The present invention relates to connectors for high-speed signal transmission. More specifically, the present invention relates to connectors in which wires are directly connected to contacts of the connectors.
High-speed cable routing has been used to transmit signals between substrates, such as printed circuit boards (PCBs), of electronic devices. Conventional high-speed cable routing often requires routing in very tight and/or low-profile spaces. However, as data rates increase (i.e., as the frequency of the high-speed signal increases), the cost of high-performance high-speed transmission systems increases as well. High-speed signals transmitted between substrates generally follow a path of:
Conventional high-speed cable assemblies typically include two connectors (i.e., the second and third connectors listed above) that are connected by high-speed cables. Accordingly, conventional high-speed cable routing also requires two additional connectors (i.e., the first and fourth connectors listed above) to connect the high-speed cables to transmitting and receiving substrates.
The signal quality is affected every time the transmitted signal transfers from each of the listed items above. That is, the signal quality is degraded when the signal is transmitted between 1) the trace on the transmitting substrate and 2) the first connector mounted to the transmitting substrate, between 2) the first connector mounted to the transmitting substrate and 3) the second connector substrate that is inserted into the first connector, etc. The signal quality can even be affected within each of the items above. For example, a signal transmitted through the trace on the transmitting or receiving substrate can suffer significant insertion loss.
High-speed cable assemblies are relatively expensive, due in part to the cost high-speed cable and the two connectors that include substrates (i.e., the second and third connectors listed above). Each connector of the high-speed cable assembly also requires processing time. Thus, the full cost of a high-speed cable assembly cable includes the cable, the high-speed-cable-assembly connectors on each end of the cable, the processing time required for each of these connectors, and the area required on a substrate for each connector.
To reduce the overall size of the high-speed cable assembly, smaller connectors and cables have been attempted. However, using smaller connectors and cables can both increase the cost and reduce the performance of high-speed cable assemblies. Eliminating the high-speed cable assembly has been attempted by transmitting the signal only on substrates. However, signals transmitted on a substrate generally have higher insertion losses compared to many cables, including, for example, micro coaxial (coax) and twinaxial (twinax) cables. Thus, eliminating the high-speed cable assembly can result in reduced signal integrity and degraded performance.
Exotic materials and RF/Microwave connectors have been used to improve the performance of high-speed cable assemblies. However, such materials and connectors increase both the cost and the size of a high-speed cable assembly. Low-cost conductors, dielectrics, and connectors have been used to reduce the overall cost of systems that rely on high-speed cable routing. However, low-cost conductors, dielectrics, and connectors decrease the performance of high-speed cable assemblies and can also increase their size.
To overcome the problems described above, preferred embodiments of the present invention provide a high-speed cable assembly that is relatively small in size, cheap, and has high performance.
Preferred embodiments of the present invention provide a high-speed cable assembly with a low-profile connection to a substrate. Because the high-speed cable assembly connects perpendicularly or substantially perpendicularly to the substrate, zero keep-out space on the substrate is needed for slide insertion. Because there is no mating connector required on the substrate, the total amount of required system space, including on the substrate, is significantly reduced. The high-speed cable assembly also uses fewer connectors, resulting in fewer transitions in the signal transmission path. Fewer transitions simplifies the signal transmission path, improves system performance, and reduces costs.
According to a preferred embodiment of the present invention, a cable assembly includes a contact ribbon made of a single stamping including a plurality of pairs of first and second signal contacts; a ground plane; a first row of ground contacts extending from the ground plane in a row along a first side of the ground plane such that a first line extending through the first row of ground contacts does not intersect with any signal contacts of the plurality of pairs of first and second signal contacts; and a second row of ground contacts extending from the ground plane in a row along a second side of the ground plane such that a second line extending through the first row of ground contacts does not intersect with any signal contacts of the plurality of pairs of first and second signal contacts; and includes a cable including a plurality of pairs of first and second center conductors, each pair of the plurality of pairs of first and second center conductors is connected to a corresponding pair of the plurality of pairs of first and second signal contacts; a plurality of insulators each surrounding a corresponding pair of the plurality of pairs of first and second center conductors; and a shield that surrounds the plurality of insulators and that is connected to the ground plane.
The plurality of pairs of first and second signal contacts are preferably arranged in a single row. A first distance between the first row of ground contacts and the second row of ground contacts is preferably greater than a second distance between the single row of the plurality of pairs of first and second signal contacts and either of the first row of ground contacts or the second row of ground contacts. The first row of ground contacts and the second row of ground contacts are preferably located on the same side of the plurality of pairs of first and second signal contacts. Preferably, the contact ribbon is included in a housing, and a support member connecting the plurality of pairs of first and second signal contacts is removed from the contact ribbon after the contact ribbon is included in the housing.
The cable is preferably a twinaxial cable. The plurality of pairs of first and second signal contacts are preferably press-fit contacts or solderable contacts.
According to a preferred embodiment of the present invention, a method of manufacturing a cable assembly includes providing a contact ribbon including a plurality of pairs of first and second signal contacts; a ground plane; a first row of ground contacts extending from the ground plane in a row along a first side of the ground plane such that a first line extending through the first row of ground contacts does not intersect with any signal contacts of the plurality of pairs of first and second signal contacts; and a second row of ground contacts extending from the ground plane in a row along a second side of the ground plane such that a second line extending through the first row of ground contacts does not intersect with any signal contacts of the plurality of pairs of first and second signal contacts, providing a cable with a plurality of pairs of first and second center conductors, a plurality of insulators each surrounding a corresponding pair of the plurality of pairs of first and second center conductors, and a shield that surrounds the plurality of insulators, connecting each pair of the plurality of pairs of first and second signal contacts to a corresponding pair of the plurality of pairs of first and second center conductors at a first end of the cable, and connecting the shield to the ground plane at the first end of the cable.
Each pair of the plurality of pairs of first and second signal contacts is preferably connected to the corresponding pair of the plurality of pairs of first and second center conductors by crimping or soldering. The shield is preferably connected to the ground plane by soldering.
The method of manufacturing a cable assembly further preferably includes forming a housing for the contact ribbon before a support member connecting the plurality of pairs of first and second signal contacts is removed. Preferably, the housing includes at least one hole, and the support member is removed by punching or cutting the support member through the at least one hole of the housing.
The method of manufacturing a cable assembly further preferably includes attaching the cable assembly to a substrate before a support member connecting the plurality of pairs of first and second signal contacts is removed. Each signal contact of the plurality of pairs of first and second signal contacts is preferably connected to a corresponding hole in the substrate by soldering.
The plurality of pairs of first and second signal contacts are preferably press-fit contacts or solderable contacts. The plurality of pairs of first and second signal contacts are preferably arranged in a single row. A first distance between the first row of ground contacts and the second row of ground contacts is preferably greater than a second distance between the single row of the plurality of pairs of first and second signal contacts and either of the first row of ground contacts or the second row of ground contacts.
The first row of ground contacts and the second row of ground contacts are preferably located on a same side of the plurality of pairs of first and second signal contacts.
The above and other features, elements, steps, configurations, characteristics and advantages of the present invention will become more apparent from the following detailed description of preferred embodiments of the present invention with reference to the attached drawings.
Preferred embodiments of the present invention will now be described in detail with reference to
As shown in
Two immediately adjacent first and second contacts of the first linear array are positioned between the first line and the second line, do not touch the first or second lines, and do not overlap the first contacts of the first or second linear arrays or the second contacts of the first or second linear arrays. Two immediately adjacent third and fourth contacts of the first linear array are positioned between the second line and the third line, do not touch the second or third lines, and do not overlap the second contacts of the first or second linear arrays or the third contacts of the first or second linear arrays.
The two immediately adjacent first and second contacts of the first linear array are each spaced apart by a third distance that is less than a fourth distance between two immediately adjacent contacts in the second linear array or between two immediately adjacent contacts in the third linear array. The contacts on the first linear array may be arranged in a first group of two, three, four, five, six, seven etc. evenly spaced pairs of contacts adjacent to a first end of the contact ribbon 10, a second group of two, three, four, five, six, seven, etc. evenly spaced pairs of contacts adjacent to a second end of the contact ribbon 10, and a distance between the first and second groups that is larger than the first distance. The first contact of the two immediately adjacent first and second contacts of the first linear array and the first contact of the second linear array both lie along a first cross-array line that forms an acute angle with the first line. The acute angle can be 1 to 89 degrees with 45 degrees preferred, the second contact of the two immediately adjacent first and second contacts of the first linear array and the second contact of the second linear array both lie along a second cross-array line that forms an acute angle with the second line. The first linear array can be signal conductors arranged into differential signal pairs, and the second and third linear arrays can be ground shield tails attached to one or more ground shields. The number of contacts in the first linear array is greater than the number of contacts in the second linear array. The number of contacts in the second and third linear arrays can be equal. For example, the first linear array can include sixteen contacts arranged into two groups of differential signal pairs, while the second or third linear arrays can each include ten contacts.
As shown in
As shown in
During the overmolding of the contact ribbon 10, both sides of each contact 12, 13 can be stabilized so that the contacts 12, 13 cannot move while the plastic is being injected around the contacts 12, 13, which can improve mechanical and electrical performance of the contacts 12, 13. Stabilizing the contacts 12, 13 can create void cores in the lower connector housing 31. These void cores can lower the dielectric constant in the region where the contacts 12, 13 are exposed to air. The void cores can be located where the cable 20 is attached to the contacts 12, 13. When the center conductors 22, 23 are soldered to the contacts 12, 13 at the void cores, the air gaps created by the void cores lower the dielectric constant while the solder balances out the local impedance with added capacitance.
Instead of using overmolding for the lower connector housing 31, any housing can be used that allows the tie bars 14 between the first contacts 12 and second contacts 13 to be removed. Such housings include, for example, pre-molded, snap-on, sonically welded, screwed-on, and glued housings. However, overmolding is preferred for the lower connector housing 31 because of its simplicity and because it is easier for a tool to remove the tie bars 14. Preferably, the lower connector housing 31 is made of plastic, for example, acrylonitrile butadiene styrene (ABS) plastic.
As shown in
The shield 21 and the first and second center conductors 22 and 23 are the conductive elements of the twinaxial cable 20. The first and second center conductors 22 and 23 are arranged to carry electrical signals, whereas the shield 21 typically provides a ground connection. The shield 21 also provides electrical isolation for the first and second center conductors 22 and 23 and reduces crosstalk between neighboring pairs of the first and second center conductors 22 and 23 and between the conductors of any neighboring cables.
The first and second center conductors 22 and 23 preferably have cylindrical or substantially cylindrical shapes. However, the first and second center conductors 22 and 23 could have rectangular or substantially rectangular shapes or other suitable shapes. The first and second center conductors 22 and 23 and the shield 21 are preferably made of copper. However, the first and second center conductors 22 and 23 and the shield 21 can be made of brass, silver, gold, copper alloy, any highly conductive element that is machinable or manufacturable with a high dimensional tolerance, or any other suitable conductive material. The insulator 24 is preferably formed of a dielectric material with a constant or substantially constant cross-section to provide constant or substantially constant within manufacturing tolerances electrical properties for the conductors 22 and 23. The insulator 24 could be made of TEFLON™, FEP (fluorinated ethylene propylene), air-enhanced FEP, TPFE, nylon, combinations thereof, or any other suitable insulating material. The insulator 24 preferably has a round, oval, rectangular, or square cross-sectional shape, but can be formed or defined in any other suitable shape. The jacket 25 protects the other layers of the twinaxial cable 20 and prevents the shield 21 from coming into contact with other electrical components to significantly reduce or prevent occurrence of an electrical short. The jacket 25 can be made of the same materials as the insulator 24, FEP, or any suitable insulating material.
As shown in
Other types of cables, such as coaxial cables, can be used in place of the twinaxial cable 20. In addition, the twinaxial cable 20 can be provided as a ribbonized twinaxial cable, and the ribbonized twinaxial cable can include a single shield that surrounds more than one pair of first and second center conductors 22 and 23.
As shown in
As similarly shown in
Two immediately adjacent first and second PTHs or solder pads of the first linear array are positioned between the first line and the second line, do not touch the first or second lines, and do not overlap the first PTHs or solder pads of the first or second linear arrays or the second PTHs or solder pads of the first or second linear arrays. Two immediately adjacent third and fourth PTHs or solder pads of the first linear array are positioned between the second line and the third line, do not touch the second or third lines, and do not overlap the second PTHs or solder pads of the first or second linear arrays or the third PTHs or solder pads of the first or second linear arrays.
The two immediately adjacent first and second PTHs or solder pads of the first linear array are each spaced apart by a third distance that is less than a fourth distance between two immediately adjacent PTHs or solder pads in the second linear array or between two immediately adjacent PTHs or solder pads in the third linear array. The PTHs or solder pads on the first linear array may be arranged in a first group of two, three, four, five, six, seven etc. evenly spaced pairs of PTHs or solder pads adjacent to a first end of the connector footprint, a second group of two, three, four, five, six, seven, etc. evenly spaced pairs of PTHs or solder pads adjacent to a second end of the connector footprint, and a distance between the first and second groups that is larger than the first distance. The first PTH or solder pad of the two immediately adjacent first and second PTHs/solder pads of the first linear array and the first PTH or solder pad of the second linear array both lie along a first cross-array line that forms an acute angle with the first line. The acute angle can be 1 to 89 degrees with 45 degrees preferred, the second PTH or solder pad of the two immediately adjacent first and second PTHs/solder pads of the first linear array and the second PTH or solder pad of the second linear array both lie along a second cross-array line that forms an acute angle with the second line. The first linear array can be signal conductors arranged into differential signal pairs, and the second and third linear arrays can be ground shield tails attached to one or more ground shields. The number of PTHs/solder pads in the first linear array is greater than the number of PTHs/solder pads in the second linear array. The number of PTHs/solder pads in the second and third linear arrays can be equal. For example, the first linear array can include sixteen PTHs/solder pads arranged into two groups of differential signal pairs, while the second or third linear arrays can each include ten PTHs/solder pads.
Preferably, the completed connector is press fit to the substrate 40 using a press-fit tool. The press-fit tool is preferably a simple tool, including, for example, a flat block attached to an arbor press, a tool with a cavity that aligns with the housing, a tap hammer, etc. That is, it is not necessary to use an expensive tool to transfer a force directly and individually to the back of each of the contacts 11, 12, and 13. Typically, the completed connector is only mated to the substrate 40 once; however, it is possible to unmate the completed connector and the substrate 40 and then to re-mate the completed connector and the substrate 40, if desired. For example, it is possible to remove the press-fit contacts 11, 12, and 13.
According to the preferred embodiments of the present invention, the first contacts 12 and the second contacts 13 are offset from ground plane 15, as shown in
Also, according to the preferred embodiments of the present invention, the first contacts 12 and the second contacts 13 are aligned in a single row, such that the overall length of the transmission for each signal is the same or substantially the same, within manufacturing tolerances. This provides “balanced” contacts with a relatively consistent characteristic impedance and low cross-talk. Preferably, the preferred embodiments of the present invention allow for communication to be performed at about 20 GHz or more, for example. In addition, the center conductors 22 and 23 of the twinaxial cable 20 preferably transmit a differential signal.
According to the preferred embodiments of the present invention, the completed connector can be used to connect the twinaxial cable to different points on the substrate 40, or to connect the substrate 40 to another substrate or to an electronic device. For example, as shown in
As another example, in an edge-to-edge application, the substrate 40 can be connected to a substrate that is co-planar or substantially co-planar and aligned along a common edge. As another example, in a right-angle application, the substrate 40 can be connected to a substrate that is perpendicular or substantially perpendicular. According to a further example, in a board-to-board application, the substrate 40 can be connected to a substrate that is parallel or substantially parallel, but not coplanar, for example, when the surfaces of the substrates are facing each other. As yet another example, in a board-to-edge-card application, one end of the completed connector can be connected to a relatively large substrate, such as a computer motherboard, while another end of the completed connector is connected to a relatively small edge-card.
The cable assemblies of the preferred embodiments of the present invention achieve a simulated insertion loss of about −1 dB at frequencies up to and including about 23 GHz and a return loss at or under −20 dB at frequencies up to about 25 GHz. The cable assembly of the preferred embodiments of the present invention achieves power sum far end crosstalk (PSFEXT) of approximately −40 dB at frequencies up to and including 10 GHz. The cable assemblies of the preferred embodiments of the present invention achieve an integrated crosstalk noise (ICN) between 5.6 and 7.5 at a frequency of about 14 GHz for all measured differential pairs. The term “about” refers to measurement tolerances. For example, a frequency of “about 30 GHz” refers to a frequency that is measured to be 30 GHz within measurement tolerances.
While preferred embodiments of the present invention have been described above, it is to be understood that variations and modifications will be apparent to those skilled in the art without departing the scope and spirit of the present invention. The scope of the present invention, therefore, is to be determined solely by the following claims.
This application claims the benefit of U.S. Provisional Application No. 62/376,765 filed Aug. 18, 2016, which is hereby incorporated by reference in its entirety.
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/US2017/043204 | 7/21/2017 | WO | 00 |
Number | Date | Country | |
---|---|---|---|
62376765 | Aug 2016 | US |