Field of the invention

The invention relates to methods of forming circuit board assemblies with an RF and DC transition between two boards and a circuit board assembly having an RF and DC transition between two boards.

Background to the invention

The invention relates to communicating RF signals between printed circuit boards (PCBs).

It is known to communicate RF signals between PCBs using solder connections. Connections of this type can have significant losses, especially at high frequencies of greater than 10 GHz.

The invention seeks to provide improved ways of transmitting high frequency RF signals between PCBs.

Summary of the invention According to an aspect of the present disclosure, there is provided a method of forming a circuit board assembly, the circuit board assembly comprising a first PCB, a second PCB, an RF signal connector for carrying RF signals therebetween, and one or more ground connectors for at least partially shielding the RF signal connector. The method comprises: providing first and second PCBs, each having an RF signal conductor and one or more ground conductors thereon; forming or attaching an RF signal connector extending between the first and second PCBs and electrically connecting the RF signal connector to the RF signal conductors of the first and second PCBs; and forming or attaching one or more ground connectors extending between the first and second PCBs and electrically connecting the one or more ground connectors to the ground conductors of the first and second PCBs. The one or more ground connectors are together arranged to at least partially surround the RF signal connector. Thus, transmission loss of the RF signal during transmission of the RF signal between the first PCB and the second PCB is low whilst being simple to manufacture.

The method may further comprise forming or attaching one or more DC connectors extending between the first and second PCB. The one or more DC connectors may be electrically connecting DC connections of the first PCB and the second PCB. The one or more DC connectors may be shielded from the RF signal connector by the one or more ground connectors which at least partially surround the RF signal connector.

It may be that the circuit board assembly formed by the methods described herein is for a microelectromechanical system, MEMS, based antenna assembly. The MEMS based antenna assembly may be a MEMS based tuneable array antenna assembly. The MEMS based tuneable array antenna assembly may be, for example, an antenna module. The MEMS based tuneable array antenna assembly may be used to distribute electromagnetic (such as radio frequency) signals, with a wavelength of below 50 millimetres, through a phase shifter. Advantageously, using MEMS devices in the MEMS based antenna assembly reduce RF losses, thereby reducing overall losses. The MEMS based tuneable antenna array may use MEMS components to tune the frequency, phase, magnitude or impedance within the said antenna array. The MEMS based antenna assembly may be a phased array antenna assembly.

Typically, the RF signal connector and the one or more ground connectors are provided in a connector assembly extending between the first and second PCBs and defining one or more openings therein. Each opening may be adjacent to at least one of the first PCB and the second PCB. The first PCB may extend in a first plane in a first direction and a second direction perpendicular to the first direction. The second PCB may extend in a second plane in the first direction and the second direction. The first plane and the second plane may be substantially parallel to one another. The first and second PCB may be spaced apart from one another in a third direction perpendicular to the first plane and the second plane. The connector assembly may be positioned between the first and second PCB and extend in the first direction, the second direction and the third direction. The one or more openings may also extend in the first direction, the second direction and the third direction within the connector assembly.

It may be that the circuit board assembly further comprises one or more (e.g. a plurality of) signal connectors extending between the first and second PCBs. The plurality of signal connectors may be a power signal connector, for example, the plurality of signal connectors may be DC power signal connectors. It will be appreciated that the one or more signal connectors may be the one or more DC connectors discussed above. The plurality of signal connectors may be arranged around the periphery of the one or more openings. The plurality of signal connectors may be connected to a plurality of signal conductors on the first and second PCBs.

Typically, the first PCB comprises a circuit component extending from a surface of the first PCB, towards the second PCB. It may be that the connector assembly is arranged such that at least one of the one or more openings accommodates the circuit component therein. The surface of the first PCB from which the circuit component extends may be a surface of the first PCB facing the connector assembly. It may be that the second PCB comprises a circuit component extending from a surface of the second PCB, towards the first PCB. Both the first PCB and the second PCB may comprise a circuit component extending from their respective surfaces. The one or more openings may each accommodate one or more circuit components, for example an integrated circuit, IC. Typically, the at least one of the one or more openings accommodates a multi-chip module therein. It will be understood that the circuit component is typically a circuit component electrically connected to one or more further components on the respective first or second PCB, via the respective first or second PCB.

The above method forms a circuit board assembly with a partially shielded RF signal connector in a connector assembly with openings to accommodate a circuit component of a first PCB. Therefore, the above circuit board assembly is more compact than other known circuit board assemblies because the connector assembly accommodates circuit components of the first PCB within the connector assembly, thereby reducing the footprint of the circuit board assembly. The provision of an RF signal connector partially surrounded by one or more ground connectors reduces transmission loss of RF signals conducted between the first PCB and a second PCB in a MEMS based phased array antenna. Advantageously, the circuit board assembly disclosed herein reduces transmission losses of RF signals in a phased array antenna whilst also reducing the size of the circuit board assembly.

The one or more ground connectors extending between the first and second PCB arranged to at least partially surround the RF signal connector may be a first group of the one or more ground connectors.

It may be that the connector assembly shields the circuit component accommodated in the one or more openings from: signal interference originating external to the connector assembly and/or signal interference originating external to the opening in which the circuit component is accommodated.

It may be that the circuit board assembly comprises a second group of one or more ground connectors extending between the first PCB and the second PCB. The second group of the one or more ground connectors may be arranged on a peripheral edge (i.e. around the periphery) of the one or more openings. The second group of the one or more (e.g. a plurality of) ground connectors may be together arranged to shield the circuit component accommodated in the one or more openings from: signal interference originating external to the connector assembly and/or signal interference originating external to the opening in which the circuit component is accommodated. For example, the second group of the one or more ground connectors may function as a faraday cage within the said opening.

It may be that least one of the one or more openings extends through an entire depth of the connector assembly. In this way, a circuit component accommodated within an opening could extend from the surface of the first PCB to substantially the surface of the second PCB facing towards the connector assembly. It may be that at least one of the one or more openings do not extend through an entire depth of the connector assembly (i.e. the depth of these openings is a portion of the depth of the connector assembly). In this way, the one or more openings may be depressions in the connector assembly, such that the one or more openings are open at only one end to accommodate the circuit component. Typically, the depth of the connector assembly is equal to or less than 1 .5 millimetres. It may be that the depth of the connector assembly is equal to or less than 1 millimetre. The depth of the connector assembly may correspond to the distance between the first PCB and the second PCB in the third direction (i.e. the direction perpendicular to the plane of the first and second PCBs). Advantageously, the depth of the connector assembly is short enough so as to reduce the footprint of the connector assembly, but is long enough to accommodate circuit components of the first PCB extending within the one or more openings of the connector assembly.

It may be that the one or more ground connectors and the RF signal connector are arranged at least 50 micrometres away from a peripheral edge of the connector assembly. It may be that the one or more ground connectors and the RF signal connector are arranged at least 1 millimetre away from a peripheral edge of the connector assembly. Advantageously, the provision of space between the peripheral edge of the connector assembly and the one or more ground connectors and the RF signal connector improves the structural stability of the circuit board assembly, because the peripheral edge of the connector assembly acts as a lateral stabiliser for the one or more ground connectors and the RF signal connector.

Typically, the connector assembly has a length of 15 millimetres to 30 millimetres and/or a width of 15 millimetres to 30 millimetres. The connector assembly may have a length and/or width of 18 millimetres to 25 millimetres. The connector assembly may have a length and/or width of 30 millimetres to 100 millimetres. The length of the connector assembly may extend in the first direction and the width of the connector assembly may extend in the second direction. The connector assembly may have the same length and the first and/or second PCB or may have a shorter length than the first and/or second PCB. The connector assembly may have the same width as the first and/or second PCB or may have a shorter width than the first and/or second PCB.

It may be that at least one of the one or more openings have a length equal to or greater than 5 millimetres and/or a width equal to or greater than 5 millimetres. It may be that at least one of the one or more openings have a length equal to or greater than 3 millimetres and/or a width equal to or greater than 3 millimetres. The one or more openings may have a length equal to or greater than a size of the circuit component to be accommodated therein. The length of the opening may refer to the first direction and the width of the opening may refer to the second direction. It may be that the method of forming the circuit board assembly comprises connecting the connector assembly to at least one of the first PCB and the second PCB using a plurality of solder connections. Typically, at least one of the plurality of solder connections may contact the respective first or second PCB over an area at least 25 microns squared (i.e. 5x10' ⁶ metres by 5x10' ⁶ metres, 25x10' ¹² m ² or 2.5x10' ¹¹ m ² or 0.000025 mm ² ). The area of contact between a solder connection and the first or second PCB may be a contact area through which the RF signal is conducted. At least one of the plurality of solder connections may contact the respective first or second PCB over an area less than 1 millimetre squared (i.e. 1x10' ³ metres by 1x10' ³ metres, 1x10' ⁶ m ² ). Typically, at least one of the plurality of solder connections may contact the respective first or second PCB over an area with a length between 80 microns to 200 microns.

Typically, the area of contact between a solder connection and the first or second PCB may be substantially square. Typically, the area of contact between a solder connection and the first or second PCB may be substantially circular. Typically, the area of contact between a solder connection and the first or second PCB may be substantially hexagonal. Typically, the area of contact between a solder connection and the first or second PCB may be a polygon of at least 4 sides. Typically, the area of contact between a solder connection and the first or second PCB may be a polygon of at least 10 sides. Typically, the area of contact between a solder connection and the first or second PCB may be a polygon of at least 16 sides.

The solder connection may be between the RF signal connector and the signal conductor of the first and/or second PCB. The solder connection may be between the one or more ground connectors and the ground conductor of the first and/or second PCB. The plurality of solder connections may be between the plurality of signal connectors and the first and/or second PCBs.

Advantageously, using solder connections with a small contact area having a length equal to or less than 100 microns increases the mechanical rigidity of the connector assembly. For comparison, a large contact area of a solder connection may be considered as a solder connection having a length equal to or greater than 400 microns. In addition, using solder connections with the dimensions discussed above reduces radiated losses of the RF signals compared to solder connections with a larger contact area. Typically, the RF signal connector is configured to conduct RF signals between the first and second PCBs. The RF signals may have a frequency equal to or greater than 1 GHz. The RF signals may be RF signals with a frequency equal to or greater than 6GHz. The RF signals may have a frequency equal to or greater than 8 GHz. The conducted RF signal may have a frequency equal to or greater than 10 GHz. The RF signals conducted between the first and second PCBs may have a wavelength of less than or equal to 500 millimetres. The RF signals may have a wavelength of less than or equal to 50 millimetres. The RF signals may have a wavelength of less than or equal to 30 millimetres.

It may be that the plurality of signal connectors are arranged to be shielded from the RF signal connector by the one or more ground connectors. That is, one or more ground connectors are arranged to, at least partially, surround the RF signal connector and the plurality of signal connectors are arranged on the opposite side of the one or more ground connectors to the RF signal connector. Advantageously, this arrangement reduces transmission loss of signal conducted along the plurality of signal connectors.

Typically, the connector assembly may comprise a first opening and a second opening. Both the first and second openings may be defined by a central portion of the connector assembly. That is, a first edge of the central portion may define the first opening and a second edge of the central portion may define the second opening. It may be that the one or more ground connectors and the RF signal connector are arranged on the central portion of the connector assembly. The central portion of the connector assembly may be a portion of the connector assembly extending in the first direction or the second direction. The RF signal connector may be positioned on the central portion with the one or more ground connectors arranged to at least partially surround the RF signal connector on the central portion of the connector assembly

It may be that the plurality of signal connectors are arranged around a periphery of the first and second openings and along the central portion of the connector assembly. For example, it may be that a first portion of the plurality of signal connectors are arranged on at least a part of the central portion. The first portion of the plurality of signal connectors may extend along the entire length of the central portion or may extend along 50% of the length of the central portion. The first portion of the plurality of signal connectors may be arranged along the central portion on either side of the RF signal connector at least partially surrounded by the one or more ground connectors. A second portion of the plurality of signal connectors may be arranged around the periphery of the first and/or second openings. The first opening may be formed by a first outer wall of the connector assembly along which a second portion of the plurality of signal connectors may be arranged. The second opening may be formed by a second outer wall of the connector assembly along which a third portion of the plurality of signal connectors may be arranged.

It may be that the second group of the one or more ground connectors are arranged around a periphery of the first and second openings and along the central portion of the connector assembly. For example, it may be that a first portion of the second group of the one or more ground connectors are arranged on at least a part of the central portion. The first portion of the second group of the one or more signal connectors may extend along the entire length of the central portion or may extend along 50% of the length of the central portion. The first portion of the second group of the one or more ground connectors may be arranged along the central portion on either side of the RF ground connector at least partially surrounded by the first group of the one or more ground connectors. A second portion of the second group of the one or more ground connectors may be arranged around the periphery of the first and/or second openings. The first opening may be formed by a first outer wall of the connector assembly along which a second portion of the second group of the one or more ground connectors may be arranged. The second opening may be formed by a second outer wall of the connector assembly along which a third portion of the second group of one or more ground connectors may be arranged.

Advantageously, this arrangement of the second group of the one or more ground connectors at least partially isolates the circuit components accommodated in the first and second openings from signals external to the opening in which the circuit component is accommodated. In this way, the second group of the one or more ground connectors functions as a faraday cage. It may be that the arrangement of the second group of the one or more ground connectors at least partially isolates the circuit component in the first opening from the second opening. It may be that the arrangement of the second group of the one or more ground connectors at least partially isolates the circuit components in the second opening from the first opening.

Typically, the method of forming the circuit board assembly may comprise flooding the one or more openings with filling material once the circuit component is accommodated in the one or more openings. Advantageously, the filling material may protect the circuit components and increase the structural rigidity of the assembly. In some examples, the filling material may be considered to substantially encase the one or more circuit components and to shield the one or more encased circuit components from RF interference.

At least one of the RF signal connector and the one or more ground connectors may each be formed as a multi-layer structure. In other words, a first layer of the RF signal connector may be formed separately to a second layer of the RF signal connector. For example, the first layer of the RF signal connector may be formed in a first processing step. Subsequently, the second layer of the RF signal connector may be formed on the first layer in a separate processing step. Alternatively, the RF signal connector may be formed in a single processing step. A first layer of the one or more ground connectors may be formed separately to a second layer of the one or more ground connectors. For example, the first layer of the one or more ground connectors may be formed in a first processing step. Subsequently, the second layer of the one or more ground connectors may be formed on the first layer in a separate processing step. Alternatively, the one or more ground connectors may (each) be formed in a single processing step.

The RF signal connector and the one or more ground connectors may be provided in a connector assembly extending between the first and second PCBs. The connector assembly may define one or more openings (e.g. depressions) therein, each adjacent one of the first PCB and the second PCB. Thus, the connector assembly can be formed to accommodate one or more components, each mounted on a one of the first PCB and the second PCB, extending in a direction towards the other of the first PCB and the second PCB, even when the components are provided close to the RF signal connectors and the one or more ground connectors. Thus, a more compact circuit board assembly can be provided. In some examples, it may be that at least one of the one or more openings extend wholly through the connector assembly from the first PCB to the second PCB. In other words, the opening may be provided by a through- channel (e.g. a through-bore). The method may comprise filling any remaining free space of the one or more openings of the connector assembly after installing the connector assembly between the first PCB and the second PCB. Thus, the opening can remain only as large (e.g. as wide, long and/or deep) as necessary for accommodating any components which extend from the first or second PCBs into the one or more openings. There may be at least one opening extending from the first PCB to the second PCB. There may be at least two openings extending from the first PCB to the second PCB.

At least one of the one or more openings, may be wholly surrounded by a boundary portion of the connector assembly, the boundary portion extending fully to one or both of the first PCB and the second PCB. In other words, the one or more openings are typically features of the connector assembly, inwards of an outer edge of the connector assembly.

The connector assembly may define a substantially rectilinear external cross-section, in a plane parallel to the first and/or second PCB. In some examples, the external crosssection may be substantially square.

The connector assembly may have a lateral extent, in a plane of the first PCB, of more than 0.5 millimetres. The connector assembly may have a lateral extent, in a plane of the first PCB, of more than 10 millimetres. The connector assembly may have a lateral extent, in a plane of the first PCB, of less than 50 millimetres. The connector assembly may have a length, in a direction normal to the plane of the first PCB, of less than 5 millimetres. The length may be less than 2 millimetres. The length may be greater than 0.5 millimetres. It will be understood that the length can be considered as a the spacing between the first PCB and the second PCB, outside the connector assembly.

At least one of the openings may have a cross-sectional area, parallel to a plane of the first PCB (or a plane of the second PCB), of greater than 0.5 square centimetres. The cross-sectional area may be greater than 1 square centimetre. Thus, an electronic component of the first or second PCB can be accommodated within the opening. The cross-sectional area may be less than 5 square centimetres. The cross-sectional area may be less than 3 square centimetres.

The first PCB may comprise at least one integrated circuit chip, extending from a surface of the first PCB, in a direction of the second PCB, within at least one of the one or more openings. The second PCB may comprise at least one integrated circuit chip, extending from a surface of the second PCB, in a direction of the first PCB, within at least one of the one or more openings.

The RF signal connector and the one or more ground connectors may be formed from an electrically conductive material, for example copper. In some examples, the RF signal connector and the one or more ground connectors may be plated in copper around a conductive core. The conductive core may be formed from copper.

In some examples, forming or attaching the RF signal connector extending between the first and second PCBs and electrically connecting the RF signal connector to the RF signal conductors of the first and second PCBs comprises: forming or attaching the RF signal connector at the first PCB and electrically connecting the RF signal connector to the RF signal conductor of the first PCB; and forming or attaching the RF signal connector at the second PCB and electrically connecting the RF signal connector to the RF signal conductor of the second PCB.

In some examples, forming or attaching the one or more ground connectors extending between the first and second PCBs and electrically connecting the one or more ground connectors to the one or more ground conductors of the first and second PCBs comprises: forming or attaching the one or more ground connectors at the first PCB and electrically connecting the one or more ground connectors to the one or more ground conductors of the first PCB; and forming or attaching the one or more ground connectors at the second PCB and electrically connecting the one or more ground connectors to the one or more ground conductors of the second PCB.

Typically, the RF signal connector and the one or more ground connectors form a coplanar waveguide type structure. This causes nonhomogeneous distribution of current density (i.e. current crowding) towards the edge of the connectors. Typically, the RF signal connector and the associated one or more ground connectors are coaxial and the one or more ground connectors are arranged around a central axis which coincides with the RF signal connectors. Typically, the RF signal connector and the one or more ground connectors (in aggregate) form an enclosed arrangement, wherein the RF signal connector is wholly surrounded by one or more ground connectors. An enclosed arrangement reduces radiated energy loss from the RF signal connector. Furthermore, this also ensures that a characteristic impedance of the transmission line is maintained through the RF signal connector. An approximation is realised from a partially enclosed arrangement, wherein the one or more ground connectors do not wholly but instead partially surround the RF signal connector. Thus a partially enclosed arrangement suppresses radiated energy loss from the one or more conductive signal pillars of the RF signal connector and at least partially maintains a characteristic impedance of the transmission line through the RF signal connector while simultaneously providing superior arrangement possibilities and/or being easier to fabricate. The one or more ground connectors may together enclose more than 50% of a perimeter of the RF signal connector. The one or more ground connectors may together enclose more than 70% of a perimeter of the RF signal connector. The one or more ground connectors may together enclose less than 70% of a perimeter of the RF signal connector.

The method typically comprises providing a dielectric material (which is typically solid) between the RF signal connector and the one or more ground connectors. The dielectric material reduces loss from the RF signal connector to the one or more ground connectors. That is, the dielectric material is used to enhance the electrical separation between the RF signal connector and the one or more ground connectors. It will be understood that the dielectric material need not have a lower electrical conductivity than air.

Typically, the one or more ground connectors are formed from a series of stacked layers. Typically, the series of stacked layers are attached between the first and second PCBs. Typically, the stacked layers are a plurality of printed layers. Each stacked layer may, for example, have a ring, a half-ring, a square or a half-square cross section. Each stacked layer may have an annular cross-section. This may allow the RF signal connector to be formed between the first and second PCBs and the one or more ground connectors to be attached between the first and second PCBs. In other words, the RF signal connector can be provided within an annulus of the stacked layers forming the one or more ground connectors.

The RF signal connector conducts RF signals between the first and second PCBs. Typically the conducted RF signal has a frequency greater than 1 GHz. Typically the conducted RF signal has a frequency greater than 10 GHz. The conducted RF signal may have a frequency greater than 17 GHz. The conducted RF signal may have a frequency greater than 31 GHz. The conducted RF signal may have a frequency greater than 60 GHz. The conducted RF signal may have a frequency less than 100 GHz. The conducted RF signal may have a frequency less than 150 GHz. The conducted RF signal may have a frequency less than 80 GHz. Thus an efficient RF transition between the first and second PCBs is obtained due to reduced losses even when conducting a RF signal with a frequency greater than 10 GHz between the first and second PCBs. Typically, the RF signal connector and/or the one or more ground connectors are rigid, thus providing additional structural strength to the circuit board assembly and efficient electrical conductivity. In other words, the RF signal connector and/or the one or more ground connectors are different to a flexible cable, e.g., different to a co-axial cable.

In some examples, each of the RF signal connector and the one or more conductive ground connectors have the same dimensions. Typically, the RF signal connector and the one or more ground connectors have a width (or diameter) of 0.4 ± 0.1 mm. Typically, each of the one or more ground connectors are equidistant from the RF signal connector. Each of the one or more ground connectors may be arranged at a distance of at least twice the width (e.g., diameter) of the one or more ground connectors from the RF signal connector. Each of the one or more ground connectors may be arranged at a distance of approximately twice the width (e.g., diameter) of the one or more ground connectors from the RF signal connector. Preferably, each of the one or more ground connectors have a width (or diameter) of at least 0.2 ± 0.1 mm and are positioned at least 0.3 ± 0.2 mm from the RF signal connector. Preferably, each of the one or more ground connectors have a width (or diameter) of 0.4 ± 0.1 mm and are positioned 0.8 ± 0.2 mm from the RF signal connector. It may be that there is a distance of 1 mm separating any two ground connectors positioned on mutually opposite sides of the RF signal connector. It may be that there is a distance of 2 mm separating any two ground connectors positioned on mutually opposite sides of the RF signal connector.

Typically, the one or more ground connectors are a plurality of ground connectors. The RF signal connector and the plurality of ground connectors may be formed in a third PCB. The third PCB may be connected between the first and second PCBs. That is, the third PCB is a transition, or intermediary, PCB. The third PCB may be considered to be a connector assembly.

Typically, the third PCB is attached between the first and second PCBs. That is, the RF signal connector and the one or more ground connectors are attached between the first and second PCBs. Thus, the third PCB may be manufactured separately, thereby reducing the time required to perform the method.

Typically, the first, third and second PCBs are formed as parallel layers. Typically, the first, third and second PCBs form a solid multilayer structure. The third PCB is typically formed by forming a PCB comprising a plurality (typically at least 1) RF signal connectors and respective one or more ground connectors surrounding each RF signal connector, and cutting the said PCB to form a plurality (typically at least 1) of said third PCBs. Thus, the RF connector can be formed efficiently.

It may be that the third PCB is typically formed by forming a PCB comprising a plurality of (typically at least 100) RF signal connectors and respective one or more ground connectors surrounding each RF signal connector, and cutting the said PCB to form a plurality (typically at least 100) of said third PCBs.

The RF signal connector may comprise a first portion formed from a first conductive signal pillar and a second portion formed from a second conductive signal pillar. An RF signal passing through the RF signal connector may pass through the second portion subsequent to passing through the first portion. Thus, the RF signal connector may be split into (at least) two portions.

The first portion may be different from the second portion. In other words, the first conductive signal pillar may differ in some way (e.g., size) from the second conductive signal pillar.

The second portion may be laterally offset relative to the first portion. In other words, the second portion may not be coaxial with the first portion. Thus, the characteristics of the RF signal connector can be changed.

In some examples, the one or more ground connectors may each comprise a first ground portion formed from a first conductive ground pillar and a second ground portion formed from a second conductive ground pillar. The first ground portion may be different from the second ground portion. The first ground portion may be laterally offset from the second ground portion.

The method may comprise forming, embedding or attaching an (e.g., one or more) electrical component between the first conductive signal pillar and the second conductive signal pillar. The electrical component may be to alter an RF signal passing through the RF signal connector. The electrical component may be provided within the third PCB. Typically, the electrical component is one or more passive or active devices. The electrical component may be an electronic component. Where the one or more electrical components generate heat, the third PCB may be configured to include a heat sink in thermal contact with the one or more electrical components to remove excess heat from the one or more electrical components. The heat sink may be provided by one or more further vias arranged between the one or more electrical components and a surface of the third PCB.

The one or more electrical components may comprise an integrated circuit. The one or more electrical components may comprise a radio frequency integrated circuit (RFIC). The one or more electrical components may comprise a lumped element component. The one or more electrical components may comprise a filter, such as a distributed filter.

In some examples, the one or more electrical components may be one or more link members for providing a lateral offset between different portions of the RF signal connector (and/or different portions of the one or more ground connectors). The one or more electrical components may be provided by a stub. The stub may be an opencircuit stub. The stub may be a short-circuit stub.

It will be understood that the term passive device means substantially any electrical device requiring less than 100mA to operate.

The one or more ground connectors may be spaced from the RF signal connector substantially uniformly between the first PCB and the second PCB. In other words, the one or more ground connectors are each spaced from the RF signal connector by substantially the same amount at any position along the RF signal connector between the first PCB and the second PCB. In other examples, the one or more ground connectors may be spaced from the RF signal connector by a first amount at a first distance along the RF signal connector between the first PCB and the second PCB and by a second amount at a second distance (different from the first distance) along the RF signal connector between the first PCB and the second PCB.

The one or more ground connectors may extend, substantially unbroken, between the first PCB and the second PCB. A first conductive ground pillar may define a first end and a second end and may extend partway between the first PCB and the second PCB. A second conductive ground pillar may define a first end and a second end and be offset from the first conductive ground pillar and extend away from the first conductive ground pillar and partway between the first PCB and the second PCB. The second end of the first conductive ground pillar may be electrically connected to the first end of the second conductive ground pillar via a conductive link member. Thus, even for relatively long distances between the first PCB and the second PCB, the one or more ground connectors can be split into portions and thermal expansion or contraction of each portion of the one or more ground connectors will not cause mechanical failure of the one or more ground connectors.

In examples, the conductive link member may extend around the RF signal connector. The conductive link member may have a thickness in an axial direction of the RF signal connector between the first PCB and the second PCB of less than 5 millimetres.

It may be that at least one of the first PCB and the second PCB are formed by a lamination process. It may be that at least one of the first PCB and the second PCB are formed by a build-up process. It may be that at least one of the first PCB and the second PCB are formed by a combination of the lamination and the build-up process. It may be that at least one of the one or more openings are formed using a router. It may be that at least one of the first PCB and the second PCB are manufactured using 3D printing.

Typically, the third PCB is a formed by a build-up process. Thus, the distance between the first and second PCBs can be easily controlled. The distance between the first and second PCBs may be determined based on the dimensions of the RF signal connector and/or the one or more ground connectors. Alternatively or additionally, the distance between the first and second PCBs may be determined based on the arrangement of the one or more ground connectors around the RF signal connector. Typically, the third PCB is formed between the first and second PCBs such that the RF signal connector and the one or more ground connectors are formed between the first and second PCBs. Typically, the third PCB is formed on the first PCB, and subsequently connected to the second PCB.

The third PCB may be formed as a multi-layer structure comprising a plurality of dielectric layers with a plurality of vias therethrough which together function as the RF signal connector and one or more ground connectors. Typically, the plurality of vias of the dielectric layers of the third PCB are aligned to form the connectors. That is, a first signal via of the plurality of vias which functions as the RF signal connector of layer ‘n’ is aligned with a second signal via of the plurality of vias which functions as the RF signal connector of layer ‘n+1 ’. The method may comprise forming a plurality of via capture pads on each layer, typically such that there is a via capture pad between each adjacent pair of dielectric layers (typically attached to one or other of each adjacent pair of dielectric layers) and on an outward side of each of the outer dielectric layers of the third PCB. Thus, the vias of the plurality of dielectric layers can be electrically connected efficiently, even when the vias are not perfectly aligned.

Typically, the thickness of each dielectric layer of the third PCB is the same. Typically, each dielectric layer of the third PCB has a layer thickness of 100 to 1000 microns, preferably 200 microns. Typically, each dielectric layer of the third PCB has a layer thickness of 100 to 300 microns, preferably 200 microns.

Typically, each dielectric layer of the third PCB is composed of a material with a coefficient of thermal expansion of 100 (x 10' ⁶ /°C) or lower, preferably lower than 20 (x 10' ⁶ /°C). It may be that each dielectric layer of the third PCB is composed of a material with a coefficient of thermal expansion of 100 (x 10' ⁶ /°C) or lower, preferably lower than 70 (x 10' ⁶ /°C). This prevents electrical disconnection between the dielectric layers when subjected to heat. It may be that typically, each dielectric layer of the third PCB is composed of bismaleimide triaizine. It may be that typically, each dielectric layer of the third PCB is composed of glass-based epoxy. It may be that typically, each dielectric layer of the third PCB is composed of ceramic-filled fluoropolymers.

It may be that the RF signal connector and the one or more ground connectors are each formed by a lamination process. It may be that the RF signal connector and the one or more ground connectors are each formed by a build-up process. It may be that the RF signal connector and the one or more ground connectors are each formed by a combination of lamination and build-up process. Thus, the shape of the RF signal connector and the one or more ground connectors may be easily configurable. That is, a range of shapes and dimensions may be created efficiently. The build-up process may be such that the RF signal connector and the one or more ground connectors are built on the first PCB. The RF signal connector and the one or more ground connectors may each be formed from continuous bodies of conductive material.

Typically, the RF signal connector and one or more ground connectors are formed of conductive material which forms a single continuous body of conductor, without a large amount of intervening solder, with the RF signal conductor and the one or more ground conductors of the first PCB. Thus good electrical conductivity can be realised, without significant losses.

The single continuous body of conductive material which forms the RF signal connector and/or the one or more ground connectors may for example have a cylindrical, semi- cylindrical, cuboid, or semi-cuboid cross-section.

It may be that the one or more ground connectors is a single ground connector. The single ground connector may be formed to completely surround the RF signal connector, thus forming an enclosed arrangement.

Typically dielectric is formed between the RF signal connector and the one or more ground connectors concurrently with the formation of the RF signal connector and one or more ground connectors by a build-up process.

It will be understood that the build-up process is a method of forming the RF signal conductor and the one or more ground conductors gradually, such as in layers. For each layer, a copper foil is first formed, and then patterned and etched. The remaining Cu structures are the “vias,” and may be circular, rectangular, or arbitrary shaped. The layer is completed with a liquid dielectric resin which flows around all structures and cures. Once cured, this layer is functionally similar to a standard fiberglass PCB core (but without the fiberglass), and multiple layers can be bonded together. The layers are directly bonded to each other, so each via has intimate contact with those on previous layers.

The first and second PCB may be connected to the third PCB by soldering, such as using a standard solder reflow technique.

Typically, the method further comprises covering a top surface (the surface opposite the first PCB) of the structure comprising the RF signal connector and the one or more ground connectors (and typically also dielectric therebetween) with a mask layer defining regions for the formation of extended portions of the RF signal connector and the one or more ground connectors, extending the RF signal connector and the one or more ground connectors, and removing the mask layer to thereby form a structure having the RF signal connector and one or more ground connectors extending therefrom. Thus, the provision of extended portions of the RF signal connector and the one or more ground connectors allow machining to planarize this side of the structure for later soldering between the multi-layer structure and the second PCB.

Typically, the dimensions of the extended portions of the RF signal connector and the one or more ground connectors is (approximately) defined by the regions in the mask layer. Typically, the mask layer has a layer thickness between 100 and 200 microns.

Typically, extending the RF signal connector and the one or more ground connectors comprises applying a conductive material onto the RF signal connector and the one or more ground connectors. Typically, applying a conductive material comprises plating, e.g. copper plating. Thus, a good electrical connection between the extension and the RF signal connector and the one or more ground connectors is formed.

Typically, once the RF signal connector and one or more ground connectors are formed on or attached to the first PCB, the method further comprises applying an overmould to the first PCB, the overmould partially or wholly covering the RF signal connector and the one or more ground connectors (and typically also dielectric therebetween), removing a top surface (i.e. the surface opposite the first PCB) of the overmould, and typically also of the RF signal connector and the one or more ground connectors, and connecting the thereby exposed RF signal connector and the one or more ground connectors to the second PCB.

Typically, the step of applying an overmould includes partially or wholly covering one or more additional circuitries on the first PCB. The one or more additional circuitries (e.g. one or more passive or active devices) are typically separate from the RF signal connector and the one or more ground connectors. Thus, the one or more additional circuitries may be protected by the overmould at the same time as the RF signal connector and the one or more ground connectors are covered by the overmould, thereby reducing the manufacturing time.

Typically, the step of removing the top surface of the overmould, and typically also removing the top surface of the RF signal connector and the one or more ground connectors, comprises one or more of grinding, etching, sawing or laser cutting.

Typically, removing the top surface of the RF signal connector and the one or more ground connectors comprises removing a portion of the top surface of a structure having the RF signal connector and the one or more ground connectors extending therefrom.

Typically the step of removing the top surface of the overmould causes the RF signal connector to be isolated from the remaining overmould. That is, the RF signal connector does not come into contact with (e.g. , the RF signal connector is not exposed to) the overmould. Thus, losses from the RF signal connector to the overmould are reduced.

Typically, overmould compounds may have high loss tangent, which could introduce significant loss if the RF signal connector is in contact with the overmould.

Typically, the RF signal connector is soldered to the signal conductor of the second PCB. Typically also, the one or more ground connectors are soldered to one or more ground conductors of the second PCB.

Typically, the RF signal connector and the one or more ground connectors are aligned in a plurality of discrete rows, at least some of the rows passing through at least two or at least three of said pillars. The method further comprises removing (typically cutting) a top surface of said pillars (and optionally said overmould) along said discrete rows. The top surface can therefore be automatically removed (e.g. cut) with ease, as the thickness of the top surface to be removed can be easily calculated. This is because the thickness of each discrete row is known. It typically also requires fewer passes for the top surface to be removed - for example, a saw or cutter may cut a plurality of discrete rows each pass. It may be that only a single pass is required.

Typically the one or more ground connectors are two ground connectors which are located on mutually opposite sides of the RF signal connector. The one or more ground connectors may be three or more ground connectors. The three or more ground connectors may be provided in a circumferentially distributed arrangement (e.g., a ring) around the RF signal connector. It may be that the three or more ground connectors are located in a half-ring, a square, or a half-square around the RF signal connector.

According to another aspect of the present disclosure, there is provided a circuit board assembly comprising: first and second PCBs, each having an RF signal conductor and one or more ground conductors thereon; an RF signal connector extending between the first and second PCBs and electrically connected to the RF signal conductors of the first and second PCB; and one or more ground connectors extending between the first and second PCBs and electrically connected to the ground conductors of the first and second PCBs. The one or more ground connectors are arranged to at least partially surround the RF signal connector. Thus, losses of the RF signal during transmission between the first PCB and the second PCB can be reduced.

The circuit board assembly disclosed herein may be formed by any of the methods described above.

Typically, the connector assembly is connected to at least one of the first PCB and the second PCB using a plurality of solder connections. It may be that at least one of the plurality of solder connections contacts the respective first or second PCB over an area of at least 25 microns squared (i.e. 5x10-6 metres by 5x10-6 metres, 25x10-12 m ² or 2.5x10-11 m ² or 0.000025 mm ² ). Advantageously, solder connections of this size reduce radiation of radio frequency energy when the RF signal is transitioning from the first PCB to the second PCB.

Typically, use of solder connections of this size improve alignment between assembled circuit boards. By using small solder connections, the alignment of the first PCB and the second PCB is improved such that the second plane of the second PCB may be approximately parallel to first plane of the first PCB. The second plane may be parallel to first plane such that it does not intersect the first plane. The second plane may have an acute angle of intersection with the first plane of less than 1 °. The second plane may have an acute angle of intersection with the first plane of less than 0.15°. The second plane may have an acute angle of intersection with the first plane between 0.014° and 0.15°. In particular, it is important for circuit boards used in tuneable antenna arrays to be aligned in this manner.

The offset may be a result of inconsistencies in the height of the solder connections (i.e. solder connections between the first PCB and the second PCB having differing heights). The height of the solder may be the distance from which the solder connections extend from the surface of the connector assembly (i.e. a direction perpendicular to the plane of the first PCB). An amount of offset of the second PCB compared to the first PCB may be defined by an offset distance. The offset distance may be a distance from a first point of the second PCB to a plane that is parallel to the first plane of the first PCB and vertically separated from the first PCB (i.e. by a distance corresponding to a lowest height of the solder connections between the connector assembly and the second PCB). The first point of the second PCB may be a point at which the distance between the second PCB and the first PCB is the greatest. The first point of the second PCB may be a point at which the distance between the second PCB and the first PCB is a mean (i.e. a mathematical average) distance of the second PCB to the first PCB.

Typically, in a circuit board assembly comprising a plurality of PCBs, the plane of each PCB may be substantially parallel to one another. Typically, a plurality of circuit board assemblies may be arranged on a surface. The plurality of circuit board assemblies may be arranged adjacent to one another on a surface. The surface may be curved or flat. In some examples, the surface is provided by the first PCB. In other words, the plurality of circuit board assemblies may each share different regions of the one or more first PCBs, and the plurality of circuit board assemblies may comprise a plurality of second PCBs, one each for each of the plurality of circuit board assemblies.

Typically, the one or more openings are flooded with filling material once the circuit component is accommodated in the one or more openings.

At least one of the RF signal connector and the one or more ground connectors may each be formed as a multi-layer structure. In other words, a first layer of the RF signal connector may be formed separately to a second layer of the RF signal connector. For example, the first layer of the RF signal connector may be formed in a first processing step. Subsequently, the second layer of the RF signal connector may be formed on the first layer in a separate processing step. Alternatively, the RF signal connector may be formed in a single processing step. A first layer of the one or more ground connectors may be formed separately to a second layer of the one or more ground connectors. For example, the first layer of the one or more ground connectors may be formed in a first processing step. Subsequently, the second layer of the one or more ground connectors may be formed on the first layer in a separate processing step. Alternatively, the one or more ground connectors may (each) be formed in a single processing step.

The RF signal connector and the one or more ground connectors may be provided in a connector assembly extending between the first and second PCBs. The connector assembly may define one or more openings (e.g. depressions) therein, each adjacent one of the first PCB and the second PCB. Thus, the connector assembly can be formed to accommodate one or more components, each mounted on a one of the first PCB and the second PCB, extending in a direction towards the other of the first PCB and the second PCB, even when the components are provided close to the RF signal connectors and the one or more ground connectors. Thus, a more compact circuit board assembly can be provided. In some examples, it may be that at least one of the one or more openings extend wholly through the connector assembly from the first PCB to the second PCB. In other words, the opening may be provided by a through- channel (e.g. a through-bore). The method may comprise filling any remaining free space of the one or more openings of the connector assembly after installing the connector assembly between the first PCB and the second PCB. Thus, the opening can remain only as large as necessary for accommodating any components which extend from the first or second PCBs into the one or more openings.

There may be at least one opening extending from the first PCB to the second PCB. There may be at least two openings extending from the first PCB to the second PCB.

At least one of the one or more openings, may be wholly surrounded by a boundary portion of the connector assembly, the boundary portion extending fully to one or both of the first PCB and the second PCB. In other words, the one or more openings are typically features of the connector assembly, inwards of an outer edge of the connector assembly.

The connector assembly may define a substantially rectilinear external cross-section, in a plane parallel to the first and/or second PCB. In some examples, the external crosssection may be substantially square.

The connector assembly may have a lateral extent, in a plane of the first PCB, of more than 1 millimetre. The connector assembly may have a lateral extent, in a plane of the first PCB, of more than 10 millimetres. The connector assembly may have a lateral extent, in a plane of the first PCB, of less than 50 millimetres. The connector assembly may have a length, in a direction normal to the plane of the first PCB, of less than 5 millimetres. The length may be less than 2 millimetres. The length may be greater than 0.5 millimetres. It will be understood that the length can be considered as a the spacing between the first PCB and the second PCB, outside the connector assembly.

At least one of the openings may have a cross-sectional area, parallel to a plane of the first PCB (or a plane of the second PCB), of greater than 0.5 square centimetres. The cross-sectional area may be greater than 1 square centimetre. Thus, an electronic component of the first or second PCB can be accommodated within the opening. The cross-sectional area may be less than 5 square centimetres. The cross-sectional area may be less than 3 square centimetres.

The first PCB may comprise at least one integrated circuit chip, extending from a surface of the first PCB, in a direction of the second PCB, within at least one of the one or more openings. The second PCB may comprise at least one integrated circuit chip, extending from a surface of the second PCB, in a direction of the first PCB, within at least one of the one or more openings.

Typically, the RF signal connector is soldered to the signal conductor of the second PCB. Typically also, the one or more ground connectors are soldered to one or more ground conductors of the second PCB.

Typically, the RF signal connector and the one or more ground connectors are separated by a dielectric material.

Typically, the first and second PCBs of the circuit board assembly are 0.5 to 1.5 mm apart, preferably 0.8 to 1 mm apart.

Typically, the RF signal connector and/or the one or more ground connectors are rigid, thus providing additional structural strength to the circuit board assembly.

Typically, the circuit board assembly comprises one or more passive or active devices embedded between the first and second PCBs.

Typically, the one or more ground connectors are a plurality of ground connectors. The RF signal connector and the plurality of ground connectors may be formed in a third PCB. The third PCB may be connected between the first and second PCBs.

Typically, the first, third and second PCBs are parallel layers. Typically, the first, third and second PCBs form a solid multilayer structure. The third PCB may comprise one or more electrical (e.g., electronic) components within the third PCB.

The third PCB may be a multi-layer structure (typically a substrate). The third PCB may comprise a plurality of dielectric layers with a plurality of vias therethrough which together function as the signal pillar and one or more ground pillars. The third PCB may comprise a plurality of via capture pads on each layer, typically such that there is a via capture pad between each adjacent pair of dielectric layers (typically attached to one or other of each adjacent pair of dielectric layers) and on an outward side of each of the outer dielectric layers of the third PCB.

Typically, the RF signal connector and the one or more ground connectors are each formed by a build-up process. The RF signal connector and the one or more ground connectors may each be formed from continuous bodies of conductive material.

Typically, the circuit board assembly comprises a dielectric material between the RF signal connector and the one or more ground connectors. The dielectric material may be a continuous body formed by the build-up process.

The RF signal connector and/or the one or more ground connectors may for example have a cylindrical, semi-cylindrical, cuboid, or semi-cuboid cross-section.

Typically the circuit board assembly further comprises an overmould which partially or wholly covers the RF signal connector and the one or more ground connectors except at a top surface (i.e. the surface opposite the first PCB) of the RF signal connector and the one or more ground connectors.

Typically, the RF signal connector and the one or more ground connectors each comprise a portion of conductive material separated by the overmould.

Typically, the first PCB comprises one or more additional circuitries (e.g. one or more passive or active devices). Typically, the overmould also covers the one or more additional circuitries.

Typically, the RF signal connector is isolated from the overmould.

Typically, the one or more ground connectors are two ground connectors which are located on mutually opposite sides of the RF signal connector. The one or more ground connectors may be three or more ground connectors. The three or more ground connectors may be distributed around the RF signal connector. In some examples, the three or more ground connectors are located in a circumferentially distributed arrangement (e.g., a ring) around the RF signal connector. The circuit board assembly may further comprise one or more further signal connectors arranged to carry electrical signals, the one or more further signal connectors each extending between the first PCB and the second PCB and arranged to be shielded from the RF signal connector by the one or more ground connectors. Thus, the same element used to transfer the RF signal between the first PCB and the second PCB can also be used to transfer other electrical signals (e.g. data signals and/or control signals) between the first PCB and the second PCB in a particularly efficient way. The one or more further signal connectors may be at least five further signal connectors.

In some examples, the one or more further signal connectors are arranged to be mutually shielded from each other. The one or more further signal connectors may be embedded within a dielectric material, arranged to substantially surround the one or more ground connectors.

The one or more further signal connectors may each comprises one or more further signal pillars.

It will be understood that the one or more further signal connectors may be formed in substantially the same way as the RF signal connector described hereinbefore.

The one or more ground connectors may be arranged to define a substantially circular inner profile surrounding the RF signal connector. An outer profile defined by the one or more ground connectors, radially outward of the inner profile may be non-circular. Specifically, the outer profile may be formed to have one or more cutout regions in which the outer profile departs from a circular shape to ensure spacing between a one of the one or more further signal connectors and the one or more ground connectors.

Where the circuit assembly comprises a connector assembly defining one or more openings therein, it may be that the one or more further signal connectors define at least a portion of a boundary of at least one of the one or more openings. In other words, the openings may be bounded on at least some sides by the one or more further signal connectors. It may be that the RF signal connector and the one or more ground connectors may be provided away from an edge of the connector assembly.

According to another aspect of the present disclosure, there is provided a circuit board assembly, for a microelectromechanical system, MEMS, based antenna assembly, comprising: first and second PCBs, each having an RF signal conductor and one or more ground conductors thereon; an RF signal connector extending between the first and second PCBs and electrically connected to the RF signal conductors of the first and second PCB; and one or more ground connectors extending between the first and second PCBs and electrically connected to the ground conductors of the first and second PCBs, wherein the one or more ground connectors are arranged to at least partially surround the RF signal connector, wherein the RF signal connector and the one or more ground connectors are provided in a connector assembly extending between the first and second PCBs and defining one or more openings therein, each opening adjacent to at least one of the first PCB and the second PCB, wherein the circuit board assembly further comprises a plurality of signal connectors extending between the first and second PCBs and arranged around a periphery of the one or more openings, wherein the first PCB comprises a circuit component extending from a surface of the first PCB, towards the second PCB, and wherein the connector assembly is arranged such that at least one of the one or more openings accommodates the circuit component therein.

According to another aspect of the present disclosure, there is provided a method of forming a circuit board assembly, for a microelectromechanical system, MEMS, based antenna assembly, the circuit board assembly comprising a first PCB, a second PCB, an RF signal connector for carrying RF signals therebetween, and one or more ground connectors for at least partially shielding the RF signal connector, the method comprising: providing first and second PCBs, each having an RF signal conductor and one or more ground conductors thereon; forming or attaching an RF signal connector extending between the first and second PCBs and electrically connecting the RF signal connector to the RF signal conductors of the first and second PCBs; and forming or attaching one or more ground connectors extending between the first and second PCBs and electrically connecting the one or more ground connectors to the ground conductors of the first and second PCBs; the one or more ground connectors together being arranged to at least partially surround the RF signal connector, wherein the RF signal connector and the one or more ground connectors are provided in a connector assembly extending between the first and second PCBs and defining one or more openings therein, each opening adjacent to the first PCB and the second PCB, wherein the circuit board assembly further comprises a plurality of signal connectors extending between the first and second PCBs and arranged around the periphery of the one or more openings, wherein the first PCB comprises a circuit component extending from a surface of the first PCB, towards the second PCB, and wherein the connector assembly is arranged such that at least one of the one or more openings accommodates the circuit component therein.

According to another aspect of the present disclosure, there is provided a MEMS based antenna assembly comprising: an input port for receiving an input RF signal; a plurality of antennas to transmit RF signals originating from the input port; one or more MEMS components in an RF signal path from the input port to each of the plurality of antennas, the one or more MEMS components comprising at least one of: one or more MEMS phase shifters to control phase shifting of the RF signals transmitted by the plurality of antennas; one or more MEMS attenuators to control amplitude of the RF signals transmitted by the plurality of antennas; one or more MEMS capacitive switches to control frequency of the RF signals transmitted by the plurality of antennas; and one or more MEMS switches to control the impedance of signal path of the RF signals transmitted by the plurality of antennas; and one or more of the circuit board assemblies of any preceding claim, wherein the RF signal connector of the or each respective circuit board assembly is provided in the RF signal path between the input port and one or more of the plurality of antennas.

Description of the Drawings

An example embodiment of the present invention will now be illustrated with reference to the following Figures in which:

Figure 1 is a side view of a circuit board assembly;

Figures 2A through 2E are plan views of a number of alternative pillar arrangements when there is a plurality of ground pillars;

Figures 3A through 3I show the steps of creating a transition line between two PCBs;

Figures 4A through 4D are plan views of a number of alternative pillar arrangements when there is a single, continuous ground pillar;

Figures 5A through 5J show the steps of creating a transition line between two PCBs using a transition PCB; Figures 6A through 6D show the steps of creating a transition line between two PCBs using a transition PCB and where a first PCB includes additional circuitry;

Figure 7 shows the steps of creating a transition line between two PCBs where a first PCB includes additional circuitry;

Figure 8 shows a representation of an example of a transition in accordance with the present disclosure;

Figure 9 shows a representation of a further example of a transition in accordance with the present disclosure;

Figure 10 shows a representation of a yet further example of a transition in accordance with the present disclosure;

Figure 11 shows a representation of a still further example of a transition in accordance with the present disclosure;

Figure 12 shows a representation of yet another example of a transition in accordance with the present disclosure; and

Figure 13 shows a representation of the transition of Figure 12, provided between two PCBs.

Detailed Description of an Example Embodiment

Figure 1 is a side view of a circuit board assembly 100, comprising a first PCB 110, a RF signal conductor 120 in the form of an RF signal pillar 120, a plurality of ground conductors 130, 140 in the form of ground pillars 130, 140 extending vertically from the first PCB (in this example three pillars are depicted, but any number more than two is envisaged), dielectric material 150 surrounding and separating the three pillars, conductive extensions 160 on the top surface of the pillars 120, 130, 140, an overmould 170, solder balls 180, and a second PCB 190.

The signal pillar 120 is an electrically conductive pillar used to transmit a RF signal between the first PCB 110 and the second PCB 190 (that is, from the first PCB 110 to the second PCB 190, and from the second PCB 190 to the first PCB 110). In particular, signal pillar 120 is electrically connected to a RF signal conductor (not shown) of the first PCB 110 and of the second PCB 190. The signal pillar 120 may be directly electrically connected to the RF signal conductor of the first PCB 110, and is electrically connected to the RF signal conductor of the second PCB 190 via the conductive extension 160 on the top surface of the signal pillar 120 and the solder ball 180.

The ground pillars 130, 140 are electrically conductive pillars which are grounded. That is, the ground pillars 130, 140 are grounded by being electrically connected to ground conductors of the first PCB 110 and/or the second PCB 190. It may be that the ground pillars 130, 140 are connected to the same ground line of either the first PCB 110 and/or the second PCB 160. Similar to the signal pillar 120, the ground pillars 130, 140 may be directly electrically connected to the RF ground conductor (not shown) of the first PCB 110, and/or may be electrically connected to the RF ground conductor (not shown) of the second PCB 190 via the conductive extensions 170 on the top surfaces of the ground pillars 130, 140 and the solder balls 180 attached thereafter. It may be that, when the ground pillars 130, 140 are only attached to the ground conductor of the first PCB 110, the circuit board assembly 100 does not include conductive extensions on the top surface of the ground pillars 130, 140 or solder balls attached thereafter.

The ground pillars 130, 140 are arranged to at least partially surround the signal pillar 120. That is, at minimum where there are only two ground pillars, the ground pillars are arranged on mutually opposite sides of the signal pillar, as shown in Figure 1. Figures 2A through 2E display a variety of alternative possibilities, although other arrangements may be used within the scope of the present disclosure. Surrounding, or at least partially surrounding, the signal pillar 120 with grounded pillars (e.g. ground pillars 130, 140) affords an enclosed or a partially enclosed arrangement, respectively.

The dielectric material 150 is used to enhance the electrical separation of the conductive pillars 120, 130, 140. This reduces loss by transfer from the signal pillar 120 to the grounded pillars 130, 140, particularly when the dielectric material has a low electrical conductivity, preferably lower than that of air. Insulation of the signal pillar 120 from the grounded pillars 130, 140 therefore reduces losses particularly when an RF signal with a frequency greater than 10 GHz is conducted through the signal pillar 120.

The conductive extensions 160, on the top surfaces of the signal pillar 120 and, typically, the ground pillars 130, 140, act as sacrificial material (which is explained further with reference to Figures 2E-G below). During the manufacturing of the circuit board assembly, the conductive extensions are provided so that a portion of them may be removed to enable conductivity (for reasons discussed below with regards to the overmould) between the conductive extension and the second PCB 190, for example, via the solder balls 180. The conductive extensions 170 are electrically connected to the top surface of the signal pillar 120 and, typically, the ground pillars 130, 140. If the circuit board assembly 100 includes additional ground pillars then conductive extensions are also on the top surfaces of these additional ground pillars.

The overmould 170 is a dielectric material. Typically, the first PCB 110 will include one or more additional circuitries such as passive or active devices (not shown in Figure 1). The overmould 170 is applied such that the one or more additional circuitries on the first PCB 110 are also covered by the overmould 170. The overmould 170 protects conductive pillars 120, 130, 140 and the one or more additional circuitries, for example, from moisture which can be detrimental to the efficiency of the pillars and the circuitries.

A portion of the overmould 170, as well as the conductive extensions 160, has been removed. This allows minimum contact of the signal pillar 120 with the overmould 170 and allows electrical connection to the second PCB 190. The portion of the overmould 170 has been removed so that solder balls 180 may be electrically connected to the conductive extensions 160 and subsequently connected to the second PCB 190. The solder balls 180 are attached to the conductive extensions 160 through reflow using techniques known in the art.

Figures 2A-2E are a number of alternative arrangements where a plurality of ground pillars are used to approximate an enclosed arrangement (that is, to have a partially enclosed arrangement) around a signal pillar 120. In Figure 2A, only two ground pillars 130, 140 are present. Alternatively, Figure 2B is a half-ring arrangement, whereas Figure 2C is a ring arrangement around the signal pillar. Figure 2D is a half-square arrangement and Figure 2E is a square arrangement. To avoid over-crowding the diagram, the ground pillars have not been labelled in Figures 2B to 2E.

Figures 3A-I are side views showing different steps of the method for constructing the circuit board assembly 100. Although Figures 3A-I depict three conductive pillars (one signal pillar 120 and two ground pillars 130, 140), any number of pillars are envisaged provided that the ground pillar or pillars at least partially surround the signal pillar. In Figure 3A, a first PCB 110 with a signal pillar 120, two ground pillars 130, 140 and a dielectric material 150 is provided. This may be constructed as per the examples provided below with reference to Figures 4 and 5.

Figures 3B-E are examples of constructing the conductive extensions 160. Here, a mask layer 210 is provided above on the top surface of the pillars and dielectric material, Figure 3B. That is, on the surface opposite the first PCB 110. The thickness of the mask layer 210 approximates the thickness of the conductive extensions 160 (shown in Figure 3D). Typically the thickness of the mask layer 210 allows the conductive extensions 160 to be thicker than what is required for the final circuit board assembly. That is, a portion of the conductive extensions 160 is to be a sacrificial portion in preparation for connection with the second PCB 190 (as described in greater detail below with reference to Figure 3G).

In Figure 3C, regions 220 in the mask layer 210 are created to thereby expose the underlying top surfaces of the pillars 120, 130, 140 underneath the mask layer 210. That is, the regions 220 in the mask layer 210 coincide with the underlying pillars 120, 130, 140. The regions 220 in the mask layer 210 may be created, for example, by etching. The conductive extensions 160 are formed by filling the regions 220 in the mask layer 210 with a conductive material, as per Figure 3D. The conductive extensions 160 may, for example, be made of copper or any other conductive material suitable for electrical connection with the pillars 120-140. Finally, as shown in Figure 3E, the mask layer 210 is removed, leaving the pillars 120-140 electrically connected to the conductive extensions 160.

Figure 3F displays an overmould 170 having been applied to cover the conductive extensions 160. The thickness of the overmould 170 is typically such that the conductive extensions 160 are only just covered. This therefore reduces the amount of overmould 170 required. The overmould 170 of Figure 3F has an exaggerated thickness for clarity.

A portion of the overmould 170 and the conductive extensions 160 is removed in Figure 3G. By removing a portion of the conductive extensions 160, it is ensured that a good electrical connection can be made. That is, it ensures that there is no overmould 170 overlapping the conductive extensions 160. The portion may be removed, for example, by machining, grinding, etching, sawing, or laser cutting; although any suitable technique may be employed. It can be seen in Figure 3G that the remaining overmould has limited contact with the signal pillar 120 and the conductive extension 160 connected to the signal pillar 120. The reduced contact therefore minimises losses from the signal pillar 120 (or the conductive extension 160 attached thereon) to the ground pillars 130, 140 (or the conductive extensions 160 attached thereon) through the overmould 170.

Figures 3H-3I display the final steps for connecting the second PCB 190. Solder balls 180 are applied in Figure 3H. This provides an electrical connection with the conductive extensions 160. The solder balls 180 are applied and attached using any suitable known techniques. As a result, a continuous, conductive path from the first PCB 110 to the solder balls 180 is obtained.

Figure 3I displays the second PCB 190 attached such that the solder ball 180 attached to the signal pillar 120 is electrically connected to the RF conductor of the second PCB 190. The solder balls 180 attached to the ground pillars 130, 140 are typically attached to ground conductors of the second PCB 190. This may be in addition to the ground pillars 130, 140 being connected to a ground conductor of the first PCB 110, or instead of.

The completed connection as per Figure 3I affords a transition line between the first PCB 110 and the second PCB 190 which minimises losses. In particular, the described transition surrounds conductive lines with dielectric material and requires the use of only small solder balls, therefore allowing a good aspect ratio to be obtained. As a result, a greater distance between the two PCBs 110, 190 may be achieved while minimising losses and maintaining a good aspect ratio for the transition.

The described method above demonstrates the basic principle of the present invention. Two possible methods of construction will now be described, particularly in describing the construction of the signal and ground pillars.

Figures 4A-4D displays a number of possible cross-sections for the signal pillar 320 and ground pillars 330a, 330b, 330c, 330d. Here, the pillars are constructed by a buildup process, resulting in a single continuous ground pillar. In this method, the pillars are progressively created on the first PCB and built up to the required height. This requires the signal pillar, ground pillar and dielectric material to be progressively built up. It is possible that the single ground pillar is formed to completely surround the signal pillar, having a cylindrical shape, as shown as ground pillar 330a in Figure 4A. Alternatively, the ground pillar may have a semi-cylindrical shape (ground pillar 330b in Figure 4B), cuboid shape (ground pillar 330c in Figure 4C), or a semi-cuboid shape (ground pillar 330d in Figure 4D), or any other suitable shape so as to at least partially surround the signal pillar 320.

Using this method, the pillars may be formed to have a number of different shapes. It also allows an enclosed arrangement (Figure 4A or 4C) or a partially enclosed arrangement (Figure 4B or 4D) to be formed to reduce any radiative losses from the signal pillar.

Once the pillars are created using the above method, the second PCB is attached as described above. That is, conductive extensions are formed on top of the pillars, an overmould is applied and a portion of the overmould (and the conductive extensions) is removed. Finally, solder balls are attached to the conductive extensions and the second PCB attached to the solder balls.

A further method is shown in Figures 5A-5J. Here, the pillars are constructed in a transition PCB 400, as shown in Figure 5A. An example is provided where the transition PCB 400 is composed of a plurality of dielectric layers 410 (i.e. cores). Each dielectric layer 410 has a plated via 412 connected to a contact pad 414. Each dielectric layer 410 is aligned such that the aligned plurality of plated vias 412 and contact pads 414 form conductive pillars.

Figures 5B-5E display the formation of conductive extensions 160 as described above with relation to Figures 3B-3E. Analogously to the situation described with reference to Figure 5B hereinbefore, the mask layer 210 is provided above an upper surface of the transition PCB 400, in contact with a contact pad 414 provided in contact with the uppermost portion of the plated vias 412 in the dielectric layer 410 providing the upper surface of the transition PCB 400. In Figure 5C, it can be seen that the exposed top surface of the pillars are the contact pads which aid in achieving good electrical connection with the conductive extensions applied in Figure 5D, with the remaining portion of the mask layer 210 removed in Figure 5E.

Figure 5F displays the transition PCB 400 being connected at a lower surface thereof to the first PCB 110, after the conductive extensions have been created. However, in other examples, it may be that the transition PCB 400 is formed on top of the first PCB 110. That is, the dielectric layers 410 are progressively created on top of one another after the first dielectric layer is attached to the first PCB 110.

Regardless of how the transition PCB 400 is formed or attached to the first PCB 110, the transition from the first PCB 110 to the second PCB 190 is completed as described above and shown in Figures 5G-5J, which are substantially similar to the technique used in relation to the method described hereinbefore with reference to Figures 3F to 3I, with the hereinafter further relevant description.

The transition PCB 400 is shown to be attached to the first PCB 110 after the conductive extensions are formed on the transition PCB 400 (for example, once any remaining masking layer is removed). However, the transition PCB 400 may alternatively be attached to the first PCB 110 before the conductive extensions are formed, that is, after any of Figures 5A-5D.

Although the Figures 5A-5J depict a transition PCB 400 having 5 dielectric layers, it will be understood that any number of layers may be used.

Figures 6A-6D displays an example of connecting a second PCB 190 when a transition PCB 400 is attached to a first PCB 110 which includes additional circuitry 500. Here, the overmould 170 is used to also cover the additional circuitry 500, as shown in Figure 6B. The overmould 170 may be used to minimise losses from the transition PCB 400 to the additional circuitry 500, as well as be used to seal the additional circuitry 500. Figure 6C shows that the portion of overmould 170 removed is localised towards the top of the transition PCB 400, rather than across the whole length of the overmould 170. It may be that the second PCB 190 attached to the top of the transition PCB 400 has a different length than the first PCB 110 as shown in Figure 6D. In other examples, the length of the first PCB 110 and the length of the second PCB 190 may be the same.

Although Figures 6A-6D depict a transition PCB 400, it will be appreciated that any method of forming the pillars may be utilised (e.g. forming the pillars by a build-up process). An example is shown in Figure 7, where the pillars are formed in accordance with the method described hereinbefore with reference to Figures 3A to 3I.

Figure 8 shows a representation of an example routing of ground connectors around an RF signal connector. Specifically, the RF signal connector 610 is surrounded by a plurality of ground connectors 620, in the form of ten ground connectors 620 as the RF signal connector 610 and the ground connectors 620 extend between a first PCB and a second PCB (first PCB and second PCB not shown for clarity). The ground connectors 620 each comprises a first portion 622 provided in the form of a first conductive ground pillar 622 extending to an intermediate axial position 630 between the first PCB and the second PCB in a direction from a first end of the ground connector 620 towards the intermediate axial position 630. The ground connectors 620 each further comprise a second portion 624 provided in the form of a second conductive ground pillar 624 extending from the intermediate axial position 630 towards a second end of the ground connector 620.

The first portion 622 is offset from the second portion 624 in a circumferential direction such that the ground connector 620 does not extend directly from the first portion 622 at the intermediate axial position 630 in the axial direction towards the second end of the ground connector 620. The first portion 622 is connected to the second portion 624 by a link member 626. In this way, axial expansion (e.g., thermal expansion) for the ground conductor 620 can be split between the first portion 622 and the second portion 624 (and any further portions of the ground connector 620). By splitting any axial expansion over multiple portions instead of a single axial extent of the ground connector 620, flex of the link member 626 can be used to mitigate each smaller portion of any axial expansion, meaning that the ground connector 620 can be formed to have a greater total length whilst still maintaining a low risk of mechanical failure of the ground connector 620 due to axial expansion. It will be understood that the link member 626 provides a flexible link between the first portion 622 and the second portion 624. In this way, axial expansion of the first portion 622 can be accommodated by flex in the link member 626, reducing, or even completely eliminating, a proportion of the axial expansion that is passed into the second portion 624.

In the example shown in Figure 8, the ground connector 620 is split into two axial portions, though it will be understood that several could be provided, each offset. It will be understood that the purpose of the ground connector 620 is to provide an RF shield around the RF signal connector 610. The RF shield need not be completely unbroken in a circumferential direction around a boundary of the RF signal connector 610. Acceptable performance may still be achieved with an RF shield including a discrete number of separated ground connectors surrounding the RF signal connector 610. Nevertheless, it will be understood that the larger the space between the ground connectors, the greater the amount of energy that will pass through the RF shield. By offsetting the first portion 622 and the second portion 624, a greater spacing can be used between a plurality of first portions 622 of the ground connectors 620 around the RF signal connector 610. Were the first portions 622 not to be offset from the second portions 624, to achieve the same shielding performance, it would be necessary to use a closer spacing between the plurality of first portions 622, therefore requiring a greater number of first portions 622. In this example, each of the link members 626 of the ground connectors 620 are connected to form a ring, providing further shielding of the RF signal connector 610. For simplicity, dielectric between the ground connectors 620 and the RF signal connector 610 is not shown.

Figure 9 shows a representation of a further example of a connection routing between PCBs (PCBs not shown in Figure 9 for simplicity). The RF signal connector 710 comprises a first portion provided by a first conductive signal pillar 712 and a second portion provided by a second conductive signal pillar 714. The second conductive signal pillar 714 is connected to the first conductive signal pillar 712 at an intermediate axial location 730 via a signal link member 716. RF shielding is provided to the RF signal connector 710 by a plurality of ground connectors 720. For the first portion of the RF signal connector 710, a first portion 722 of the plurality of ground connectors 720 are circumferentially distributed around the first conductive signal pillar 712 to provide RF shielding thereto. The first portions 722 of the plurality of ground connectors 720 are provided by first ground pillars 722. At the intermediate axial location 730, the plurality of first portions 722 of the plurality of ground connectors 720 are connected to a plurality of second portions 724 of the plurality of ground connectors 720 via a ground link member 726. The ground link member 726 laterally surrounds the signal link member 716. The second portions 724 of the plurality of ground connectors 720 extend from the link member 726 in a circumferentially distributed arrangement to partially surround the second conductive signal pillar 714. The second portions 724 of the plurality of ground connectors 720 are each provided by a ground pillar 724. In this way, it can be understood that the transition between the first PCB and the second PCB can include a lateral offset in the RF signal connector, whilst still maintaining suitable RF shielding. As described elsewhere herein, it will also be understood that one or more electrical (e.g., electronic) components can be provided in place of at least part of the signal link member 716 and/or the ground link member 726.

Figure 10 shows a representation of a yet further example of a transition in accordance with the present disclosure. In particular, the transition 800 is shown as a cross-section, showing all elements which extend through the transition 800 for connecting between signal conductors and ground conductors of a first PCB and a second PCB. The transition 800 comprises an RF signal connector 810 and a ground connector 820, each substantially as described hereinbefore apart from the herein noted distinctions. The ground connector 820 is provided by a plurality of ground pillars 822 (only one labelled), arranged to surround the RF signal connector 810. An inner profile 825 defined by at least a portion of the ground connector 820 (e.g., at each end of the transition 800) has a substantially circular shape. An outer profile 827 defined by at least a portion of the ground connector 820 (e.g., at each end of the transition 800) has a non-circular shape, including a plurality of concave cut-outs, each extending radially inward between the plurality of ground connectors 822. In this way, the region of the ground connector 820 closest to the RF signal connector 810 has an isotropic spacing from the RF signal connector 810 at the ends of the transition 800, where it is of most consequence, but is permitted to have a non-isotropic profile for the region of the ground connector 820 further from the RF signal connector 810.

The transition 800 is also provided with a plurality of (in this example, twelve) further signal connectors 880 (only two labelled in Figure 10 to aid clarity). Each of the further signal connectors 880 is provided radially outward of the ground connector 820 and is for providing signal communication of electrical signals (e.g., data signals and/or control signals) between the first PCB and the second PCB via the transition 800. In other words, the further signal connectors 880 are shielded from the RF signal connector 810 by the ground connector 820. The inclusion of the cut-outs allows the further signal connectors 880 to be located more closely to the RF signal connector 810 than would otherwise be possible. Accordingly, a size of the transition 800 can be reduced compared to a version of the transition not having the cut-outs. Viewed another way, more further signal connectors 880 can be provided on a transition 800 of a given size.

It will be understood that the spacing between the connectors 810, 820, 880 of the transition 800 is typically filled by a dielectric material to provide structural stability to the transition 800.

Figure 11 shows a simplified representation of a still further example of a transition in accordance with the present disclosure The transition 900 is viewed from a side angle, showing only the route of an RF signal connector 910 from a first PCB contact pad 912 to be connected to a first PCB (not shown in Figure 11) to a second PCB contact pad 914 to be connected to a second PCB (also not shown in Figure 11). An electrical component 930 (e.g., an electronic component in the form of a radio-frequency integrated circuit, RFIC) is provided in the RF signal path within the RF signal connector 910. The RF signal connector 910 comprises a first portion 911a between the first PCB contact pad 912 and the electrical component 930. The RF signal connector also comprises a second portion 911 b between the second PCB contact pad 914 and the electrical component 920. In this example, the transition 900 comprises three layers of dielectric material 921a, 921 b, 921c in which the RF signal connector 910 and the electrical component 930 is provided. Specifically, the electrical component 930 is provided within the second layer 921 b of dielectric material.

In this example, the electrical component 930 is a heat-generating electrical component 930. Further vias 931a, 931b are provided to extend from the electrical component 930 to an end of the transition 900 to conduct excess heat away from the electrical component 930, off the transition 900 and to the second PCB. It will be understood that the heat could alternatively or additionally be conducted to the first PCB via other vias.

Figure 12 shows a representation of yet another example of a transition in accordance with the present disclosure. The transition 1000 is viewed from a top-down perspective, showing contact pads of multiple connectors for connecting between first and second PCBs (not shown in Figure 12). The transition 1000 is provided in the form of a connector assembly 1000. The connector assembly 1000 has a substantially square cross-sectional shape, having two internal voids 1001a, 1001 b, provided therein. The voids 1001a, 1001b can sometimes be referred to as openings or depressions. In this example, each void 1001a, 1001 b extends through the entire depth of the connector assembly 1000. The first void 1001a is defined by a first portion of a first wall component 1003, a first portion of a second wall component 1005, opposite the first wall component 1003, a third wall component 1007 and a central portion 1009. The second void 1001b is defined by a second portion of the first wall component 1003, a second portion of the second wall component 1005, opposite the first wall component 1003, a fourth wall component 1011 , and the central portion 1009. The third wall component 1007 and the fourth wall component 1011 are opposite and parallel. The central portion 1009 extends between the first wall component 1003 and the second wall component 1005. In some examples, circuit components are embedded within a dielectric region of the connector assembly 1000, with contact pads provided on an external surface of the connector assembly 1000 to provide an external connection to the circuit components (not shown in Figure 12). The connector assembly 1000 includes an RF signal connector 1010 and a plurality of ground connectors 1020, specifically six ground connectors, as part of the central portion 1009. The central portion 1009, the third wall component 1007 and the fourth wall component 1011 , also each include a plurality of further signal connectors 1080. For clarity, only a single further signal connector 1080 in each of the central portion 1009, the third wall component 1007 and the fourth wall component 1011 has been labelled in Figure 12. Nevertheless, in this example it can be seen that there are 34 further connectors 1080 provided in each of the third wall component 1007 and the fourth wall component 1011 , and 29 further connectors 1080 provided in the central portion 1009. In this way, circuit components mounted on either of the first or second PCBs having the connector assembly 1000 connected therebetween, can still be extend from the respective first or second PCB, towards the other PCB, within the footprint of the connector assembly 1000, by being provided within one of the voids 1001a, 1001 b. In this example, the connector assembly has a length of approximately 20 millimetres, a width of approximately 20 millimetres, and a depth (defining a spacing between the PCBs) of approximately 1 millimetre. Typically, the voids 1001a, 1001 b may be flooded with filling material during manufacture of the circuit board assembly comprising the first PCB, the second PCB and the connector assembly, once the circuit components are accommodated in the voids.

Figure 13 shows a representation of the transition of Figure 12, provided between two PCBs. Figure 13 shows a cross-section through the transition and the PCBs. Specifically, the assembly 1100 of Figure 13 includes a first PCB 1102, separated from a second PCB 1104 by a transition assembly, formed from a third component wall 1107, a fourth component wall 1111 and a central portion 1109, each substantially as described with reference to Figure 12 hereinbefore. The third component wall 1107 and the fourth component wall 1111 are each showing a cross-section through a further signal connector 1180. The central portion 1109 is showing a cross section through two ground connectors 1120 and the RF signal connector 1110. A first void 1101a and a second void 1101 b are also substantially as described with reference to Figure 12 hereinbefore. A first PCB component 1194, in the form of an integrated circuit chip 1194, extends from the first PCB 1102 into the first void 1101a, part way towards the second PCB 1104. A second PCB component 1192, also in the form of an integrated circuit chip 1192, extends from the second PCB 1104 into the second void 1101b, part way towards the first PCB 1102. In this way, it can be seen that the connector assembly can be provided even in regions where components are mounted on the PCBs. It will be appreciated that the connector assembly may have dimensions other than the substantially square shape illustrated in Figure 12. In addition, although Figure 12 illustrates two openings, i.e. the first void 1001a and the second void 1001b, the connector assembly may have less than two or more than two openings. Although each opening is depicted as accommodating one circuit component, the one or more openings may each comprise a plurality of circuit components.

In summary, there is provided a method of forming a circuit board assembly, the circuit board assembly comprising a first PCB (110), a second PCB (190), an RF signal connector (120) for carrying RF signals therebetween, and one or more ground connectors (130, 140) for at least partially shielding the RF signal connector. The method comprises: providing first and second PCBs, each having an RF signal conductor and one or more ground conductors thereon, forming or attaching an RF signal connector (120) extending between the first and second PCBs and electrically connecting the RF signal connector to the RF signal conductors of the first and second PCBs, forming or attaching one or more ground connectors (130, 140) extending between the first and second PCBs and electrically connecting the one or more ground connectors to the ground conductors of the first and second PCBs. The one or more ground connectors are arranged to at least partially surround the RF signal connector. The RF signal connector is formed from one or more conductive signal pillars. The or each of the one or more ground connectors are formed from one or more conductive ground pillars.

It may be that an aspect of the invention extends to the following numbered clauses.

1. A method of forming a circuit board assembly, the circuit board assembly comprising a first PCB, a second PCB, an RF signal connector for carrying RF signals therebetween, and one or more ground connectors for at least partially shielding the RF signal connector, the method comprising: providing first and second PCBs, each having an RF signal conductor and one or more ground conductors thereon; forming or attaching an RF signal connector extending between the first and second PCBs and electrically connecting the RF signal connector to the RF signal conductors of the first and second PCBs; and forming or attaching one or more ground connectors extending between the first and second PCBs and electrically connecting the one or more ground connectors to the ground conductors of the first and second PCBs; the one or more ground connectors together being arranged to at least partially surround the RF signal connector, wherein the RF signal connector and the one or more ground connectors are provided in a connector assembly extending between the first and second PCBs and defining one or more openings therein, each opening adjacent one of the first PCB and the second PCB.

2. The method of clause 1 , wherein the first PCB comprises a circuit component extending from a surface of the first PCB, towards the second PCB, and wherein the connector assembly is arranged such that at least one of the one or more openings accommodates the circuit component therein.

3. The method of clause 1 or clause 2, wherein the one or more ground connectors are a plurality of ground connectors, the RF signal connector and the plurality of ground connectors are formed in a third PCB, and wherein the third PCB is connected between the first and second PCBs.

4. The method of clause 3, wherein the third PCB is formed by a build-up process.

5. The method of any one of clauses 3 or 4, wherein the third PCB is formed as a multi-layer structure comprising a plurality of dielectric layers with vias therethrough which function as the RF signal connector and the plurality of ground connectors.

6. The method of clause 1 or clause 2, wherein the RF signal connector and the one or more ground connectors are each formed by a build-up process.

7. The method of any preceding clause, wherein the RF signal connector and the one or more ground connectors are each formed as a multi-layer structure.

8. The method of any one preceding clause comprising the steps of: covering a top surface (the surface opposite the first PCB) of the structure comprising the RF signal connector and the one or more ground connectors (and typically also dielectric therebetween) with a mask layer defining regions for the formation of extended portions of the RF signal connector and the one or more ground connectors; extending the RF signal connector and the one or more ground connectors; and removing the mask layer to thereby form a structure having the RF signal connector and one or more ground connectors extending therefrom.

9. The method of any one preceding clause comprising, after the RF signal connector and one or more ground connectors are formed on or attached to the first PCB: applying an overmould to the first PCB, the overmould partially or wholly covering the RF signal connector and the one or more ground connectors (and typically also dielectric therebetween); removing a top surface (i.e. the surface opposite the first PCB) of the overmould, and typically also of the RF signal connector and the one or more ground connectors; and connecting the thereby exposed RF signal connector and the one or more ground connectors to the second PCB.

10. The method of clause 9, wherein the RF signal connector and the one or more ground connectors are aligned in a plurality of discrete rows, at least some of the rows passing through at least two or at least three said pillars, and the method comprises removing (typically cutting) a top surface of said pillars (and optionally also said overmould) along said discrete rows.

11. The method of any one preceding clause, wherein the one or more ground connectors are two ground connectors which are located on mutually opposite sides of the RF signal connector, or three or more ground connectors which are located in a circumferentially distributed arrangement around the RF signal connector.

12. The method of any one preceding clause, wherein the RF signal connector comprises a first portion formed from a first conductive signal pillar and a second portion formed from a second conductive signal pillar, wherein an RF signal passing through the RF signal connector passes through the second portion subsequent to passing through the first portion.

13. The method of clause 12, further comprising forming or attaching an electrical component between the first conductive signal pillar and the second conductive signal pillar, wherein the electrical component is to alter an RF signal passing through the RF signal connector.

14. The method of clause 13, wherein the electrical component is an integrated circuit component.

15. The method of any of clauses 12 to 14, wherein the first conductive signal pillar is offset from the second conductive signal pillar.

16. A circuit board assembly comprising: first and second PCBs, each having an RF signal conductor and one or more ground conductors thereon; an RF signal connector extending between the first and second PCBs and electrically connected to the RF signal conductors of the first and second PCB; and one or more ground connectors extending between the first and second PCBs and electrically connected to the ground conductors of the first and second PCBs; wherein the one or more ground connectors are arranged to at least partially surround the RF signal connector, and wherein the RF signal connector and the one or more ground connectors are provided in a connector assembly extending between the first and second PCBs and defining one or more openings therein, each opening adjacent one of the first PCB and the second PCB.

17. The circuit board assembly of clause 16, wherein the first PCB comprises a circuit component extending from a surface of the first PCB, towards the second PCB, and wherein the connector assembly is arranged such that at least one of the one or more openings accommodates the circuit component therein.

18. The circuit board assembly of clause 16 or clause 17, wherein the one or more ground connectors are a plurality of ground connectors, the RF signal connector and the plurality of ground connectors are formed in a third PCB, and wherein the third PCB is connected between the first and second PCBs. 19. The circuit board assembly of clause 16 or clause 17, wherein the RF signal connector and the one or more ground connectors are each formed by a buildup process.

20. The circuit board assembly of any one of clauses 16 to 19, the circuit board assembly further comprising an overmould which partially or wholly covers the RF signal connector and the one or more ground connectors except at a top surface (i.e. the surface opposite the first PCB) of the RF signal connector and the one or more ground connectors.

21 . The circuit board assembly of any one of clauses 16 to 20, wherein the one or more ground connectors are two ground connectors which are located on mutually opposite sides of the RF signal connector, or three or more ground connectors which are located in a circumferentially distributed arrangement around the RF signal connector.

22. The circuit board assembly of any one of clauses 16 to 21 , wherein the RF signal connector comprises a first portion formed from a first conductive signal pillar and a second portion formed from a second conductive signal pillar, wherein an RF signal passing through the RF signal connector passes through the second portion subsequent to passing through the first portion.

23. The circuit board assembly of clause 22, further comprising an electrical component provided between the first conductive signal pillar and the second conductive signal pillar, the electrical component to alter an RF signal passing through the RF signal connector.

24. The circuit board assembly of clause 23, wherein the electrical component is an integrated circuit component.

25. The circuit board assembly of any one of clauses 16 to 24, further comprising one or more further signal connectors arranged to carry electrical signals, the one or more further signal connectors each extending between the first PCB and the second PCB and arranged to be shielded from the RF signal connector by the one or more ground connectors. Throughout the description and claims of this specification, the words “comprise” and “contain” and variations of them mean “including but not limited to”, and they are not intended to and do not exclude other components, integers or steps. Throughout the description and claims of this specification, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise.

Features, integers, characteristics or groups described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein unless incompatible therewith. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and/or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and/or steps are mutually exclusive. The invention is not restricted to the details of any foregoing embodiments. The invention extends to any novel one, or any novel combination, of the features disclosed in this specification (including any accompanying claims, abstract and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed.