bottom of the U-shaped cutting. The tab plate 14 is assembled with the receptacle member 13 by engaging the latch tab 14e with the projection 13g in the side of the receptacle member 13.

[0106] 2. Installing the OSAs in the Receptacle Member

[0107] Next, as shown in FIG. 4A, the TOSA 15 and the ROSA 16 are installed in the receptacle member 13. Respective OSAs, 15 and 16, provide a pair of flanges, 15c and 16c, in the head portion thereof, 15d and 16d, which is the so-called sleeve. These flanges, 15c and 16c, sandwich the wall 13q therebetween that forms the U-shaped cutting. At that time, the tab plate 14 is sandwiched between the wall 13q and one of the flanges in respective OSAs, 15 and 16. The width of the U-shaped cutting in the tab plate 14 is configured to be slightly smaller than the diameter of the sleeves, 15d and 16d. Accordingly, although the OSAs are easily set within the cutting by deforming the tab plate 14, the OSAs once set are hard to be disassembled. On the other hand, the width of the U-shaped cutting of the receptacle member 13 is slightly wider than the diameter of the sleeve, accordingly, to set the OSAs, 15 and 16, within the receptacle member 13 may be carried out without any problem. The tab plate 14 may configure an identification mark in the side where the TOSA is to be installed to distinguish the TOSA 15 from the ROSA 16. The assembly of the OSA unit 6 has completed after installing the OSAs, 15 and 16, in the receptacle member 13 with the tab plate 14. The OSA unit 6 is illustrated in FIG. 4B.

[0108] 3. Installing the Substrate in the Base

[0109] Next, the primary substrate 17, on which the electronic circuit is mounted in advance, is to be assembled with the base 20. Prior to the assembling of the substrate 17, the thermal sheet 27 is put on the portion where the TOSA is to be mounted and the insulating film 19 is attached to the supplementary substrate 18 in the surface opposite to the surface the electronic component is mounted (FIG. 5B).

[0110] The primary substrate 17 is temporarily fixed with the OSA unit 6. This temporal fixing is carried out by putting the substrate 17 between the lead pins, 15b and 16b, when the OSAs, 15 and 16, provide the lead pins, without soldering. While, when the OSAs, 15 and 16, are connected to the substrate 17 with the FPC board, the soldering between the FPC board and the substrate, and between the FPC board and the OSAs, 15 and 16, may be carried out. Occasionally, the OSAs, 15 and 16, soldered with the FPC board may be connected or soldered with the substrate 17 after assembling the OSA unit 6 with the base. Some fixture to assist the assembly of the OSA unit 6 with the base 20 may be utilized in this process.

[0111] Sliding the rear part of the substrate 17 under the coupling surface 20i of the base 20, the step 17d formed in the side of the substrate 17 abuts against the side of the coupling portion 20h, which automatically positions the substrate 17 on the base 20 in the longitudinal direction of the transceiver 1. Pressing down the substrate 17 as abutting against the coupling portion 20h, the projection 20k in the side 20c of the base 20 fits in the notch 17e formed in the side of the substrate 17. By configuring the width of the projection 20k slightly wider than that of the notch 17e, the relative position of the substrate 17 to the base 20 may be automatically and precisely determined. In such configuration, once fitting the substrate 17 in the base 20, the substrate 17 is hard to disassemble. A portion of the side 20c of the base 20 is bent inward to stabilize the position of the substrate 17 put thereon.

[0112] Subsequently, the OSA unit 6 is set to the base 20, as shown in FIGS. 6A and 6B. Pressing the base 20 downward, as the OSA unit 6 is temporarily fixed with the substrate 17, the projection 13m provided in the side of the OSA unit 6, practically provided in the side of the receptacle member 13, engages with the opening 20a of the support 20d. In the meantime of this engagement between the projection 13m and the opening 20e, the center post 20f formed in the front end of the base 20 inserts into the opening 13r in the center of the receptacle member 13. Thus, the OSA unit 6 is fixed to the base 20 by three positions, i.e., two positions in the side and one position in the front end thereof. Moreover, these three fixings are complementary to each other because the fixing in the side is between the convex in the OSA unit 6 and the concavity in the base 20, on the other hand, the fixing in the front end is between the concavity in the OSA unit 6 and the convex in the base 6. The bracket 25 is fit to the ROSA 16 after assembling the OSA unit 6 and the primary substrate 17 with the base 20.

[0113] 4. Installing the Heat Conducting Plate

[0114] Next, the heat conducting plate 21 is installed on the base 20, which is illustrated in successive from FIG. 7A to FIG. 8B. First, the thermal sheet 24 is attached to the IC 17c in direct thereto or via some resin material with flexibility and good thermal conductivity to enhance the heat conductance to the heat conducting plate 21. Subsequently, inserting the rear end portion 20s of the base 20 into the opening between the surface 21h and the spacer tab 21l, of the heat conducting plate 21 and rotating the heat conducting plate 21 as touching the inner surface 21s of the third portion 21 to the surface 20r of the base 20, the surface 21i fits to the front end 20p of the coupling surface 20i. Moreover, the spacer tab 21l widens the gap between the coupling surface 20i while the legs 21k put the tabs 20j of the base 20 therebetween. That is, four sides, 21s, 21i and 21k, surrounding the second portion 21b of the plate 21 wrap and fit to the coupling surface 20i of the base 20. The first portion 21a of the plate 21 in the back surface 21e thereof comes in direct contact to the thermal sheet 24 on the IC 17c. The front legs 21j of the plate 21 presses down the primary substrate 17 by fixing the plate 17 to the base 20.

[0115] Thus, the assembly of the heat conducting plate 21 with the base 20 may be carried out without any screws only by mechanically fitting of two members using the resiliency inherently attributed to metal materials. Therefore, the present transceiver has the configuration quite superior productivity,

[0116] The heat generated in the IC 17c is conducted by the heat conducting plate 21 to the rear end of the transceiver 1, and finally dissipated to the cage 2 at the rear end surface 21h. A substantial gap between the transceiver 1 and the cage is necessary to ensure the smooth insertion into/extraction from the cage 2. Therefore, even the heat generated within the transceiver 1 is effectively conducted to the cover 23; it is effective to radiate thus conducted heat to the cage 2. However, only the rear end of the transceiver 1 may touch to the cage 2 when the transceiver 1 is adequately set within the cage 2 and the plug of the transceiver 1 mates with the connector in the inner end of the cage 2. Accordingly, to conduct heat generated within the transceiver 1 to the rear end thereof may enhance the efficiency of the heat dissipation. The heat conducting plate adopted in the present transceiver 1 may provide a solution for the subject above mentioned.

[0117] 5. Soldering the Lead Pins of the OSAs

[0118] Next, the lead pins, 15b and 16b, extending from the OSAs, 15 and 16, are soldered with the primary substrate 17. The processes thus described determine the relative position of the primary substrate 17 and the OSA unit 6 through the base 20, accordingly, the soldering, which is a type of the permanent fixing, between the lead pins, 15b and 16b, and the electrodes, 17a and 17b, on the substrate 17 may be carried out. The OSAs, 15 and 16, are necessary to position relatively to the receptacle member 13 to provide the optical receptacle 10 following the standard. On the other hand, the primary substrate 17 is necessary to position relatively to the covet 23 to ensure the smooth insertion into the electrical connector provided in the inner end of the cage 2. Thus, to connect the substrate 17 with the lead pins, 15b and 16b, in permanent after the positions of the substrate 17 and the OSAs, 15 and 16, may prevent the mechanical distortion from localizing in the soldering point.

[0119] When the bracket 25 is attached to the ROSA 16, the soldering between this bracket 25 and the ROSA 16 is also carried out simultaneously to the soldering of the lead pins, 15b and 16b. When the FPC board is used for connecting to the substrate 17, the soldering thereof may be optionally carried out after the assembly of the OSA unit 6. The flexibility inherently attributed to the FPC board can compensate the positional discrepancy between the substrate 17 and the OSA unit 6. The soldering for the electrodes, 17a and 17b formed on the primary surface of the substrate 17 may be carried out in the portion exposed between the heat conducting plate 17 and the tab plate 14, while through the opening 201 formed in the bottom 20g of the base 20 for the electrodes in the back side of the substrate 17.

[0120] 6. Installing the Supplementary Substrate

[0121] The supplementary substrate 18 connects to the one side edge of the primary substrate 17 with the FPC board 18d, namely, the FPC board 18d passes through the cutting portion 21p of the heat conducting plate 21 to connect to the primary substrate 17. The supplementary substrate 18 is installed on the heat conducting plate 21 in the first portion 21a thereof by folding the FPC board 18d as the support post 21m passes the opening 18e in the supplementary substrate 18. Since the back surface of the supplementary substrate 18 exposes the interconnections, the insulating film 19 is put thereon, as shown in FIG. 5B.

[0122] The supplementary substrate 18, which is installed on the first portion 21a, is fixed by the supporting plate 22, as shown in FIGS. 11A and 11B. The tip 21j of the first portion 21a of the supporting plate 22 is inserted under the second portion 21b from the cutting 21q as the fourth portion 22d is lifted up. The first portion 22a positions within the cutting 21q, and the step 22g in the leg 22l of the first portion 22a fit to the notch portion 18f of the supplementary substrate 18. Pressing down the second to fourth portions, 22b to 22d, the tab 22i in the second portion 22b abuts against the side of the support post 21m, and the step 22m in the leg 22e of the fourth portion 22d fits to the cutting 18g of the supplementary substrate 18. Therefore, the supporting plate 22 is assembled with the heat conducting plate 21 and the base 20 by the step 22g, the tab 22i and the step 22m. Setting the cover 23 to put the body unit of the transceiver 1 therein, the beam portion 22k bent upward by about 0.2 mm abuts against the inner surface of the cover 23 such that the cover 23 presses this beam portion 22k down, consequently, the supplementary substrate 18 is fixed to the heat conducting plate 21.

[0123] 7. Assembling the Actuator

[0124] Next, the actuator 12 is attached with the transceiver 1 as shown in FIGS. 12B to 12D. The actuator 12, which is made of resin, is assembled with the receptacle member 13 with the axis 12g thereof being fit to the beaked hook 13f and the hollow 12j receiving the center projection 13e of the receptacle member 13. Sliding the actuator 12 on the bottom surface of the receptacle member 13 from the front side, the center projection 13e is set within the hollow 12j as deforming the beaked hook 13f and the projection 13e itself. In the same time, the axis 12g is set within the beaked hook 13f. Due to the tip shape of the beaked hook 13f, the actuator 12 one set within the hook 13f is hard to disassemble, namely, the actuator does not fall even the bottom of the transceiver 1 directs downward.

[0125] 8. Setting the Cover

[0126] Prior to the setting of the cover 23, the thermal sheet 26 is attached, if necessary, to the portion of the TOSA 15 and the ROSA 16 exposed between the receptacle member 13 and the supplementary substrate 18. The thickness of this thermal sheet 26 is adjusted such that the top surface of the sheet 26 comes in directly contact to the ceiling of the cover 23 when the cover 23 is set to the body unit. The thermal sheet 26 may enhance the heat dissipating efficiency from the TOSA 15 and the ROSA to the cover 23. The heat thus conducted to the cover 23 may not only radiate to the cage 2 but also conduct to the rear end of the cover 23 and may be conducted to the inner end of the cage 2.

[0127] The cover 23 configures the latch tab 23f with an opening in both front sides thereof, as shown in FIG. 14A, while, forms the stopper 23h in the rear end. The opening provided in the latch tab 23f is to be engaged with the projection 13g provided in the side of the receptacle member 13. The cover 23 is set to the body unit such that the stopper 23h abuts against the outer side portion 21u of the rear end surface 21h. The fin 23d in the front bottom of the cover 23 extends forward, and the projection 12e formed in the actuator 12 protrudes from the opening 23e in the root of this fin 23d. The fin 23d supports the actuator 12.

[0128] As shown in FIG. 1, on the upper surface of the cover 23 is configured with the structure with two slits and the portion 23c between these slits being bent inward. This bent portion 23c comes in contact to the upper surface 21r and the abutting surface 21 in of the heat conducting plate 21 to press the plate 21 downward. Since this bent portion 23c has an arched shape, it makes no problem for setting/removing the cover 23 from the body unit.

[0129] 9. Installing the Bail

[0130] The bail 11 is attached to the receptacle member 13. The projection 13d in the side wall of the receptacle member 13 inserts into the opening 11c formed in the leg portion 11b of the bail 11, as shown in FIGS. 13A and 13B. The attachment of the bail 11 can be easily carried out by spreading the leg portions 11b in outward. Setting the actuator 12 in the neutral position, the inner and outer sliding surfaces, 12i and 12h, is put between the cam surface 11f and the sliding surface 11h in the tip 11g of the leg 11b. The projection 12e of the actuator 12 can be simply pulled inward by rotating the bail 11 or by pressing the front end 12d of the actuator 12 downward, which may release the transceiver 1 from the cage 2.

[0131] Finally, labels, 28a to 28c, are stuck on the top and side surfaces of the transceiver 1 to complete the assembly. The label 28a on the top surface may be covered by a transparent film 28d to prevent thereof from peeling off by the insertion/extraction of the transceiver 1 to the cage 2.

[0132] The configuration and the assembling process of the present transceiver are thus described with referring to accompanying drawings. Since the present transceiver 1 assembles metal components made by cutting, bending or tapping the single metal plate and fits these metal components without any screws or other joining means, the cost not only the components themselves but also the assembling process can be reduced. Moreover, since the present transceiver 1 provides the heat conducting plate to dissipate heat from the IC 17c installed in the center of the transceiver 1 to the outside thereof, besides the heat conducting plate made of copper alloy, the heat dissipation efficiency can be enhanced for the transceiver 1 to be apply to the high speed application over GHz frequency band.