Connectors

Abstract

This invention discloses a cable termination that provides low signal distortion even at high frequencies. The conductive elements of the cable are fused to the edges of signal conductors in a cable connector or other component terminating the cable. For terminating a differential pair, the conductive elements of the cable can be terminated in the cable termination to the opposite edges of a pair of signal conductors. The conductive elements can be shaped such that the spacing between the conductive elements passing through the cable and the signal paths in the signal conductors of the cable termination is uniform.

Description

Technical Field

[0001] This patent application generally relates to interconnection systems for interconnecting electronic assemblies, such as those that include electrical connectors.

[0002] Cross-reference to related applications

[0003] This application claims priority to and the benefit of U.S. Provisional Patent Application Serial No. 62 / 541,017, filed on August 3, 2017, and entitled “CONNECTOR FOR LOW LOSS ORTHOGONAL INTERCONNECTION SYSTEM,” which is incorporated herein by reference in its entirety. Prior Art

[0004] Electrical connectors are used in many electronic systems. It is often easier and more cost-effective to manufacture the system as separate electronic assemblies, such as printed circuit boards (PCBs), which can be joined together using electrical connectors. A known configuration for joining multiple PCBs uses one PCB as a backplane. Additional PCBs (called "daughterboards" or "daughter cards") can be connected via the backplane.

[0005] A backplane is a printed circuit board (PCB) on which numerous connectors are mounted. Conductive traces in the backplane can be electrically connected to signal conductors in the connectors, allowing signals to be routed between the connectors. Daughter cards can also have connectors mounted on them. The daughter card connectors can be plugged into connectors mounted on the backplane. In this way, signals can be routed between daughter cards via the backplane. Daughter cards can be inserted into the backplane at right angles. Therefore, connectors used in these applications may include right-angle bends and are often referred to as "right-angle connectors."

[0006] Connectors can also be used in other configurations for interconnecting printed circuit boards. Some systems use a midplane configuration. Similar to a backplane, a midplane has connectors mounted on one surface that are interconnected by conductive traces within the midplane. The midplane also has connectors mounted on a second side, allowing daughter cards to be plugged into both sides of the midplane.

[0007] Daughter cards inserted from opposite sides of the midplane typically have an orthogonal orientation. This orientation positions one edge of each printed circuit board adjacent to the edge of each circuit board inserted into the opposite side of the midplane. The traces within the midplane connecting the circuit boards on one side of the midplane to the circuit boards on the other side can be very short, resulting in desirable signal integrity properties.

[0008] A variation of the midplane configuration is called "direct attach." In this configuration, daughter cards are inserted from opposite sides of the chassis that enclose the system's printed circuit boards. These boards are also oriented orthogonally, so that the edge of a board inserted from one side of the chassis abuts the edge of a board inserted from the opposite side of the system. These daughter cards also have connectors. However, instead of plugging into a connector on the midplane, the connector on each daughter card plugs directly into a connector on the printed circuit board inserted from the opposite side of the system.

[0009] Connectors of this configuration are sometimes referred to as orthogonal connectors. Examples of orthogonal connectors are shown in U.S. Patents 7,354,274, 7,331,830, 8,678,860, 8,057,267, and 8,251,745.

[0010] Other connector configurations are also known. For example, RAM connectors are sometimes included in a connector product line, where a daughter card connector has a mating interface with a socket. A RAM connector may have conductive elements that bend through right angles, as in a daughter card connector. However, unlike conventional daughter card connectors, a RAM connector may have a mating interface with mating contact elements that are complementary to and mate with the socket. For example, a RAM connector may have a mating interface with pins or blades, or other mating contacts that may be used in a backplane connector. A RAM connector may be mounted near the edge of a daughter card and accommodate a daughter card connector mounted to another daughter card. Alternatively, a cable connector may be plugged into the RAM connector. Summary of the Invention

[0011] Various aspects of the present invention relate to a low-loss interconnect system.

[0012] In another aspect, some embodiments may relate to a cable assembly having a cable and a cable termination assembly. The cable may include two conductive elements. The cable termination assembly may include a pair of signal conductors having contact tails. The contact tails may have broadsides and edges that join the broadsides. The signal conductors may be held together with the contact tails in an edge-to-edge configuration, with a first edge of a first signal conductor of the pair facing a second edge of a second signal conductor of the pair. The two conductive elements of the cable may each be attached to a respective edge of the first and second edges.

[0013] In some embodiments, a method of manufacturing a cable assembly is provided. The method may include stamping a conductive member into a structure. The structure may include first and second signal conductors and a connecting rod. The first and second signal conductors may be joined by the connecting rod. Each of the first and second signal conductors may include a contact tail having a broadside and an edge joining the broadsides. After stamping the conductive member into the structure, the two conductive elements of a cable may be welded to opposing edges of the first and second signal conductors of the structure, respectively. The method may further include cutting the connecting rod. In some embodiments, the method may also include molding a housing over the structure.

[0014] Some embodiments may relate to a method of manufacturing a cable assembly including first and second signal conductors. Each of the first and second signal conductors may include a contact tail having a broadside and an edge joining the broadsides. The method may include attaching the two conductive elements of the cable to the edges of the first and second signal conductors, respectively. The method may further include shaping portions of the two conductive elements of the cable to reduce impedance discontinuity at the cable attachment point.

[0015] The foregoing summary is provided by way of illustration and is not intended to be limiting. Simple diagram description

[0016] The drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component illustrated in various figures may be represented by a like number. For clarity, not every component may be labeled in every figure. In the drawings: [FIG. 1] is an isometric view of an illustrative electrical interconnect system configured as a right-angle backplane connector according to some embodiments; [Figure 2] is a partially cutaway isometric view of the backplane connector of Figure 1; [Figure 3] is an isometric view of the pin assembly of the backplane connector of Figure 2; [Figure 4] is a partially exploded isometric view of the daughter card connector of Figure 1; [Figure 5] is an isometric view of the wafer assembly of the daughter card connector of Figure 4; [Figure 6] is an isometric view of the wafer module of the wafer assembly of Figure 5; [Figure 7] is an isometric view of a portion of the insulating housing of the wafer assembly of Figure 5; [FIG. 8] is a partially exploded isometric view of the wafer module of the wafer assembly of FIG. 5; FIG9 is a partially exploded isometric view of a portion of the wafer module of the wafer assembly of FIG5 ; FIG10 is a partially exploded isometric view of a portion of the wafer module of the wafer assembly of FIG5; [Figure 11] is an isometric view of an expander module according to some specific examples; [Figure 12A] is an isometric view of a portion of the expander module of Figure 11; [Figure 12B] is an isometric view of a portion of the expander module of Figure 11; [Figure 12C] is an isometric view of a portion of the expander module of Figure 11; [Figure 13] is a partially exploded isometric view of the expander module of Figure 11; [Figure 14] is an isometric view of two expander modules oriented 180 degrees in a rotated manner; [Figure 15A] is an isometric view of the assembly of the two expander modules of Figure 14; [FIG. 15B] is a schematic representation of one end of the assembly of FIG. 15A taken along line BB; [FIG. 15C] is a schematic representation of one end of the assembly of FIG. 15A taken along line CC; [Figure 16] is an isometric view of the connector and expander module assembly of Figure 15A; [Figure 17A] is an isometric view of the expander housing; [Figure 17B] is a partially cutaway perspective view of the expander housing of Figure 17A; [Figure 18A] is a partially exploded isometric view of an orthogonal connector; [Figure 18B] is an isometric view of the assembled orthogonal connector; [Figure 19] is a cross-sectional view of the orthogonal connector of Figure 18B; [Figure 20] is an isometric view of a portion of the orthogonal connector of Figure 18B; [FIG. 21] is a partially exploded isometric view of the electronic system including the orthogonal connector of FIG. 18B and the daughter card connector of FIG. 4; [FIG. 22] is an isometric view of an illustrative electrical interconnect system according to some embodiments; FIG23A is an isometric view of a connector of the electrical interconnect system of FIG22, showing the mating interface of the connector; [FIG. 23B] is an isometric view of the connector of FIG. 23A, showing the mounting interface of the connector; [Figures 24A] to [Figure 24C] are partially exploded isometric views of the connector of Figure 23A; [Figure 24D] is an isometric view of the connector unit of the connector of Figure 23A; [Figure 24E] is an isometric view of the connector unit of the connector of Figure 23B; [FIG. 25A] is an isometric view of a cable assembly according to some embodiments; [Figure 25B] is a partially cutaway isometric view of the cable assembly of Figure 25A; FIG. 25C is an isometric view of a cable assembly module according to an embodiment that can be used to form the cable assembly of FIG. 25A ; [FIG. 26A] is an isometric view of a signal conductor according to one of some specific examples; [FIG. 26B] is an isometric view of a cable attached to the signal conductor pair of FIG. 26A according to some embodiments; [FIG. 27A] is a plan view of a mounting interface between a signal conductor and a cable according to some embodiments; [FIG. 27B] is an elevational view of the mounting interface between the signal conductor pair and the cable of FIG. 27A; FIG. 27C is a plan view of a mounting interface between a signal conductor and a cable according to some embodiments; and [Figure 27D] is an elevational view of the mounting interface between the signal conductor pair and the cable of Figure 27C. Implementation Method

[0017] The inventors have recognized and appreciated that a low-loss interconnect system can be constructed by integrating discrete types of connector elements within a connector. For example, traces routed through a printed circuit board can typically have low loss over relatively short distances. One type of connector element can therefore be configured to attach to traces within the printed circuit board at the footprint where the connector is mounted to the printed circuit board. To route signals to components on the printed circuit board separated by a longer distance from the connector, the loss of the signals routed through the traces on the printed circuit board may be undesirable. A second type of connector element, configured to attach to a cable, can be used to couple to the signals to be routed over longer distances. The cable can then route the signals to locations near the components, providing a lower-loss connection than conventional designs, in which all connections to components on the printed circuit board are made via traces within the circuit board.

[0018] Furthermore, the inventors have recognized and understood techniques for connecting a cable to a conductive element in a second-type connector unit. In some embodiments, the cable can be attached at opposing edges of a conductive element configured to carry differential pair signals. This configuration can orient the signal conductors in the cable in the same plane as the mating contact portion of the conductive element. The edges can be configured so that the conductors of the cable align with the mating contact portion of the conductive element. This configuration can reduce geometric changes at the cable attachment interface that could otherwise result in impedance changes that could affect signal integrity. Alternatively, or in addition, this attachment interface can reduce the amount of metal at the attachment interface, reducing changes in inductance relative to conventional designs in which the cable is soldered to the broadside of the signal conductor, which also reduces changes in impedance.

[0019] Furthermore, the inventors have recognized and appreciated connector designs that reduce the density of wiring shunts within the connector footprint on a printed circuit board. Specifically, the inventors have recognized and appreciated that the footprint of high-density connectors leads to increased density of wiring shunts at the board level, a problem that can be exacerbated by increasing circuit density on the board. The inventors have also recognized and appreciated that the retention force between the connector and the board may need to be enhanced, for example, by means of a hold-down screw, which consumes additional real estate on the board.

[0020] In some embodiments, a connector may be manufactured with multiple types of connector units held by one or more support members. For example, the connector may include a first type connector unit and a second type connector unit. Each connector unit may include at least one row of signal conductors. Each signal conductor may include a contact tail, a mating contact portion, and an intermediate portion extending between the contact tail and the mating contact portion. The contact tails of the signal conductors of the first type connector unit may be configured for attachment to a circuit board, while the contact tails of the signal conductors of the second type connector unit may be configured for attachment to a cable. In some embodiments, the connector may further include a plurality of cables. Each cable may include one or more wires. Each wire may be attached to a contact tail of a signal conductor of a module of the second type unit.

[0021] In some embodiments, the connector unit may be a subassembly formed from a plurality of modules. The modules may be held in a housing and aligned in rows. In some embodiments, each module may include a pair of signal conductors. At the mating interface, the signal conductor pairs in each module may be separated in a direction perpendicular to the rows.

[0022] In some embodiments, each of the signal conductor pairs of the second-type connector unit can include a broadside and an edge joining the broadsides. The signal conductor pairs can be formed from stamped sheet metal, with one or more tie bars connecting two opposing edges of the pair, such that the spacing between the pairs is controlled by the size of the tie bars. The size of the tie bars can be selected based on the size of the wires in the cable to be attached (e.g., AWG 28, AWG 30, etc.). The tie bars can set the spacing between opposing edges of the signal contact tails so that when a wire is attached to each edge, the spacing between the wires produces a desired impedance, such as an impedance that matches the impedance of another portion of the cable or interconnect.

[0023] Before or after the wires are attached to the signal conductors, a housing can be molded around the signal conductor pair so that the contact tails of the signal conductor pair are retained in the housing in an edge-to-edge configuration. The tie bars can then be severed to isolate the signal conductor pair. The inventors have recognized and appreciated that attaching the cables to the edge allows for precise control of the spacing between cables and reduces impedance mismatches, compared to conventional methods of terminating the cables at the surface, thereby forming large bumps. Impedance control can also be provided by attachment, which has a minimal impact on the inductance at the conductor-to-wire interface.

[0024] In some embodiments, the first type connector unit may have signal conductors with a right-angled middle portion. The first type connector unit may be direct attach, orthogonal, RAM, or other desired configuration. In some embodiments, the connector may be given the desired configuration at the mating interface by attaching the expander module to its mating contact portion. Each expander module may include a pair of signal conductors with surrounding shielding. To form an orthogonal connector, the orientation of the signal pair at one end of the expander module may be orthogonal to the orientation at the other end of the module. Both ends of the pair of signal conductors may be terminated with mating contact portions that are adapted to mate with mating contact portions of another connector. The expander module may be held in place by a housing or other suitable retaining structure that is mechanically coupled to the connector unit.

[0025] In some embodiments, each first-type connector unit may include two rows of signal conductors. The mating contact portions of two adjacent signal conductors in the same row may define a pair that is complementary to the signal-conducting pair of the expander module. Thus, attaching the expander module to the first-type connector unit creates a mating interface in which the orientation of the signal-conducting pair is orthogonal to the orientation of the mating contact portion pair of the first-type connector unit.

[0026] In some embodiments, the mating contact portions of the first-type connector unit may form a first array, and the mating contact portions of the second-type unit may form a second array. In some embodiments, the first and second arrays may collectively form the mating interface of the connector. In other embodiments, the mating contact portions of the first-type connector unit may mate with the mating contact portions of the expander module. The mating contact portions at the end of the expander module opposite the end attached to the first-type connector unit may form a third array. In this embodiment, the second and third arrays may collectively form the mating interface of the connector.

[0027] The above principles are illustrated using an embodiment, such as the interconnect system shown in Figure 1. Figure 1 illustrates a form of electrical interconnect system that can be used in an electronic system. In this embodiment, the electrical interconnect system includes a right-angle connector and can be used, for example, to electrically connect a daughter card to a backplane. These figures illustrate two mating connectors. In this embodiment, connector 200 is designed to attach to a backplane, and connector 600 is designed to attach to a daughter card.

[0028] As shown in FIG1 , a modular connector can be constructed using any suitable technology. Additionally, as described herein, the module used to form connector 600 can be used in conjunction with an expander module to form an orthogonal connector. This orthogonal connector can mate with a daughter card connector (e.g., connector 600).

[0029] As shown in FIG. 1 , daughtercard connector 600 includes contact tails 610 designed to attach to a daughtercard (not shown). Backplane connector 200 includes contact tails 210 designed to attach to a backplane (not shown). These contact tails form one end of a conductive element that passes through the interconnect system. When the connector is mounted on a printed circuit board (PCB), these contact tails electrically connect to conductive structures within the PCB that carry signals or connect to reference potentials. In the illustrated embodiment, the contact tails are press-fit "eye-of-the-needle" contacts designed to fit into through-holes in the PCB. However, other contact tail types may be used.

[0030] Each of the connectors also has a mating interface, where one connector can mate with—or detach from—the other connector. Daughter card connector 600 includes mating interface 620. Backplane connector 200 includes mating interface 220. Although not fully visible in the view shown in FIG. 1 , the mating contacts of the conductive elements are partially exposed at the mating interface.

[0031] Each of these conductive elements includes an intermediate portion that connects the contact tail to the mating contact portion. The intermediate portion can be retained within a connector housing, at least a portion of which can be dielectric to provide electrical isolation between the conductive elements. Additionally, the connector housing can include conductive or lossy portions that, in some embodiments, can provide a conductive or partially conductive path between some of the conductive elements. In some embodiments, the conductive portion can provide shielding. The lossy portion can also, in certain circumstances, provide shielding and / or provide desirable electrical properties within the connector.

[0032] In various embodiments, the dielectric member may be molded or overmolded from a dielectric material such as plastic or nylon. Examples of suitable materials include, but are not limited to, liquid crystal polymer (LCP), polyphenylene sulfide (PPS), high-temperature nylon or polyphenylene oxide (PPO), or polypropylene (PP). Other suitable materials may be used, as aspects of the present invention are not limited in this respect.

[0033] All of the materials described above are suitable for use as binder materials in connector manufacturing. According to some embodiments, one or more fillers may be included in some or all of the binder materials. As a non-limiting example, thermoplastic PPS filled with glass fiber to 30% by volume can be used to form the entire connector housing or the dielectric portion of the housing.

[0034] Alternatively or additionally, portions of the housing may be formed from a conductive material, such as machined metal or pressed metal powder. In some embodiments, portions of the housing may be formed from metal or other conductive material, with dielectric members separating the signal conductors from the conductive portions. In the illustrated embodiment, for example, the housing of backplane connector 200 may have a region formed from a conductive material, with dielectric members separating the intermediate portions of the signal conductors from the conductive portions of the housing.

[0035] The housing of daughtercard connector 600 can also be formed in any suitable manner. In the illustrated embodiment, daughtercard connector 600 can be formed from multiple units, which can be subassemblies that can include one or more "wafers" and, in some embodiments, one or more expander modules, as well as one or more support members to hold the components together. Each of the wafers (700, FIG. 5) can include a housing portion, which can similarly include dielectric, lossy, and / or conductive portions. One or more members can hold the wafers in a desired position. For example, support members 612 and 614 can respectively hold the top and back of multiple wafers in a side-by-side configuration. Support members 612 and 614 can be formed from any suitable material, such as a sheet of metal stamped with tabs, openings, or other features that engage corresponding features on the individual wafers.

[0036] Other components that may form part of the connector housing can provide mechanical integrity to daughtercard connector 600 and / or hold the wafer in a desired position. For example, front housing portion 640 ( FIG. 4 ) can receive the portion of the wafer that forms the mating interface. Any or all of these portions of the connector housing can be dielectric, lossy, and / or conductive to achieve the desired electrical characteristics of the interconnect system.

[0037] In some embodiments, each wafer may hold a row of conductive elements forming signal conductors. These signal conductors may be shaped and spaced to form single-ended signal conductors. However, in the embodiment illustrated in FIG1 , the signal conductors are shaped and spaced in pairs to provide differential signal conductors. Each row may include or be bounded by a conductive element serving as a ground conductor. It should be understood that the ground conductor need not be connected to ground, but is shaped to carry a reference potential, which may include ground, a DC voltage, or other suitable reference potential. The "ground" or "reference" conductor may have a different shape than the signal conductors and be configured to provide suitable signal transmission properties for high-frequency signals.

[0038] The conductive elements can be made of metal or any other material that is electrically conductive and provides suitable mechanical properties for use in electrical connectors. Phosphor bronze, beryllium copper, and other copper alloys are non-limiting examples of materials that can be used. The conductive elements can be formed from these materials in any suitable manner, including by stamping and / or forming.

[0039] The spacing between adjacent rows of conductors can be within a range that provides the desired density and desired signal integrity. As a non-limiting example, the conductors can be stamped from a 0.4 mm thick copper alloy, and the conductors within each row can be spaced 2.25 mm apart, with the rows of conductors spaced 2.4 mm apart. However, higher density can be achieved by placing the conductors closer together. In other embodiments, smaller dimensions can be used to provide higher density, such as, for example, a thickness between 0.2 and 0.4 mm or a spacing of 0.7 to 1.85 mm between posts or between conductors within a post. Furthermore, each row can include four pairs of signal conductors, enabling a density of 60 or more pairs per linear inch for the interconnect system illustrated in FIG. 1 . However, it should be understood that more pairs per row, closer spacing between pairs within a row, and / or smaller distances between rows can be used to achieve a higher density connector.

[0040] The wafer can be formed in any suitable manner. In some embodiments, the wafer can be formed by stamping rows of conductive elements from a sheet of metal and overmolding a dielectric portion over the intermediate portions of the conductive elements. In other embodiments, the wafer can be assembled from modules, each module including a single single-ended signal conductor, a single pair of differential signal conductors, or any suitable number of single-ended or differential pairs.

[0041] The inventors have recognized and appreciated that assembling wafers from modules can help reduce "skew" in higher frequency signal pairs, such as those between approximately 25 GHz and 40 GHz or higher. In this context, skew refers to the difference in electrical propagation time between signals in a pair operating as differential signals. Modular configurations designed to reduce skew are described, for example, in U.S. Patents 9,509,101 and 9,450,344, which are incorporated herein by reference.

[0042] According to the technology described in those patents incorporated by reference, in some embodiments, the connector can be formed from modules, each module carrying a signal pair. The modules can be individually shielded, such as by attaching shielding members to the modules and / or inserting the modules into an organizer or other structure that can provide electrical shielding between the pairs and / or ground structures around the signal-carrying conductive elements.

[0043] In some embodiments, the signal conductor pairs within each module can be broadside coupled over a majority of their length. Broadside coupling enables the signal conductors in a pair to have the same physical length. To facilitate routing of signal traces within the connector footprint of a printed circuit board to which the connector is attached and / or construction of the connector's mating interface, the signal conductors can be aligned by edge-to-edge coupling in one or both of these regions. Thus, the signal conductors can include transition regions where coupling changes from edge-to-edge to broadside, or vice versa. As described below, these transition regions can be designed to prevent mode conversion or suppress undesirable propagation modes that could interfere with the signal integrity of the interconnect system.

[0044] Modules can be assembled into a wafer or other connector structure. In some embodiments, a different module can be formed for each row position that will be assembled into a right-angle connector. These modules can be used together to construct a connector with as many rows as desired. For example, a module of one shape can be formed for a pair positioned at the shortest row of a connector (sometimes referred to as the ab row). A separate module can be formed for the conductive elements in the next longest row (sometimes referred to as the cd row). The interior of the module with the cd row can be designed to conform to the exterior of the module with the ab row.

[0045] This pattern can be repeated for any number of pairs. Each module can be shaped to work with the module carrying the pair for shorter and / or longer rows. To produce connectors of any suitable size, connector manufacturers can assemble multiple modules onto a wafer to provide the desired number of pairs. In this way, connector manufacturers can introduce a connector family for a widely used connector size—such as two pairs. As customer requirements change, connector manufacturers can purchase tools for each additional pair, or modules containing multiple pairs or groups of pairs to produce larger connector sizes. The tools used to produce modules for smaller connectors can be used to produce modules for shorter rows of even larger connectors. This modular connector is illustrated in Figure 6.

[0046] 1 is provided in further detail, which shows a partial cross-sectional view of backplane connector 200. In the embodiment illustrated in FIG2, the front wall of housing 222 is cut away to expose an interior portion of mating interface 220.

[0047] In the illustrated embodiment, the backplane connector 200 also has a modular structure. A plurality of pin modules 300 are organized to form an array of conductive elements. Each of the pin modules 300 can be designed to mate with a module of the daughter card connector 600.

[0048] In the illustrated embodiment, four columns and eight rows of pin modules 300 are shown. With each pin module having two signal conductors, the four columns 230A, 230B, 230C, and 230D of pin modules form rows with a total of four pairs of signal conductors, or eight. However, it should be understood that the number of signal conductors per column or row is not a limitation of the present application. A greater or fewer number of columns of pin modules may be included within the housing 222. Similarly, a greater or fewer number of rows may be included within the housing 222. Alternatively or additionally, the housing 222 may be considered a module for a backplane connector, and multiple such modules may be aligned side-by-side to extend the length of the backplane connector.

[0049] In the embodiment illustrated in FIG2 , each of the pin modules 300 contains conductive elements that serve as signal conductors. These signal conductors are held within insulating members that may serve as part of the housing of the backplane connector 200. The insulating portion of the pin modules 300 may be positioned to separate the signal conductors from the rest of the housing 222. In this configuration, the rest of the housing 222 may be conductive or partially conductive, such as may result from the use of lossy materials.

[0050] In some embodiments, the housing 222 may contain both conductive and lossy portions. For example, the shield, including the walls 226 and the plate layer 228, may be formed from powdered metal or any other suitable conductive material. The pin module 300 may be inserted into an opening in the plate layer 228.

[0051] Lossy or conductive components can be positioned adjacent to the rows 230A, 230B, 230C, and 230D of the pin modules 300. In the embodiment of FIG. 2 , separators 224A, 224B, and 224C are shown between adjacent rows of pin modules. Separators 224A, 224B, and 224C can be conductive or lossy and can be formed as part of the same operation or from the same components as the walls 226 and the plate layer 228. Alternatively, separators 224A, 224B, and 224C can be separately inserted into the housing 222 after the walls 226 and the plate layer 228 are formed. In embodiments where separators 224A, 224B, and 224C are formed separately from the walls 226 and the plate layer 228 and subsequently inserted into the housing 222, separators 224A, 224B, and 224C can be formed from a different material than the walls 226 and / or the plate layer 228. For example, in some embodiments, walls 226 and plate layers 228 can be conductive, while separators 224A, 224B, and 224C can be lossy or partially lossy and partially conductive.

[0052] In some embodiments, other lossy or conductive components may extend into mating interface 220, perpendicular to board layer 228. Component 240 is shown adjacent to the endmost rows 230A and 230D. Separator components 240, which are approximately the same width as a row, are positioned in rows adjacent to rows 230A and 230D, compared to separators 224A, 224B, and 224C, which extend across mating interface 220. Daughtercard connector 600 may include slots in its mating interface 620 for receiving separators 224A, 224B, and 224C. Daughtercard connector 600 may include openings similarly accommodating component 240. Component 240 may have similar electrical effects as separators 224A, 224B, and 224C, as both can suppress resonance, crosstalk, or other undesirable electrical effects. Because member 240 fits within an opening in daughter card connector 600 that is smaller than separators 224A, 224B, and 224C, the housing portion of daughter card connector 600 may have greater mechanical integrity at the sides that receive member 240.

[0053] FIG3 illustrates pin module 300 in greater detail. In this embodiment, each pin module includes a pair of conductive elements serving as signal conductors 314A and 314B. Each of the signal conductors has a mating interface portion shaped as a pin. In FIG3 , the mating interface is located on a module configured for a backplane connector. However, it should be understood that in the embodiments described below, similar mating interfaces can be formed at either end of the signal conductors of an expander module, or, in some embodiments, at both ends of the signal conductors of an expander module.

[0054] As shown in FIG3 , where the module is configured for use in a backplane connector, the opposite ends of the signal conductors have contact tails 316A and 316B. In this embodiment, the contact tails are shaped as press-fit flexible portions. The middle portion of the signal conductor that connects the contact tails to the mating contact portion passes through the pin module 300.

[0055] Conductive elements serving as reference conductors 320A and 320B are attached to opposing outer surfaces of pin module 300. Each reference conductor has a contact tail 328 shaped for electrical connection to a through-hole in a printed circuit board. The reference conductors also have mating contact portions. In the illustrated embodiment, two types of mating contact portions are described. Flexible member 322 can serve as the mating contact portion, pressing against the reference conductor in daughtercard connector 600. In some embodiments, surfaces 324 and 326 can alternatively or additionally serve as the mating contact portion, where a reference conductor from the mating conductor can press against reference conductor 320A or 320B. However, in the illustrated embodiment, the reference conductors can be shaped so that electrical contact is made only at flexible member 322.

[0056] Turning to FIG4 , further details of daughter card connector 600 are shown in a partially exploded view. The components shown in FIG4 can be assembled into a daughter card connector configured to mate with a backplane connector, as described above. Alternatively or additionally, a subset of the connector components shown in FIG4 can be combined with other components to form an orthogonal connector. This orthogonal connector can mate with a daughter card connector, as shown in FIG4 .

[0057] As shown, connector 600 includes multiple dies 700A held together in a side-by-side configuration. Here, eight dies are shown, corresponding to the eight rows of pin modules in backplane connector 200. However, as with backplane connector 200, the size of the connector assembly can be configured by integrating more rows per die, more dies per connector, or more connectors per interconnect system.

[0058] Conductive elements within wafer 700A may include mating contact portions and contact tails. Contact tails 610 are shown extending from a surface of connector 600 adapted for mounting against a printed circuit board. In some embodiments, contact tails 610 may pass through member 630. Member 630 may include insulating, lossy, and / or conductive portions. In some embodiments, contact tails associated with signal conductors may pass through the insulating portion of member 630. Contact tails associated with reference conductors may pass through either the lossy or conductive portion.

[0059] In some embodiments, the conductive or lossy portion can be flexible, such as from a conductive elastomer or other material known in the art for forming gaskets. The flexible material can be thicker than the insulating portion of member 630. This flexible material can be positioned to align with pads on the surface of the daughter card to which connector 600 will be attached. These pads can be connected to reference structures within a printed circuit board so that when connector 600 is connected to the printed circuit board, the flexible material contacts the reference pads on the surface of the printed circuit board.

[0060] The conductive or lossy portions of member 630 can be positioned to electrically connect to reference conductors within connector 600. Such connections can be made, for example, by passing contact tails of the reference conductor through the lossy conductive portion. Alternatively or additionally, in embodiments where the lossy or conductive portions are flexible, they can be positioned to press against mating reference conductors when the connector is attached to a printed circuit board.

[0061] The mating contact portion of wafer 700A is held in front housing portion 640. The front housing portion can be made of any suitable material, which may be insulating, lossy, and / or conductive, or may include any suitable combination or combinations of such materials. For example, using materials and techniques similar to those described above for housing wall 226, the front housing portion can be molded from a filled, lossy material, or formed from a conductive material. As shown, the wafer is assembled from modules 810A, 810B, 810C, and 810D ( FIG. 6 ), each having a pair of signal conductors surrounded by a reference conductor. In the illustrated embodiment, front housing portion 640 has multiple channels, each positioned to receive a pair of such signal conductors and an associated reference conductor. However, it should be understood that each module may contain a single signal conductor or more than two signal conductors.

[0062] In the illustrated embodiment, the front housing 640 is shaped to fit within the wall 226 of the backplane connector 200. However, in some embodiments, as described in more detail below, the front housing can be configured to connect to an extender housing.

[0063] FIG5 illustrates wafer 700. Multiple such wafers can be aligned side by side and held together using one or more support members or any other suitable means to form a daughter card connector, or, as described below, an orthogonal connector. In the illustrated embodiment, wafer 700 is a subassembly formed from a plurality of modules 810A, 810B, 810C, and 810D. These modules are aligned to form a row of mating contact portions along one edge of wafer 700 and a row of contact tails along another edge of wafer 700. In the embodiment where the wafer is designed for a right-angle connector, those edges are vertical, as illustrated.

[0064] In the illustrated embodiment, each of the modules includes a reference conductor that at least partially encloses the signal conductor. The reference conductor can similarly have a mating contact portion and a contact tail.

[0065] The modules can be held together in any suitable manner. For example, the modules can be held within a housing, which, in the illustrated embodiment, is formed with components 900A and 900B. Components 900A and 900B can be formed separately and then secured together, capturing modules 810A...810D therebetween. Components 900A and 900B can be held together in any suitable manner, such as by attaching components that form an interference fit or a snap fit. Alternatively or additionally, adhesives, welding, or other attachment techniques can be used.

[0066] Components 900A and 900B can be formed from any suitable material. Such material may be an insulating material. Alternatively or additionally, such material may be or include lossy or conductive components. For example, components 900A and 900B may be formed by molding such materials into the desired shape. Alternatively, components 900A and 900B may be formed in place around modules 810A...810D, for example, via insert molding. In this embodiment, components 900A and 900B do not need to be formed separately. Rather, the housing portion used to hold modules 810A...810D may be formed in a single operation.

[0067] Figure 6 shows modules 810A...810D without components 900A and 900B. In this view, the reference conductor is visible. Signal conductors (not visible in Figure 6) are enclosed within the reference conductor, forming a waveguide structure. Each waveguide structure includes a contact tail region 820, a middle region 830, and a mating contact region 840. Within the mating contact region 840 and contact tail region 820, the signal conductors are positioned edge-to-edge. Within the middle region 830, the signal conductors are positioned for broadside coupling. Transition regions 822 and 842 are provided to transition between edge-coupling and broadside coupling orientations.

[0068] The transition regions 822 and 842 in the reference conductor can correspond to transition regions in the signal conductor, as described below. In the illustrated embodiment, the reference conductor forms an enclosure around the signal conductor. In some embodiments, the transition regions in the reference conductor can maintain a substantially uniform spacing between the signal and reference conductors over the length of the signal conductor. Thus, the enclosure formed by the reference conductor can have different widths in different regions.

[0069] The reference conductor provides shielding coverage along the length of the signal conductor. As shown, coverage is provided along substantially the entire length of the signal conductor, including coverage in the mating contact portion and the midsection of the signal conductor. The contact tails are shown exposed so that they can contact the printed circuit board. However, in use, these mating contact portions will be adjacent to ground structures within the printed circuit board, so exposure, as shown in FIG6 , does not compromise shielding coverage along substantially the entire length of the signal conductor. In some embodiments, the mating contact portion may also be exposed for mating with another connector. Thus, in some embodiments, shielding coverage may be provided over more than 80%, 85%, 90%, or 95% of the midsection of the signal conductor. Similarly, shielding coverage may be provided in the transition region, such that shielding coverage may be provided over more than 80%, 85%, 90%, or 95% of the combined length of the midsection and transition region of the signal conductor. In some embodiments, as illustrated, the mating contact region and some or all of the contact tails may also be shielded such that, in various embodiments, the shield coverage may exceed 80%, 85%, 90%, or 95% of the length of the signal conductor.

[0070] In the illustrated embodiment, the waveguide-like structure formed by the reference conductor has a wider dimension in the row direction of the connector in the contact tail region 820 and the mating contact region 840 to accommodate the wider dimension of the signal conductors arranged side by side in the row direction in these regions. In the illustrated embodiment, the contact tail region 820 and the mating contact region 840 of the signal conductor are separated by a distance that aligns them with the mating contacts of the mating connector or the contact structures on the printed circuit board to which the connector is attached.

[0071] This equal spacing requirement means that the waveguides will be wider in the row dimension than in the lateral direction, thereby providing an aspect ratio of the waveguides in these regions that can be at least 2: 1, and in some embodiments can be approximately at least 3: 1. In contrast, in the intermediate region 830, the signal conductors are oriented such that the wide dimensions of the signal conductors overlap in the row dimension, resulting in an aspect ratio of the waveguides that can be less than 2: 1, and in some embodiments can be less than 1.5: 1 or approximately 1: 1.

[0072] With this smaller aspect ratio, the maximum dimension of the waveguide in the middle region 830 will be smaller than the maximum dimension of the waveguide in regions 830 and 840. Because the lowest frequency of waveguide propagation is inversely proportional to the length of its shortest dimension, the lowest frequency propagation mode that can be excited in the middle region 830 is higher than the lowest frequency propagation mode that can be excited in the contact tail region 820 and the mating contact region 840. The lowest frequency mode that can be excited in the transition region will be somewhere in between. Because the transition from edge coupling to broadside coupling has the potential to excite undesirable modes in the waveguide, signal integrity can be improved if the frequencies of these modes are above the connector's expected operating range, or at least as high as possible.

[0073] These regions can be configured to avoid mode conversion when switching between coupling orientations, which would excite undesirable signals propagating through the waveguide. For example, as shown below, the signal conductor can be shaped so that it occurs in the intermediate region 830 or the transition regions 822 and 842, or partially in both. Additionally or alternatively, the module can be structured to suppress undesirable modes excited in the waveguide formed by the reference conductor, as described in more detail below.

[0074] While the reference conductors can substantially enclose each pair, the enclosure is not required to be free of openings. Thus, in an embodiment shaped to provide a rectangular shield, the reference conductors in the middle region can be aligned with at least a portion of all four sides of the signal conductors. The reference conductors can be combined to provide, for example, 360-degree coverage around the signal conductor pairs. This coverage can be achieved, for example, by overlapping or physically contacting the reference conductors. In the illustrated embodiment, the reference conductors are U-shaped shells that together form the enclosure.

[0075] Regardless of the shape of the reference conductor, 360-degree coverage can be provided. For example, this coverage can be provided with a circular, elliptical, or any other suitable reference conductor shape. However, complete coverage is not required. For example, coverage can have an angular range between approximately 270 and 365 degrees. In some embodiments, coverage can be in the range of approximately 340 to 360 degrees. This coverage can be achieved, for example, by slots or other openings in the reference conductor.

[0076] In some embodiments, shield coverage can vary in different regions. In the transition region, shield coverage can be greater than in the intermediate region. In some embodiments, shield coverage can have an angular range greater than 355 degrees, or even 360 degrees in some embodiments, due to direct contact or even overlap of the reference conductor in the transition region, even though less shield coverage is provided in the transition region.

[0077] The inventors have recognized and appreciated that, in a sense, completely enclosing a signal pair within a reference conductor in a central region can produce undesirable effects that affect signal integrity, particularly when used in conjunction with the transition between edge coupling and broadside coupling within a module. The reference conductor surrounding the signal pair can form a waveguide. Signals on the pair, particularly within the transition region between edge coupling and broadside coupling, can induce energy from differential propagation modes between the edges to excite signals that can propagate within the waveguide. According to some embodiments, one or more techniques can be used to avoid excitation of these undesirable modes, or to suppress them if they are excited.

[0078] Some techniques can be used to increase the frequency at which undesirable modes are excited. In the illustrated embodiment, the reference conductor can be shaped to leave openings 832. These openings can be located in the narrower wall of the enclosure. However, in embodiments with wider walls, the openings can be located in the wider wall. In the illustrated embodiment, openings 832 extend parallel to the midsection of the signal conductors and are located between the signal conductors forming a pair. These slots reduce the angular range of shielding, such that the angular range of shielding adjacent to the midsection where the broadsides of the signal conductors are coupled can be less than 360 degrees. For example, it can be within a range of 355 degrees or less. In embodiments where components 900A and 900B are formed by overmolding a lossy material onto the module, the lossy material can be allowed to fill openings 832, extending or not into the interior of the waveguide, which can suppress the propagation of undesirable signal propagation modes that can reduce signal integrity.

[0079] In the embodiment illustrated in FIG6 , the opening 832 is slot-shaped, effectively dividing the shield in half in the middle region 830. The lowest frequency that can be excited in a structure acting as a waveguide (e.g., the effect of a reference conductor substantially surrounding a signal conductor, as illustrated in FIG6 ) is inversely proportional to the size of each side. In some embodiments, the lowest-frequency waveguide mode that can be excited is a TEM mode. By effectively shortening the sides by incorporating the slot-shaped opening 832, the frequency of the TEM mode that can be excited is increased. A higher resonant frequency can mean that less energy within the operating frequency range of the connector is coupled to unwanted propagation within the waveguide formed by the reference conductor, improving signal integrity.

[0080] In region 830, a pair of signal conductors are broadside coupled, and opening 832 (with or without lossy material therein) suppresses the TEM co-propagating mode. While not being bound by any particular theory of operation, the inventors speculate that opening 832, combined with the edge coupled to the broadside coupling transition, helps provide a balanced connector suitable for high-frequency operation.

[0081] FIG7 illustrates component 900, which may be a representation of component 900A or 900B. As can be seen, component 900 is formed with channels 910A...910D shaped to receive modules 810A...810D shown in FIG6 . With the modules positioned within the channels, component 900A can be secured to component 900B. In the illustrated embodiment, components 900A and 900B are attached by threading a post, such as post 920, in one component through a hole, such as hole 930, in the other component. The post may be welded or otherwise secured within the hole. However, any suitable attachment mechanism may be used.

[0082] Components 900A and 900B can be molded from or include lossy materials. Any suitable lossy material can be used for these and other structures that are "lossy." Materials that conduct electricity but have some loss, or that absorb electromagnetic energy within the frequency range of interest through another physical mechanism, are generally referred to herein as "lossy" materials. Electrically lossy materials can be formed from lossy dielectric materials and / or poorly conductive and / or lossy magnetic materials. For example, magnetically lossy materials can be formed from materials traditionally considered ferromagnetic, such as those having a magnetic loss tangent greater than approximately 0.05 within the frequency range of interest. "Magnetic loss tangent" is the ratio of the imaginary part to the real part of a material's composite electrical permeability. Actual lossy magnetic materials or mixtures containing lossy magnetic materials may also exhibit useful amounts of dielectric loss or conductive loss effects over portions of the frequency range of interest. Electrically lossy materials can be formed from materials traditionally considered dielectrics, such as those having an electrical loss tangent greater than approximately 0.05 within the frequency range of interest. "Electrical loss tangent" is the ratio of the imaginary part to the real part of a material's complex permittivity. Electrically lossy materials can also be formed from materials typically considered conductors but relatively poor conductors within the frequency range of interest, containing conductive particles or regions that are sufficiently dispersed that they do not provide high electrical conductivity or are otherwise fabricated with properties that result in relatively poor bulk conductivity within the frequency range of interest compared to good conductors such as copper.

[0083] Electrically lossy materials typically have a bulk conductivity of about 1 Siemens / meter to about 100,000 Siemens / meter, and preferably about 1 Siemens / meter to about 10,000 Siemens / meter. In some embodiments, materials with a bulk conductivity between about 10 Siemens / meter and about 200 Siemens / meter may be used. As a specific example, a material with a conductivity of about 50 Siemens / meter may be used. However, it should be understood that the conductivity of the material can be selected empirically or through electrical simulation using known simulation tools to determine a suitable conductivity that provides both suitably low crosstalk and suitably low signal path attenuation or insertion loss.

[0084] Electrically lossy materials can be partially conductive materials, such as materials having a surface resistivity between 1 Ω / square and 100,000 Ω / square. In some embodiments, the electrically lossy material has a surface resistivity between 10 Ω / square and 1000 Ω / square. As a specific example, the material can have a surface resistivity between approximately 20 Ω / square and 80 Ω / square.

[0085] In some embodiments, an electrically lossy material is formed by adding a filler containing conductive particles to a binder. In this embodiment, the lossy component can be formed by molding the filler-containing binder or otherwise shaping it into the desired form. Examples of conductive particles that can be used as fillers to form electrically lossy materials include carbon or graphite formed into fibers, flakes, nanoparticles, or other types of particles. Metals in powder, flake, fiber, or other particulate form can also be used to provide suitable electrically lossy properties. Alternatively, a combination of fillers can be used. For example, metal-plated carbon particles can be used. Silver and nickel are suitable metals for electroplating fibers. Coated particles can be used alone or in combination with other fillers, such as carbon flakes. The binder or matrix can be any material that will solidify, cure, or otherwise hold the filler material in place. In some embodiments, the binder can be a thermoplastic material conventionally used in the manufacture of electrical connectors to facilitate molding the electrically lossy material into the desired shape and position as part of the connector's fabrication. Examples of such materials include liquid crystal polymers (LCPs) and nylon. However, many alternative forms of binder materials may be used. Curable materials such as epoxy resins may be used as binders. Alternatively, materials such as thermosetting resins or adhesives may be used.

[0086] Furthermore, while the binder material described above can be used to form an electrically lossy material by forming a binder around a conductive particulate filler, the present application is not limited thereto. For example, the conductive particles can be impregnated into a formed matrix material, or coated onto a formed matrix material, such as by applying a conductive coating to a plastic or metal component. As used herein, the term "binder" encompasses materials that encapsulate fillers, impregnate fillers, or otherwise serve as a base circuit board for retaining fillers.

[0087] Preferably, the filler will be present in a sufficient volume percentage to allow for the formation of a conductive path from particle to particle. For example, when metal fibers are used, the fibers may be present in an amount of approximately 3% to 40% by volume. The amount of filler may affect the conductive properties of the material.

[0088] Filler materials are commercially available, such as those sold by Celanese Corporation under the trade name Celestran®, which can be filled with carbon fiber or stainless steel filaments. Lossy materials, such as lossy conductive carbon-filled adhesive preforms, such as those sold by Techfilm of Billerica, Massachusetts, can also be used. This preform can include an epoxy adhesive filled with carbon fiber and / or other carbon particles. The adhesive encapsulates the carbon particles, which act as reinforcement for the preform. This preform can be inserted into a connector wafer to form all or part of the housing. In some embodiments, the preform can be bonded via an adhesive in the preform, which can be cured during a heat treatment process. In some embodiments, the adhesive can take the form of a separate conductive or non-conductive adhesive layer. In some embodiments, the adhesive in the preform can alternatively or additionally be used to secure one or more conductive elements, such as foil strips, to the lossy material.

[0089] Various forms of reinforcing fibers can be used, in woven or non-woven form, coated or uncoated. Non-woven carbon fiber is one suitable material. As the present invention is not limited in this respect, other suitable materials may be used, such as custom blends sold by RTP Company.

[0090] In some embodiments, the lossy component can be manufactured by stamping a preform or a sheet of lossy material. For example, the insert can be formed by stamping a preform as described above with an appropriate opening pattern. However, other materials can be used instead of or in addition to such a preform. For example, a sheet of ferromagnetic material can be used.

[0091] However, lossy components can also be formed in other ways. In some embodiments, a lossy component can be formed by alternating layers of lossy and conductive materials, such as metal foil. These layers can be rigidly attached to one another, such as by using epoxy or other adhesives, or can be held together in any other suitable manner. The layers can be given the desired shape before being secured to one another, or they can be stamped or otherwise shaped after being held together.

[0092] Figure 8 shows further details of the construction of wafer module 1000. Module 1000 can represent any module in a connector, such as any of modules 810 through 810D shown in Figures 5 and 6. Each of modules 810A through 810D can have the same general construction, and some features can be common to all modules. For example, the contact tail region 820 and the mating contact region 840 can be common to all modules. Each module can include a central region 830, but the length and shape of central region 830 can vary depending on the module's location within the wafer.

[0093] In the illustrated embodiment, module 100 includes a pair of signal conductors 1310A and 1310B ( FIG. 9 ) held within insulating housing portion 1100. Insulating housing portion 1100 is at least partially enclosed by reference conductors 1010A and 1010B. This subassembly can be held together in any suitable manner. For example, reference conductors 1010A and 1010B can include features that engage with each other. Alternatively or in addition, reference conductors 1010A and 1010B can include features that engage with insulating housing portion 1100. As yet another example, once components 900A and 900B are secured together, the reference conductors can be held in place, as shown in FIG. 5 .

[0094] In the embodiment illustrated in Figures 3 and 8, sub-region 1042 is designed to overlap with pin module 300 when module 1000 is fully pressed against pin module 300. Protruding insulating members 1042A and 1042B are sized to fit within spaces 342A and 342B, respectively. When the modules are pressed together, the distal ends of insulating members 1042A and 1042B press against surface 450 (Figure 3). These distal ends can have a shape that complements the tapered shape of surface 450, such that insulating members 1042A and 1042B fill spaces 342A and 342B, respectively. This overlap creates a relative positioning of signal conductors, dielectric, and reference conductors that is accessible to structures within sub-region 340. When modules 300 and 1000 are fully pressed together, these components can be sized to provide the same impedance as in sub-region 340. When the modules are fully pressed together, in this embodiment the nominal mating position, the signal conductors will have the same impedance across the mating area formed by sub-regions 340, 1040 and where sub-regions 342 and 1042 overlap.

[0095] These components may also be sized and may have material properties that provide impedance control as a function of the separation of modules 300 and 1000. Impedance control may be achieved by providing substantially the same impedance through sub-regions 342 and 1042 (even if those sub-regions do not completely overlap), or by providing a gradual impedance transition regardless of the separation of the modules.

[0096] In the illustrated embodiment, this impedance control is provided in part by protruding insulating members 1042A and 1042B, which fully or partially overlap module 300, depending on the separation between modules 300 and 1000. These protruding insulating members reduce the magnitude of changes in the relative dielectric value of the material surrounding the pins of pin module 300. Impedance control is also provided by protrusions 1020A and 1022A, and 1020B and 1022B, in reference conductors 1010A and 1010B. These protrusions affect the separation between portions of the signal conductor pair and reference conductors 1010A and 1010B in a direction perpendicular to the axis of the signal conductor pair. This separation, combined with other characteristics (such as the width of the signal conductors in those portions), controls the impedance in those portions so that it approaches the nominal impedance of the connector or does not abruptly change in a manner that could cause signal reflections. Other parameters of either or both of the paired modules may be configured for this impedance control.

[0097] Turning to FIG9 , further details of exemplary components of module 1000 are described. FIG9 is an exploded view of module 1000, not showing reference conductors 1010A and 1010B. In the illustrated embodiment, insulating housing portion 1100 is fabricated from multiple components. Central member 1110 may be molded from an insulating material. Central member 1110 includes two recesses 1212A and 1212B into which conductive elements 1310A and 1310B may be inserted. In the illustrated embodiment, conductive elements 1310A and 1310B form a pair of signal conductors.

[0098] Covers 1112 and 1114 can be attached to opposite sides of the core member 1110. Covers 1112 and 1114 can help retain the conductive elements 1310A and 1310B within the recesses 1212A and 1212B and maintain a controlled separation from the reference conductors 1010A and 1010B. In the illustrated embodiment, covers 1112 and 1114 can be formed from the same material as the core member 1110. However, the materials are not required to be the same, and in some embodiments, different materials can be used, such as to provide different relative permittivities in different regions to provide the desired impedance for the signal conductors.

[0099] In the illustrated embodiment, recesses 1212A and 1212B are configured to hold a pair of signal conductors for edge coupling at the contact tail and mating contact portions. The pair is held for broadside coupling over a substantial portion of the midsection of the signal conductors. To transition between edge coupling at the ends of the signal conductors and broadside coupling in the midsection, a transition region may be included in the signal conductors. The recesses in the center member 1110 may be shaped to provide transition regions in the signal conductors. Protrusions 1122, 1124, 1126, and 1128 on the covers 1112 and 1114 may press the conductive elements against the center member 1110 in these transition regions.

[0100] In the embodiment illustrated in FIG9 , the transition between broadside and edge coupling can be seen above region 1150. At one end of this region, the signal conductors are aligned edge-to-edge along the row direction in a plane parallel to the row direction. Towards the middle portion across region 1150, the signal conductors nudge in opposite directions perpendicular to that plane and toward each other. Thus, at the ends of region 1150, the signal conductors lie in a single plane parallel to the row direction. The middle portions of the signal conductors are aligned perpendicular to those planes.

[0101] Region 1150 includes a transition region, such as 822 or 842, where the waveguide formed by the reference conductor transitions from its widest dimension to a narrower dimension in the middle portion, plus a portion of the narrower middle region 830. Thus, at least a portion of the waveguide formed by the reference conductor in at least this region 1150 has a widest dimension, W, that is the same as in the middle region 830. Having at least a portion of the physical transition in the narrower portion of the waveguide reduces undesirable coupling of energy into the propagation modes of the waveguide.

[0102] Fully 360-degree shielding of the signal conductors in region 1150 can also reduce coupling of energy to undesirable waveguide propagation modes. Therefore, in the illustrated embodiment, opening 832 does not extend into region 1150.

[0103] In the illustrated embodiment, conductive elements 1310A and 1310B each have an edge and a wider side between their edges. Contact tails 1330A and 1330B are aligned in row 1340. With this alignment, the edges of conductive elements 1310A and 1310B face each other at contact tails 1330A and 1330B. Other modules in the same wafer will similarly have contact tails aligned along row 1340. Contact tails from adjacent wafers will be aligned in parallel rows. The spaces between the parallel rows form routing channels on the printed circuit board to which the connector is attached. Mating contact portions 1318A and 1318B are aligned along row 1344.

[0104] In the embodiment of FIG. 9 , conductive elements for a right-angle connector are illustrated, as reflected by the right angle between row 1340 representing attachment points to the daughter card and row 1344 representing the locations of mating pins attached to the backplane connector.

[0105] FIG10 shows further details of module 1000. In this view, conductive elements 1310A and 1310B are shown separated from core member 1110. For clarity, covers 1112 and 1114 are not shown. Transition region 1312A between contact tail 1330A and middle portion 1314A is visible in this view. Similarly, transition region 1316A between middle portion 1314A and mating contact portion 1318A is also visible. Similar transition regions 1312B and 1316B are visible for conductive element 1310B, allowing for edge coupling at contact tail 1330B and mating contact portion 1318B, as well as broadside coupling at middle portion 1314B.

[0106] The mating contact portions 1318A and 1318B can be formed from the same sheet metal as the conductive element. However, it should be understood that in some embodiments, the conductive element can be formed by attaching a separate mating contact portion to another conductor to form an intermediate portion. For example, in some embodiments, the intermediate portion can be a cable, such that the conductive element is formed by terminating the cable with the mating contact portion.

[0107] In the illustrated embodiment, the mating contact portion is tubular. This shape can be achieved by stamping a conductive element from sheet metal and then forming the mating contact portion into a tubular shape. The circumference of the tube can be large enough to accommodate pins from a mating pin module, but can mate with the pins. The tube can be divided into two or more sections to form a flexible beam. Two such beams are shown in Figure 10. Bumps or other protrusions can be formed on the distal ends of the beams to form contact surfaces. These contact surfaces can be coated with gold or other conductive, ductile materials to enhance electrical contact reliability.

[0108] When conductive elements 1310A and 1310B are installed in central member 1110, mating contact portions 1318A and 1318B fit within openings 1220A and 1220B. The mating contact portions are separated by wall 1230. Distal ends 1320A and 1320B of mating contact portions 1318A and 1318B can align with openings in platform 1232, such as opening 1222B. These openings can be positioned to receive pins from mating pin module 300. Wall 1230, platform 1232, and insulating protruding members 1042A and 1042B can be formed as part of portion 1110, for example, in a single molding operation. However, any suitable technique can be used to form these components.

[0109] FIG10 illustrates another technique that can be used, instead of or in addition to the techniques described above, to reduce energy in undesired propagation modes within the waveguide formed by the reference conductors in transition region 1150. Conductive or lossy materials can be integrated into each module to reduce the excitation of or dampen undesired modes. For example, FIG10 illustrates lossy region 1215. Lossy region 1215 can be configured to fall along the centerline between signal conductors 1310A and 1310B in some or all of region 1150. Because signal conductors 1310A and 1310B are nudge-twisted in different directions through the region to implement the edge-to-broadside transition, lossy region 1215 may not be bounded by surfaces parallel or perpendicular to the walls of the waveguide formed by the reference conductors. Rather, it can be contoured to provide surfaces that are equidistant from the edges of signal conductors 1310A and 1310B as they twist through region 1150. In some embodiments, lossy region 1215 can be electrically connected to a reference conductor. However, in other embodiments, lossy region 1215 can be floating.

[0110] Although illustrated as lossy region 1215, similarly positioned conductive regions can also reduce the coupling of energy into undesirable waveguide modes that degrade signal integrity. In some embodiments, such a conductive region having a surface that twists through region 1150 can be connected to a reference conductor. While not being bound by any particular theory of operation, a conductor that acts as a wall separating signal conductors and twists to follow the twist of the signal conductor in the transition region can couple ground current into the waveguide in this manner to reduce undesirable modes. For example, current can be coupled to flow in a different mode through the wall of the reference conductor parallel to the broadside-coupled signal conductor, rather than exciting a common mode.

[0111] FIG11 illustrates one embodiment of an expander module 1500 that can be used in an orthogonal connector. The expander module includes a pair of signal conductors having first mating contact portions 1510A and 1512A, and second mating contact portions 1510B and 1512B. The first and second mating contact portions are positioned at a first end 1502 and a second end 1504, respectively, of the expander module. As illustrated, the first mating contact portion is positioned along a first line 1550 that is orthogonal to a second line 1552, and the second mating contact portion is positioned along the second line. In the illustrated embodiment, the mating contact portions are shaped as pins and are configured to mate with corresponding mating contact portions of the connector module 810; however, it should be understood that other mating interfaces, such as beams, blades, or any other suitable structures, may also be used for the mating contact portions, as the present invention is not limited thereto. As described in more detail below, conductive shield elements 1520A and 1520B are attached to opposite sides of the expander module 1500 in the middle portion 1510 between the first end 1502 and the second end 1504. The shield elements surround the middle portion so that the signal conductors within the expander module are fully shielded.

[0112] Figures 12A-12C illustrate further details of signal conductors 1506 and 1508 disposed within expander module 1500. Insulated portions of the expander module are also visible, as shield elements 1520A and 1520B are not visible in these views. As shown in Figure 12A, the first and second signal conductors are each formed from a single piece of conductive material, with mating contact portions 1510 and 1512 connected by intermediate portions 1514 and 1516. The intermediate portion includes a 90° bend, such that the first mating portion is orthogonal to the second mating portion, as discussed above. Furthermore, as illustrated, the bends in the first and second signal conductors are offset so that the lengths of the two signal conductors are substantially the same; this configuration can advantageously reduce and / or eliminate skew in the differential signals carried by the first and second signal conductors.

[0113] Referring now to Figures 12B and 12C , the middle portions 1514 and 1516 of signal conductors 1506 and 1508 are disposed within insulating material 1518. First and second portions of insulating material 1518A and 1518B are formed adjacent to mating contact portions 1510 and 1512, and a third insulating portion 1522 is formed between the first and second portions around the middle portions of the signal conductors. While the insulating material is formed into three separate portions in the depicted embodiment, it should be understood that in other embodiments, the insulating material may be formed into a single portion, two portions, or more than three portions, as the present invention is not limited in this regard. Insulating portions 1518 and 1522 define orthogonal planar regions 1526 and 1528 on each side of the expander module to which conductive elements 1520A and 1520B are attached. Furthermore, the expander module is not required to be formed using the operations in the sequence illustrated in Figures 12A through 12C . For example, insulating portions 1518A and 1518B may be molded around signal conductors 1506 and 1508 before such conductive elements are bent at right angles.

[0114] FIG13 shows an exploded view of expander module 1500 and illustrates further details of conductive shield elements 1520A and 1520B. The shield elements are shaped to conform to insulating material 1518. As illustrated, first shield element 1520A is configured to cover the exterior surface of the expander module, and second shield element 1520B is configured to cover the interior surface. Specifically, the shield elements include first and second planar portions 1530A and 1530B, shaped to attach to planar regions 1526 and 1528, respectively, with the planar portions separated by a 90° bend 1532, such that the planar portions are orthogonal. The shield elements further include retaining clips 1534A and 1534B, and tabs 1536, each of which attaches to a corresponding feature on insulating material 1518 or an opposing shield element to secure the shield element to the expander module.

[0115] In the illustrated embodiment, conductive shield elements 1520A and 1520B include mating contact portions formed as four flexible beams 1538A...1538D. When assembled ( FIG. 11 ), two flexible beams 1538A and 1538B are adjacent to first end 1502 of extender module 1500; the other two flexible beams 1538C and 1538D are adjacent to second end 1504. Each pair of flexible beams is separated by an elongated notch 1540.

[0116] In some embodiments, the conductive shield elements 1520A and 1520B may have the same configuration at each end, such that the shield elements 1520A and 1520B may have the same shape but different orientations. However, in the illustrated embodiment, the shield elements 1520A and 1520B have different configurations at the first end 1502A and the second end, respectively, such that the shield elements 1520A and 1520B have different shapes. For example, as illustrated in FIG13 , the flexible beams 1538C and 1538D adjacent to the second end include fingers 1542 housed in corresponding pockets 1544. The fingers and pockets are constructed and configured to introduce a preload into the flexible beams, which can help provide a reliable mating interface. For example, the preload can cause the flexible beams to bend or flex outward from the expander module to facilitate mating contact as the second end of the expander module is received in the corresponding connector module.

[0117] Referring now to FIG. 14 , two identical expander modules 1900A and 1900B are illustrated rotated 180° relative to one another along each module's longitudinal axis. As described in greater detail below, the expander modules are shaped so that, when rotated in this manner, the two modules interlock to form expander module assembly 2000 ( FIG. 15A ). When interlocked in this manner, first planar portion 1926A and second planar portion 1928A on the first module are adjacent to and parallel to first planar portion 1926B and second planar portion 1928B, respectively, on the second module.

[0118] Figure 15A shows an expander module assembly including two expander modules 1900A and 1900B of Figure 14 . As illustrated, the mating portions of signal conductors 1910A...1910D and 1912A...1912D form two square arrays of mating contacts at the ends of the assembly. Figures 15B-15C illustrate schematic top and bottom views, respectively, of the square arrays, showing the relative orientation of the mating portions of each signal conductor in the expander module. In the illustrated embodiment, the assembly has a centerline 2002 parallel to the longitudinal axis of each expander module, with the center of each of the square arrays aligned with the centerline.

[0119] FIG16 illustrates one embodiment of an orthogonal connector 2100 during the manufacturing stage. Similar to daughter card connector 600, the orthogonal connector is assembled from connector modules and includes contact tails 2110 extending from a surface of the connector adapted for mounting to a printed circuit board. However, connector 2100 further includes a front housing 2140 adapted to accommodate a plurality of expander modules. The front housing also includes retention features 2150 that engage corresponding features on the expander housing 2300, as described below. As shown, the expander module assembly 2000 can be simply slid into the front housing to facilitate easy assembly of connector 2100.

[0120] FIG16 shows two interlocked expander modules inserted into a connector assembly. Inserting an interlocked pair of expander modules avoids the complexity of interlocking the expander modules after they have been inserted, but it should be understood that other techniques can be used to assemble the expander modules into the connector assembly. As another alternative embodiment, multiple pairs of expander modules can be inserted in a single operation.

[0121] FIG17A illustrates one embodiment of an expander housing 2300 for use with a direct-attach orthogonal connector. The expander housing has a first side 2302 adapted to attach to the front housing 2140 of the orthogonal connector 2100. As shown, the first side includes a cutout 2350 in an outer wall 2306 adapted to engage with a retaining feature 2150 on the front housing 2140. As discussed below, a second side 2304 of the expander housing is configured for detachable mating with a daughter card connector (e.g., a RAF connector). Furthermore, the expander housing includes mounting holes 2310 that can be used to connect the expander housing to additional components of the interconnect system, such as a printed circuit board. A cross-sectional view of the expander housing is shown in FIG17B. Similar to the backplane connector 200, the expander housing includes lossy or conductive separators 2320 and 2322 disposed in the first and second sides of the expander housing, respectively.

[0122] Referring now to Figures 18A-18B, direct-attach connector 2400 includes an orthogonal connector 2100 having a front housing 2140 adapted to mate with an expander housing 2300. A plurality of expander modules are configured into assembly 2000, wherein shielded signal contacts are positioned in a square array, and the first ends of the expander modules are received in the front housing. As illustrated, the expander housing is placed over the expander modules and then secured to form connector 2400; the connector includes a mating end 2410 that can be attached and mated with a connector, such as daughter card connector 600, on an orthogonal printed circuit board, as discussed below.

[0123] Figure 19 is a cross-sectional view of the assembled connector 2400. The mating end of expander module 1500 is received in corresponding connector modules 810A...810D on wafer 700. In the illustrated embodiment, the expander module is housed within an expander housing. Furthermore, the mating contact portion of the expander module that mates with the connector module is orthogonal to the mating contact portion, extending into the mating end 2410 of the connector, allowing the connector to function as a direct-connect orthogonal connector.

[0124] Figure 20 is a detailed view of the mating end 2410 of the connector 2400. The pins forming the mating contact portion of the expander module are organized into an array of differential signal pairs to form the mating interface. As discussed above, lossy or conductive separators 2320 separate the rows of signal pins.

[0125] FIG21 depicts one embodiment of an assembled orthogonal connector 2400, which can be directly attached to a RAF connector, such as daughtercard connector 600, via a detachable interface 2700. As shown, the contact tails 2210 of connector 2400 are oriented orthogonally to the contact tails 610 of daughtercard connector 600. In this manner, the printed circuit board (not shown for simplicity) to which the connectors can be attached via their contact tails can be oriented orthogonally. It should be understood that while one orthogonal configuration of connectors 2400 and 600 is depicted, in other embodiments, the daughtercard connectors can be rotated 180° to form a second orthogonal configuration. For example, the depicted configuration may correspond to a 90° rotation of connector 600 relative to connector 2400, and a second orthogonal configuration (not shown) may correspond to a 270° rotation.

[0126] An orthogonal configuration, illustrated in Figure 21, forms a signal path to a printed circuit board (PCB) via a connector attached to the edge of the board. The connector is attached to the board at a footprint. The trace "breaks" from that footprint and is routed to other points on the board where it connects to other components. In this way, signals can be routed through the connector to components anywhere on the board. However, in some cases, the board may be large enough that the distance between the connector and the component receiving the signal is long, for example, greater than 6 inches. These long traces may undesirably degrade the signal carried by the trace. Figure 22 illustrates a configuration that avoids long traces by facilitating the consolidation of cables to carry signals over long distances.

[0127] FIG22 illustrates an electrical interconnect system 2800 including connectors 2802, 2804, 2900, a cable 2806, and printed circuit boards (PCBs) 2801 and 2803. Connector 2900 may include a first-type connector unit 2902 and a second-type connector unit 2904. The first-type connector unit may be mounted directly to PCB 2801. The second-type connector unit may be coupled to PCB 2801 via cable 2806. In the illustrated embodiment, cable 2806 is connected to connector 2802, which in turn is connected to PCB 2801. In the illustrated embodiment, connector 2900 includes two types of connector units: in this embodiment, a plurality of first-type connector units and one second-type connector unit. However, the present invention is not limited in this respect. In some embodiments, the connector may include more than two types of connector units, for example, three, four, or more types. It should also be understood that the connector may include any suitable number of connector units of any type. Connector 2804 can be mounted to PCB 2803 at one end and mate with connector 2900 at the other end, such that PCB 2803 is orthogonal to PCB 2801.

[0128] Cable 2806 may have a first end 2808 attached to second-type connector unit 2904 and a second end 2810 attached to connector 2802, thereby coupling the cable to PCB 2801. The second end of the cable may be coupled to PCB 2801 at a distance D from the first end of the cable. Any suitable value may be selected for distance D. In some embodiments, D may be at least 6 inches, within a range of 1 to 20 inches, or any value within this range, such as between 6 and 20 inches. However, the upper limit of this range may depend on the size of PCB 2801 and the distance between components (not shown) and connector 2900 on PCB 2801, as connector 2802 may be mounted near components that receive or generate signals passing through cable 2806. As a specific example, connector 2802 may be mounted within 6 inches of these components, and in some embodiments, within 4 inches or within 2 inches of these components.

[0129] The inventors also recognize and appreciate that the footprint of connector 2900 on PCB 2801 is the combined footprint of the first-type connector unit and the second-type connector unit. When the first-type connector unit is mounted, the connector footprint occupies less substrate area on the circuit board than the combined footprint. This frees up space for routing circuit traces and / or placing hold-down screws to enhance retention. Alternatively or additionally, PCB 2801 may have fewer layers than required to route all signals from the first and second-type units out of the connector footprint. Furthermore, the restrictions on traces routed out of the footprint of the second-type connector unit may be less restrictive. The footprint of the second-type connector unit on PCB 2801 is defined by the mounting end 2812 of connector 2802. The mounting end of connector 2802 can be configured to facilitate routing with high signal integrity. For example, the footprint associated with mounting end 2812 can be spaced from the edge of PCB 2801, allowing traces to be routed out of that portion of the footprint in all directions. Furthermore, the illustrative second connector unit 2904 in FIG22 is attached with 12 cables aligned in a row at the first end. The cable rows are arranged in a 3×4 array at the second end attached to connector 2802. This configuration, or other configurations selected for mounting end 2812, can result in a relatively short breakout area, which maintains signal integrity when connecting to adjacent components, compared to the routing pattern that might be required for the same 12 signals from the first and second type units together to form a larger footprint.

[0130] Connector 2900 can mate with any suitable type of connector. In some embodiments, mating connector 2804 can be an orthogonal connector configured similarly to connector 600 illustrated in FIG. 4 . In this embodiment, connector 2900 can be a direct-attach orthogonal connector. However, the present invention is not limited in this respect, and connector 2900 can be used in a midplane configuration or any other suitable configuration. Furthermore, connectors with multiple types of units, such as connector 2900, can be used in other configurations, such as RAM or RAF configurations.

[0131] Figures 23A and 23B illustrate isometric views of connector 2900 from a mating interface 2920 and a mounting interface 2910, respectively. Connector 2900 may include an expander housing 2906 that holds both a first-type connector unit 2902 and a second-type connector unit 2904. The connector units may include signal conductors having mating contact portions 2912, contact tails 2914, and intermediate portions (not shown) extending between the contact tails and the mating contact portions. In the illustrated embodiment, the mating contact portions are shaped as pins. The pins are arranged to form pairs, with each pair extending parallel to direction 3006, and the pairs aligned in row direction 3002 to form an array of pairs. The expander housing may include a separator 2908. The pairs of mating contact portions in each row may be separated by the separator.

[0132] In this embodiment, the first-type connector unit 2902 includes a wafer 3008, which can be configured similarly to the wafer 700 illustrated in FIG. 5 . The rear portion of the wafer can be held by a support member 3014 . In the illustrated embodiment, the connector 2900 includes ten wafers 3008 . The wafer 3008 comprises six wafer modules held by a housing made of two halves 3018A and 3018B. Each module includes a pair of differential signal conductors 2916A and 2916B.

[0133] For example, as illustrated in FIG16 , the signal conductor pairs within each of wafers 3008 can be aligned in the row direction 3002. To achieve the orientation of the pins at the mating interface illustrated in FIG24A , an orthogonal expander module, such as expander module 2000 ( FIG16 ), can be attached to the mating interface of wafer 3008. FIG24A through FIG24E illustrate that connector 2900 can further include a plurality of expander modules 3010 attached to the mating end of wafer 3008. Expander modules 3010 can be configured similarly to expander module 1500 illustrated in FIG11 through FIG13 . Two identical expander modules 3010 can also form an expander module assembly 3012 similar to expander module 2000 illustrated in FIG15A through FIG15B .

[0134] A plurality of wafers and a plurality of expander modules can be held by one or more support members 3004. In the illustrated embodiment, support member 3004 is implemented as at least two separate components 2902A and 2902B. However, any suitable number and shape of components can be used to form the support member. For example, additional components can hold the wafers at opposing surfaces and / or sides of the illustrated structure. Alternatively or additionally, support member 3004 can be a housing having an opening for receiving and securing the wafers.

[0135] In the embodiment of Figure 24A, component 2902A holds six wafers, and component 2902B holds four wafers. The wafers held by 2902A are collectively attached to 24 expander modules 3010, and the wafers attached to component 2902B are collectively attached to 36 expander modules 3010. When each row of expander modules is attached to two wafers, those two wafers and the attached expander modules can be considered a first type of "unit," and any suitable number of such units can be used to form a connector.

[0136] However, it should be understood that each first-type connector unit can be a subassembly of any suitable number of components to implement any suitable number of rows of conductive elements, or can be implemented as a single component or implemented in any other suitable manner. Using the wafers and expander modules described above, each first-type connector unit can be formed from a multiple of two wafers (such as two, four, six, or eight wafers) and a multiple of that number of expander modules, where the multiple is equal to the number of signal conductors in a wafer, but the present application is not limited in this respect.

[0137] If multiple units are used, the connector units can be held together by a support member. In the illustrated embodiment, expander housing 2906 serves as the support member. Support member 3004 can include retention features 2950 that engage with corresponding features 2960 on expander housing 2906. However, it should be understood that support member 3004 can be omitted in some embodiments if the wafer is attached directly to expander housing 2906, or if other support structures are used to hold the connector components together.

[0138] In Figures 24A to 24E , the mating contact portion of wafer 3008 is covered by support member 3004 and is not shown. However, the mating contact portion can be configured similarly to the mating contact portion of wafer 700 illustrated in Figure 5 . Each wafer module of wafer 3008 can include a pair of different signal conductors. The mating contact portion of the wafer module can be configured as a receptacle adapted to receive the first mating contact portion of an expander module, which can be configured as pins. The wafer-molded mating contact portions in the wafer can be aligned along row 3002 . Adjacent wafer modules, each of the two wafers 3008 in the first-type connector unit 2902 can receive the first mating contact portion of expander module assembly 3012 . Thus, the second mating contact portion of the expander module assembly can form assembly 3202 , in which the differential signal conductor pairs can be aligned along row 3006 , which is perpendicular to row 3002 .

[0139] In the illustrated embodiment, there is one second-type unit 2904. To complement the first-type unit, the illustrative second-type unit 2904 includes twelve cables 2806 aligned in the direction of row 3002. Each second-type unit 2904 may include a plurality of modules 3100 held by a unit housing 3102. The plurality of modules in the second-type unit may be aligned in the direction of row 3002. Each module 3100 may include a module housing 3112 that holds a pair of signal conductors 3104A and 3104B. The pair of signal conductors are separated in the direction of row 3006. The mating contact portions of the second-type unit may form an array 3204. Arrays 3202 and 3204 together may form the mating interface 2920 of connector 2900.

[0140] The mating contact portions of the signal conductors are illustrated as pins. However, other configurations, such as receptacles, are possible. The contact tails of the signal conductors (not shown) are attached by cables 2806. The attachment interface between the contact tails and the cables is protected by at least the unit housing. Each cable may include a pair of wires, each of which is attached to a respective contact tail of a pair of signal conductors in the module. In some embodiments, the cables may be twinaxial cables. The shield surrounding the conductors of the twinaxial cables may be attached to the shield surrounding the conductive elements in the respective modules 3100. The unit housing 3102 may extend further along the length of the cable than the support member 3004, so that the attachment interface between the module 3100 and the cable 2806 is covered.

[0141] 25A-25B illustrate isometric views of a second-type connector unit with cables attached to form a cable assembly 3400. Cable assembly 3400 may include an assembly housing 3402 that holds a plurality of cable assembly modules 3420. Here, housing 3402 is made of two halves 3402A and 3402B that are secured together, capturing modules 3420 therebetween. These components may be held together using adhesives, interference fits, heat staking, or other suitable means.

[0142] The housing 3402 and the modules 3420 can form a second type connector unit. In the illustrated embodiment, each of the modules 3420 has a pair of signal conductors, and the modules 3420 are configured so that the second type connector unit has two rows of signal conductors.

[0143] 25C illustrates an isometric view of a cable assembly module 3420, which may include a module 3408 of a second-type connector unit 3404 and a cable 3406. Module 3408 may include a pair of signal conductors 3410A, 3410B held by a module housing 3412. Module 3408 may provide a mating interface that mates with the mating interface provided by each expander module used in forming the first-type connector unit.

[0144] The cable conductors can be attached to the signal conductors within module 3408 in any suitable manner. However, according to some embodiments, the cable conductors can be attached to the edges of the signal conductors to provide a conductive structure with substantially uniform thickness and / or substantially uniform spacing between conductive elements. For example, the thickness (including the thickness of the cable conductors, the signal conductors, and any welds, solders, or other materials used to weld the two) can be no more than 10% greater than the thickness of the stock material used to form the signal conductors. In some embodiments, the thickness variation between the cable attachment and the stock thickness can be less than 25% or less than 50%. More generally, the thickness variation can be less than that resulting from conventional methods of connecting the cable conductors to the connector signal conductors at their broadsides, which can increase the thickness of the conductive path by 100% or more. Similarly, the separation at the attachment location can be relatively small, such as no more than 10% or, in some embodiments, no more than 5% compared to the separation at the mating interface.

[0145] This connection is illustrated in Figures 26A and 26B. Figure 26A illustrates an isometric view of a pair of signal conductors 3410A and 3410B. Signal conductors 3410A and 3410B may represent signal conductors within module 3408 or any other cable connector. The signal conductors may include a contact tail 3510, a mating contact portion 3520, and an intermediate portion 3530 extending between the contact tail and the mating contact portion. The signal conductors may be slightly moved in opposite directions in a transition region 3514, resulting in a space s1 between the contact tails that differs from a space s2 between the intermediate portions and, in the illustrated embodiment, between the mating contact portions. In some embodiments, s1 may be greater than s2. The contact tail 3510 may include a broadside 3502 and an edge 3504 joining the broadsides. The pair of signal conductors can be held together with the contact tails in an edge-to-edge configuration, with edge 3504A of signal conductor 3410A facing edge 3504B of signal conductor 3410B. The mating contact portion 3520 can be configured as a pin. In some embodiments, the pin can be made from rolled metal sheet.

[0146] FIG26B illustrates an isometric view of a cable 3406 attached to the pair of signal conductors 3410A, 3410B. The cable 3406 may include a pair of conductive elements 3510A, 3510B insulated by a dielectric portion 3512. The cable 3406 may also include a shield surrounding the conductive elements 3510A, 3510B, which is not shown for simplicity. However, the shield may be attached to a shield or ground conductor in the cable connector.

[0147] A portion of the pair of conductive elements may be exposed outside the dielectric portion. The exposed portion of conductive element 3510A may be attached to edge 3504A of signal conductor 3410A. The exposed portion of conductive element 3510B may be attached to edge 3504B of signal conductor 3410B. Attachment may be performed in any suitable manner, such as by welding or brazing. For example, laser welding may be used. For example, a laser welding operation may be performed in which a laser is aimed along a path along the edge of a conductive element. As the laser's focus is changed, the wires in the cable are welded to the edge of the conductive element.

[0148] In some embodiments, the laser can be controlled to form a continuous, rounded joint between each conductive element of the cable and the edge of the signal conductor in the connector. The inventors have found that such joints can be more reliable and repeatable than soldering via wire. Commercially available green lasers can be used to form suitable welds, but any suitable welding equipment can be used.

[0149] Operations such as welding or brazing, which result in the direct fusion of the cable's conductive elements to the connector's conductive elements, can avoid the large amount of conductive material that may be present when using other attachment techniques (e.g., soldering). Reducing the amount of conductive material required for attachment can reduce changes in impedance, which can contribute to desirable electrical properties. However, in some embodiments, solder or other fusible materials may be added to facilitate attachment.

[0150] The cable conductors can be attached to the edges of any suitably shaped conductive element in the connector. Figures 27A through 27D illustrate a method for manufacturing a cable connector. Figure 27A illustrates a plan view of a mounting interface 3640 between structure 3630 and cable 3606. Figure 27B is an elevation view of mounting interface 3640, illustrating the relatively small additional thickness at the attachment location. Structure 3630 can include a pair of signal conductors 3610A and 3610B joined by a tie bar 3602. The contact tails of the signal conductors can be slightly moved in opposite directions and away from the tie bar through a transition region 3614. Structure 3630 can be stamped from sheet metal, allowing its dimensions to be precisely controlled by the stamping die.

[0151] Cable 3606 may include a pair of conductive elements 3620A, 3620B, each attached to one of the opposing edges of signal conductors 3610A, 3610B. The pair of signal conductors 3610A, 3610B are spaced apart by a distance d1 to accommodate cable 3606. Distance d1 can be controlled by the width W of tie bars 3602 and / or the degree of slope in transition region 3614. This distance can be precisely controlled by stamping.

[0152] FIG27C illustrates a plan view of the mounting interface 3642 between structure 3630 and cable 3606. FIG27C illustrates that insulating housing 3650 has been molded over structure 3630. Housing 3650 can be molded using an insert molding operation or any other suitable method. Connecting rod 3602 has then been severed. In this configuration, conductive elements 3610A and 3610B have been separated. However, the spacing between conductive elements 3610A and 3610B is maintained because both are embedded within housing 3650.

[0153] With connecting rod 3602 severed, mating contacts 3604A and 3604B on conductive elements 3610A and 3610B can be formed to provide any suitable shape. Any suitable metal forming technique can be used. For example, the edges can be stamped to provide the mating contacts as blades. Alternatively or additionally, the mating contacts can be rolled to provide the mating contacts as pins. As yet another variation, the mating contacts can be shaped as single-beam, dual-beam, or multi-beam contacts. As another alternative, separate components can be attached to conductive elements 3610A and 3610B, such as to form a multi-beam structure or to provide a socket.

[0154] The forming operation can separate the mating contacts 3604A and 3604B from each other by a distance d2 (measured edge to edge). In the illustrated embodiment, d2 can be approximately equal to d1. For example, d2 can differ from d1 by 10% or less, or in some embodiments, by 25% or 50% or less.

[0155] However, the spacing between the edges is not required to be uniform along the entire length of the contacts. The edges of the contacts at the attachment area may taper toward each other, or may taper toward each other along the elongated axis of the mating contacts 3604A and 3604B. This configuration can provide a gradual impedance transition from the cable to the mating interface of the connector. Alternatively or additionally, the shape of the conductive elements 3610A and 3610B may vary along their length to provide a wider or narrower width within the housing relative to the exterior. As another example of a variation, even if the opposing edges of the conductive elements 3610A and 3610B are shaped to provide a uniform spacing d2 along their length, the width of the conductive elements during attachment can be controlled by varying the profile of the outer edges of the conductive elements 3610A and 3610B, even varying the width along their length. For example, the outer edges may taper toward or away from each other.

[0156] Having thus described several embodiments, it will be appreciated that various alternatives, modifications and improvements will readily occur to those skilled in the art. Such alterations, modifications and improvements are intended to be within the spirit and scope of the present application.

[0157] Therefore, the above description and figures are by way of example only. Various modifications may be made to the illustrative structures shown and described herein. For example, a method for manufacturing a module for a cable connector is described with reference to Figures 27A through 27D. The method steps may be performed in an order other than that described. The cable conductors may be attached after the housing is formed and / or after the connecting rods are cut.

[0158] As another example, conductive elements in a cable (such as conductive elements 3620A and 3620B) can be shaped before or after being attached to signal conductors (such as signal conductors 3610A and 3610B), respectively. In some embodiments, the conductive elements in the cable can be mechanically deformed by pressing them against the signal conductors. This operation can strengthen the bond between the conductive elements and the signal conductors once they are fused. Alternatively or additionally, the conductive elements from the cable can be flattened to form opposing surfaces parallel to the broadsides of the signal conductors. This flattening operation can improve electrical performance by reducing variations in the thickness of the conductive paths. Alternatively or additionally, the conductive elements from the cable can be flattened to form one or more planar surfaces perpendicular to the broadsides of the signal conductors. This shape can facilitate attachment of the conductive elements to the signal conductors and / or provide more precise control over the separation between signal paths.

[0159] As another example, techniques for improving signal quality at the mating interface of an electrical interconnect system are described. These techniques may be used individually or in any suitable combination. Furthermore, the size of the connector may be increased or decreased from the size shown. Furthermore, it is possible to construct the connector using materials other than those explicitly mentioned. As another example, a connector having four differential signal pairs in a row is used for illustrative purposes only. Any desired number of signal conductors may be used in the connector.

[0160] As another embodiment, an embodiment is described in which a different front housing portion is used to hold the connector module in a daughter card connector configuration versus an orthogonal configuration. It should be understood that in some embodiments, the front housing portion can be configured to support either use.

[0161] Manufacturing techniques may also vary. For example, the embodiment described herein is one in which daughter card connector 600 is formed by assembling multiple wafers onto a stiffener. An equivalent structure may be formed by inserting multiple shields and signal receptacles into a molded housing.

[0162] As another example, a connector formed from modules is described, each of which contains a pair of signal conductors. Each module need not contain exactly one pair, nor must the number of signal pairs be the same across all modules in the connector. For example, modules with two or three pairs can be formed. Furthermore, in some embodiments, a core module can be formed that has two, three, four, five, six, or some greater number of rows in a single-ended or differential pair configuration. In embodiments where the connector is wafer-based, each connector or each wafer can include such a core module. To fabricate a connector with more rows than included in a base module, additional modules (e.g., each having a smaller number of pairs, such as a single pair per module) can be coupled to the core module.

[0163] Furthermore, while many of the inventive aspects are shown and described with reference to an orthogonal connector having a right-angle configuration, it should be understood that the various aspects of the present invention are not limited in this respect, as any of the inventive concepts (whether alone or in combination with one or more other inventive concepts) may be used with other types of electrical connectors, such as backplane connectors, daughterboard connectors, midplane connectors, cable connectors, stacking connectors, mezzanine connectors, I / O connectors, chip sockets, etc.

[0164] In some embodiments, the contact tails are described as press-fit "eye-of-the-needle" flexible portions designed to fit within through-holes in a printed circuit board. However, other configurations, such as surface mount components, spring contacts, solderable pins, etc., may also be used, as aspects of the present invention are not limited to using any particular mechanism for attaching the connector to the printed circuit board.

[0165] Furthermore, signal and ground conductors are described as having specific shapes. In the above embodiments, signal conductors are routed in pairs, with each conductive element of the pair having substantially the same shape to provide a balanced signal path. The signal conductors of the pair are positioned closer together than other conductive structures. Those skilled in the art will appreciate that other shapes can be used, and signal or ground conductors can be identified by their shape or measurable characteristics. In many embodiments, signal conductors can be narrow relative to other conductive elements, which can serve as reference conductors to provide low inductance. Alternatively or additionally, signal conductors can have a shape and position relative to wider conductive elements, which can serve as references to provide a characteristic impedance suitable for electronic systems, such as in the range of 50 to 120 ohms. Alternatively or additionally, in some embodiments, signal conductors can be identified based on the relative positioning of conductive structures used as shields. For example, a signal conductor can be substantially surrounded by a conductive structure that can serve as a shielding member.

[0166] Furthermore, the configuration of the connector module and expander module described above provides shielding of the signal path through the interconnection system formed by the connector module and expander module in the first connector and the connector module in the second connector. In some embodiments, slight gaps in the shielding components or spacing between shielding components may exist without materially affecting the effectiveness of the shielding. For example, in some embodiments, it may be impractical to extend the shielding to the surface of the printed circuit board so that there is a gap of approximately 1 mm. Despite the presence of such separations or gaps, such configurations can still be considered fully shielded.

[0167] Furthermore, the expander embodiment is described as a module having an orthogonal configuration. It should be understood that if the expander module has pins or blades at its second end, the expander module can be used to form a RAM without a 90-degree twist. Other types of connectors can alternatively be formed with a module having a socket or other configurations of mating contacts at the second end.

[0168] Furthermore, the expander module is described as forming a separable interface with the connector module. This interface may include gold plating or plating with some other metal or other material that prevents oxide formation. For example, this configuration may enable the same module used in the daughter card connector to be used with the expander module. However, the interface between the connector module and the expander module is not required to be separable. In some embodiments, for example, the mating contacts of the connector module or the expander module may generate sufficient force to scrape oxide from the mating contacts and form an airtight seal when mated. In this embodiment, the gold and other platings may be omitted.

[0169] Connectors configured as described herein can provide desired signal integrity properties over a frequency range of interest. The frequency range of interest can depend on the operating parameters of the system in which the connector is used, but can typically have an upper limit between approximately 15 GHz and 50 GHz, such as 25 GHz, 30 GHz, or 40 GHz, although higher or lower frequencies may be of interest in certain applications. Some connector designs may have a frequency range of interest that spans only a portion of this range, such as 1 GHz to 10 GHz, 3 GHz to 15 GHz, or 5 GHz to 35 GHz.

[0170] The operating frequency of an interconnect system can be determined based on the frequency range over which the interconnect can pass with acceptable signal integrity. Signal integrity can be measured according to a number of criteria depending on the application for which the interconnect system is designed. Some of these criteria may relate to signal propagation along a single-ended signal path, a differential signal path, a hollow waveguide, or any other type of signal path. Two examples of these criteria are signal attenuation along the signal path or reflections of the signal from the signal path.

[0171] Other criteria may relate to the interaction of multiple different signal paths. These criteria may include, for example, near-end crosstalk (NEXT), which is defined as the portion of a signal injected on a signal path at one end of an interconnect system that is measurable on any other signal path on the same end of the interconnect system. Another such criterion may be far-end crosstalk (FEXT), which is defined as the portion of a signal injected on a signal path at one end of an interconnect system that is measurable on any other signal path on the other end of the interconnect system.

[0172] As a specific example, it may be required that the signal path attenuation does not exceed 3 dB power loss, the reflected power ratio does not exceed -20 dB, and the individual signal paths contribute no more than -50 dB to the signal path crosstalk. Because these characteristics are frequency-dependent, the operating range of the interconnect system is defined as the frequency range that meets the specified criteria.

[0173] Described herein are designs for electrical connectors that can provide desired signal integrity for high-frequency signals, such as frequencies in the GHz range, including up to about 25 GHz or up to about 40 GHz or higher, while maintaining high density, such as spacing between adjacent mating contacts of about 3 mm or less, including, for example, spacing between adjacent contacts in a row between 1 mm and 2.5 mm or between 2 mm and 2.5 mm on center. The spacing between rows of mating contact portions can be similar, but the spacing between all mating contacts in a connector is not required to be the same.

[0174] Therefore, the present invention is not limited to the details of construction or the arrangement of components set forth in the following description and / or accompanying drawings. The various specific examples are provided for illustrative purposes only, and the concepts described herein can be practiced or implemented in other ways. Furthermore, the phraseology and terminology used herein are for descriptive purposes and should not be construed as limiting. The use of the words "include," "comprising," "having," "containing," or "involving," and variations thereof, herein, is intended to encompass the items listed thereafter (or their equivalents) and / or as additional items.

[0175] 200: Connector 210: Contact tail 220: Coordination interface 222: Shell 224A:Separator 224B:Separator 224C:Separator 226:Wall 228:Layer 230A: Column 230B: Column 230C: Column 230D: Column 240:Component 300:Pin module 314A:Signal conductor 314B:Signal conductor 316A: Contact tail 316B: Contact tail 320A: Reference conductor 320B: Reference conductor 322: Flexible component 326: Surface 328: Contact tail 340: Sub-area 342: Sub-area 342A: Space 342B: Space 450: Surface 600:Daughter card connector / connector 610: Contact tail 612: Support member 614: Support member 620: Coordination interface 630: Component 640: front housing part 700: Wafer 700A: Wafer 810A:Module 810B:Module 810C:Module 810D:Module 820: Contact tail area 822: Transition Area 830: Middle area 832: Open 840: Mating contact area 842: Transition Area 900: Component 910A: Component / Channel 910B: Component / Channel 910C: Channel 910D: Channel 920: column 930: hole 1000:Wafer module 1010A: Reference conductor 1010B: Reference conductor 1020A:Protrusion 1020B: protrusion 1022A: protrusion 1022B: protrusion 1040: Sub-area 1042: Sub-area 1042A: Protruding insulating member 1042B: Protruding insulating member 1100: Insulation shell part 1110: Central component 1112: Cover 1114: Cover 1122: protrusion 1124: protrusion 1126: protrusion 1128: protrusion 1150: Area 1212A: Groove 1212B: Groove 1215: Lossy area 1220A: Open 1220B: Open 1222B: Opening 1230: wall 1232:Platform 1310A:Signal conductor 1310B:Signal conductor 1312A: Transition Area 1312B: Transitional Area 1314A: Middle part 1314B: Middle part 1316A: Transition Zone 1316B: Transition Zone 1318A: Mating contact part 1318B: Mating contact part 1320A: Remote 1320B: Remote 1330A: Contact tail 1330B: Contact tail 1340: OK 1344: OK 1500: Expander module 1502: First End 1504: Second end 1506:Signal conductor 1508:Signal conductor 1510:Mating contact part 1510A: First mating contact portion 1510B: Second mating contact portion 1512A: First mating contact portion 1512B: Second mating contact portion 1514: Middle part 1516: Middle part 1518: Insulation material 1518A: Insulation material 1518B: Insulation material 1520A: Conductive components 1520B: Conductive components 1522: Third insulation part 1526: Orthogonal plane area 1528: Orthogonal plane area 1530A: First plane portion 1530B: First plane portion 1532:90° bending parts 1534A:Retaining clip 1534B:Retaining clip 1536: Tab 1538A: Flexible beam 1538B: Flexible beam 1538C: Flexible beam 1538D: Flexible beam 1540: Notch 1542: Finger 1544:bag 1900A: Expander Module 1900B: Expander Module 1910A:Signal conductor 1910B:Signal Conductor 1910C:Signal Conductor 1910D:Signal Conductor 1912A:Signal Conductor 1912B:Signal Conductor 1912C:Signal Conductor 1912D:Signal Conductor 1926A: First plane section 1926B: First plane section 1928A: Second plane section 1928B: Second plane section 2000: Expander module assembly 2002: Centerline 2100: Orthogonal Connector 2110:Contact tail 2140:Front housing 2150: Maintain characteristics 2300: Expander housing 2302: First side 2304: Second side 2306:Outer wall 2310:Mounting hole 2320: Lossy or conductive separators 2322: Lossy or conductive separators 2350:Incision 2400: Connector 2410:Mating end 2700: Interface 2800:Electrical Interconnection System 2801:Printed Circuit Board 2802: Connector 2803:Printed Circuit Board 2804:Connector 2806: Cable 2808: First End 2810: Second end 2812: Installation end 2900: Connector 2902: First type connector unit 2902A: Components 2902B: Components 2904: Second type connector unit 2906: Expander housing 2908:Separator 2910: Installation interface 2912:Mating contact part 2914:Contact tail 2916A: Differential signal conductor 2916B: Differential signal conductor 2920:Matching interface 2950: Maintain characteristics 2960: Features 3002: row direction 3004:Supporting member 3006: Direction 3008: Wafer 3010: Expander module 3012: Expander module assembly 3014:Supporting member 3018A: Half Body 3018B: Half body 3100:Module 3102:Unit housing 3104A:Signal Conductor 3104B:Signal Conductor 3112:Module housing 3202:Array 3204:Array 3400:Cable assembly 3402:Assembly housing 3402A: Half body 3402B: Half body 3404: Second type connector unit 3406: Cable 3408:Module 3410A:Signal Conductor 3410B:Signal Conductor 3412:Module housing 3420: Cable assembly module 3502: wide edge 3504:Edge 3504A:Edge 3504B:Edge 3510:Contact tail 3510A:Signal conductor 3510B:Signal Conductor 3512: Dielectric part 3514: Transition Zone 3520:Mating contact part 3530: Middle part 3602: Connecting rod 3604A:Mating contacts 3604B:Mating contact 3606: Cable 3610A:Signal Conductor 3610B:Signal Conductor 3614: Transition Area 3620A: Conductive components 3620B: Conductive components 3630:Structure 3640: Installation interface 3642: Installation interface 3650: Shell d1: distance d2: distance S1: Space S2: Space W: widest size

Claims

1. A connector comprising: a plurality of connector modules, each connector module including a conductor pair, each conductor having a contact tail, a mating end, and an intermediate portion connecting the contact tail and the mating end; and one or more support members holding the plurality of connector modules such that the plurality of connector modules are aligned in a plurality of rows, wherein: In each of the plurality of rows, each connector module includes a pair of conductors separated in a direction not parallel to the row, the pair of conductors having mating ends.

2. The connector as requested in item 1, wherein: The plurality of connector modules includes a plurality of first-type connector modules and a plurality of second-type connector modules, wherein the contact tails of the plurality of first-type connector modules extend beyond the bottom of the connector, and the contact tails of the plurality of second-type connector modules extend beyond the rear of the connector.

3. The connector as described in request item 2, wherein: The middle portions of the conductors of the plurality of first-type connector modules are bent such that, for each conductor, the mating end and the contact tail extend at a right angle, and the mating ends and the contact tails of the conductors of the plurality of second-type connector modules extend parallel to each other.

4. The connector of claim 1, further comprising: a plurality of biaxial cables attached to the plurality of connector modules, each biaxial cable comprising a pair of conductors, each conductor being attached to the contact end of the conductor pair of the respective connector module.

5. A connector comprising: a plurality of connector units, each of the plurality of connector units comprising at least one row of conductors, each conductor comprising a contact tail, a mating end, and an intermediate portion connecting the contact tail and the mating end; and one or more support members holding the plurality of connector units, wherein the mating ends form an array, wherein: Each of the plurality of connector units comprises a plurality of modules aligned in rows, and for each connector unit, each of the plurality of modules comprises a conductor pair separated in a direction not parallel to the rows, the conductor pair having mating ends.

6. The connector as requested in item 5, wherein, For each of the plurality of connector units: each of the plurality of modules includes a module housing that holds the conductor pairs separated from the direction which is not parallel to the direction in which the conductor pairs have the mating ends.

7. The connector as requested in item 5, wherein: The plurality of connector units include at least one first type connector module and at least one second type connector module, wherein the contact tail of the conductor of the at least one first type connector module is configured to be attached to a printed circuit board, and the contact tail of the conductor of the at least one second type connector module is configured to be attached to a cable.

8. The connector of claim 7, wherein the one or more support members hold the plurality of connector units, wherein the rows are aligned in parallel, and the mating ends of the at least one first type connector module form a first array, and the contact tails of the at least one second type connector module form a second array.

9. The connector as requested in item 7, wherein: The connector further comprises a plurality of biaxial cables, each biaxial cable being attached to one of the plurality of modules of the at least one second type connector module, and each of the plurality of biaxial cables having a pair of conductors, each conductor being attached to a contact end of a conductor of the module.

10. The connector as described in request item 9, wherein: The conductor pair of each of the plurality of biaxial cables is attached to the opposite edge of the contact tail of the individual conductor pair of the module.

11. The connector as requested in item 7, wherein: Each of the at least one first-type connector module contains two rows of conductors.

12. The connector as requested in item 7, wherein: The middle portion of the conductor of the at least one first-type connector module is bent at 90 degrees.

13. The connector of claim 5, wherein the mating end includes pins.

14. The connector of any one of claims 5 to 13, comprising: a plurality of cables, each cable comprising a first end attached to the contact tail of an individual conductor of the connector and a second end opposite to the first end and coupled to a printed circuit board.

15. The connector as described in request item 14, wherein: The second end of the cable is coupled to the printed circuit board at a distance of at least 6 inches from the first end of the cable.

Images