High Density Flip Chip Package for Wireless Transceivers.

Abstract

An RF flip chip is provided that includes a balun with a centrally located bump in a local bump area adjacent to a die corner. The centrally located bump can be floating for signaling within the balun. The local bump area includes a crackstop area at the die corner where the bump is excluded. The local bump area also includes at least one output bump for the balun.

Description

[Technical field]

[0001] Cross-reference to related art This application claims priority to and the benefit of U.S. Nonprovisional Patent Application No. 16 / 875,972, filed May 15, 2020, which is incorporated by reference in its entirety as if fully set forth below and for all applicable purposes.

[0002] This application relates to integrated circuit packaging, and more particularly to high density flip chip packages for wireless transmitters and receivers. [Background technology]

[0003] Compared to purely digital devices, radio frequency (RF) systems for wireless applications generally require several passive devices for filtering and matching purposes. These passive devices are difficult to integrate on a single semiconductor die with other transceiver components. Similarly, various active components in a wireless transceiver, such as power amplifiers, are often created using semiconductor technologies that require different semiconductor technology compared to more digitally based components such as modems. Thus, a transceiver for a wireless device such as a mobile phone has several diverse semiconductor dies and passive components. It is beneficial to integrate these various individual dies and passive components into an RF system-in-package (SiP). For example, RF SiP offers several advantages, such as improved power efficiency, streamlined design, and reduced noise. Despite these advantages, the design goal of increasing density must overcome many challenges. Summary of the Invention [Means for solving the problem]

[0004] According to a first aspect of the present disclosure, an RF flip chip die is provided that includes a crack stop region including a corner of the die and a local bump region abutting the crack stop region, the local bump region including a first balun formed of a coil arranged about a central region and a die interconnect located within the central region.

[0005] According to a second aspect of the present disclosure, a method of RF signaling is provided, the method including: injecting a differential RF signal into a first coil of a balun located in a local bump area abutting a corner area for an RF flip chip die, the first coil being formed around a central balun area including a floating copper pillar for the local bump area; and producing a single-ended RF signal in a second coil of the balun in response to driving the differential RF signal.

[0006] According to a third aspect of the present disclosure, an RF flip chip is provided that includes a semiconductor substrate having a corner region around a corner of the semiconductor substrate, a plurality of metal layers adjacent the corner region and configured to form a first balun having a coil, a first copper pillar, and a top dielectric layer having a portion located between a central region of the coil and the first copper pillar.

[0007] According to a fourth aspect of the present disclosure, a method of RF signaling is provided, the method including: injecting a single-ended RF signal into a first coil of a balun located in a local bump area abutting a corner area for an RF flip chip die, the first coil being formed around a central balun area including a floating copper pillar for the local bump area; and producing a differential RF signal in a second coil of the balun in response to driving the single-ended RF signal.

[0008] These and other advantageous features will be better understood through the following detailed description of the invention. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 illustrates a conventional RF flip chip in which the local bump area has a uniform distribution of bumps. [Diagram 2] FIG. 2 illustrates an RF system including an RF flip chip with local bump areas according to one embodiment of the present disclosure. [Diagram 3] 1 illustrates mounting an RF flip chip with local bump areas onto a carrier substrate according to one embodiment of the present disclosure. [Figure 4A] FIG. 2 is a simplified perspective view of a balun relative to a local bump area according to one aspect of the present disclosure. [Figure 4B] FIG. 4B is a plan view of the balun of FIG. 4A. [Diagram 5] FIG. 2 is a plan view of a local bump region according to one aspect of the present disclosure. [Figure 6] FIG. 1 illustrates a method of RF signaling using a balun in a local bump region according to one embodiment of the present disclosure. [Figure 7] 1A-1C illustrate several exemplary electronic systems each incorporating an RF flip chip according to an aspect of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] The embodiments of the present disclosure and their advantages are best understood by referring to the following detailed description, it being understood that like reference numerals are used to identify like elements shown in one or more of the figures.

[0011] The integration of a die into an RF integrated circuit package, such as a SiP, may use several techniques, such as wire bonding and flip chip technology. Flip chip mounting of a die is particularly attractive due to the increased density of interconnects it offers compared to wire bonding. In both types of packaging, the die has an active side on which active devices, such as transistors, are formed. This active side faces away from the underlying side, such as the package carrier in wire bonding applications. However, in flip chip packaging, the active side of the die faces the package carrier. Thus, the "flip" in flip chip packaging comes from the fact that the die is "flipped" upside down compared to its orientation in traditional wire bonding applications.

[0012] The interconnects between the flip chip die and the underlying surface may be formed using solder balls. To further increase density, these interconnects may be formed using metal pillars (e.g., copper pillars). In the following description, it will be understood that the term "bump" refers to the generic flip chip interconnect, whether the interconnect is a solder ball or a metal pillar. However, it should be noted that solder bumps have an inherent spherical shape that limits the bump density. In contrast, metal pillars such as copper pillars can be spaced apart with greatly improved bump density. Thus, the following description is directed, without loss of generality, to an embodiment in which the bumps are copper pillars. However, even with the use of copper pillars to form the bumps, several obstacles prevent further improvements in density. For example, traditionally, dies have been formed from a semiconductor wafer that contains the layout for many other dies. To separate the wafer into individual dies, the wafer is sectioned in the Cartesian x and y directions such that each die is rectangular. The rectangular shape for the die results in mechanical stresses at the die corners. Therefore, to limit this mechanical stress, there is a crack stop area at each die corner. Within the crack stop area, no bumps are allowed. However, across the remaining active surface of the die, the bump density must meet a minimum threshold density to properly secure the die to the underlying surface. In addition, there is also a minimum separation distance between each bump. These two factors of minimum bump density and minimum bump separation create density problems, for example, in the square area of ​​the die that includes the die corners and its crack stop area (referred to herein as local bump area). The definition of the local bump area depends on the semiconductor manufacturing requirements, but it is assumed herein that when the minimum bump separation is defined by the pitch p, the width and length of the local bump area is 3*p in some semiconductor processing technologies. The minimum bump separation or pitch p may be defined as the minimum separation of the center of a bump to the center of an adjacent bump. In one semiconductor processing node, p is equal to 130 μm.However, it will be appreciated that the pitch p and the width and length of the local bump area will vary according to semiconductor processing node or technology.

[0013] The crack stop regions may form right triangles with the right vertices being the die corners. If the bumps are evenly distributed across the 3*p length and width of the local bump region, such a local bump region has six bumps. This may be better understood with respect to the exemplary die 100 shown in FIG. 1. The die 100 has four corners and therefore four crack stop regions 105 and four associated local bump regions 110, each with six bumps 115. If the bump distribution is not uniform, additional bumps may be added. Regardless of the number of bumps 115, their inclusion within each local bump region 110 has traditionally limited the types of components that may be integrated into the active surface (not shown) of the die below each local bump region 110.

[0014] In general, a local bump area may be defined by a rectangular area of ​​the die that includes the die corners and has a length and width that are greater than the bump pitch p. Because the base and length of the crackstop area are the bump pitch p, the resulting rectangular area of ​​the die will therefore include the crackstop area. The local bump area may be further defined to include what is left of the rectangular area when the crackstop area is excluded. Thus, the local bump area may be the area of ​​the die that abuts the crackstop area. Because the local bump area abuts the crackstop area that includes the die corners, the local bump area may also be referred to herein as a corner area of ​​the die. In some embodiments, the local bump area may have a shape other than a rectangle or a rectangle with the corners removed.

[0015] The local bump regions 110 and the example die 100 are not necessarily drawn to scale in Figure 1. The die 100 may be sized and / or shaped such that the local bump regions 110 are spaced further apart than shown. Additionally, the aspect ratio and / or shape of the die 100 may differ from that shown.

[0016] One type of component commonly used in RF front-end applications is a balun. For example, the die 100 may be a transceiver die that includes various RF components, such as mixers and / or filters. Such transceivers may use differential signaling, for example, due to their favorable common-mode interference rejection. However, commonly used antennas, such as patch antennas in mobile phone applications, may not be suitable to be driven by differential RF signals, or it may not be beneficial to route differential signals to such elements. In these configurations, a single-ended RF signal may be used to drive the antenna. Similarly, power amplifiers for amplifying RF signals before they drive the antenna are also typically single-ended. Power amplifiers for wireless applications often use semiconductor processes that are not suitable for the processes used to build transceivers, so the transceiver and power amplifier may be located on separate dies. The balun converts the differential RF signal to a single-ended RF signal, so that the transceiver die can feed the resulting single-ended RF signal into the power amplifier die. The resulting die may be integrated into an RF front-end system-in-package, or alternatively may remain as separate devices or may be packaged into modules coupled via circuit boards.

[0017] An exemplary RF front-end system 200 is shown in Figure 2. The transceiver die 205 includes an RF driver or amplifier 210 that drives a differential RF signal through a first coil 245 in a balun 215. In response to the differential signaling, a second coil 250 in the balun 215 creates a single-ended RF signal that is sent out from the transceiver die 205 on the bump 235. The other end of the second coil 250 is grounded.

[0018] Power amplifier die 230 receives a single-ended RF signal on bump 240. Alternatively, die 230 may be wire bonded. In a SiP embodiment for system 200, dies 205 and 230 may both be flip-chip mounted to a carrier (not shown). Alternatively, dies 205 and 230 may be packaged separately, whereby their integration in system 200 is via a circuit board. Power amplifier 220 in die 230 amplifies the single-ended RF signal to drive antenna 225 through another bump 255.

[0019] It should be noted that the balun 215 may instead function during a receive mode of operation. For example, the die 230 may include a low noise amplifier (not shown) that amplifies the receive RF signal from the antenna 225. The resulting amplified receive RF signal drives a second coil 250 in the balun 215 to create a differential receive RF signal at the first coil 245.

[0020] The balun 215 is a passive component that is not integrated into the active surface of the die. Instead, the balun 215 may be formed in a metal layer adjacent to the active surface of the transceiver die 205. Some exemplary metal layers in a flip chip die 300 are shown in FIG. 3. The die 300 has an active surface 305 on which various active components, such as transistors, are integrated. The active surface 305 is separated from various metal layers by corresponding dielectric layers. For clarity, only a first metal layer M1 and a second metal layer M2 are shown, but it will be appreciated that a larger number of metal layers are typically provided. As is known in the semiconductor art, metal layers such as metal layers M1 and M2 are patterned to form various conductors for signals to the die 300, as well as to provide power and ground. Additionally, the metal layers may be patterned to form passive components, such as the inductive coil (FIG. 2) in the balun 215. After the metal layers are patterned as desired, pillars 310 may be deposited, for example, via electroplating. The pillars 310 are attached to the final or outermost dielectric layer via under bump metallization (UMB) regions 315. Referring again to FIG. 1, each crack stop region 105 may also be referred to as a UBM prohibited region since there are no UMB features within the crack stop region 105. The die 300 is then mounted onto the substrate 320 via fusing the substrate 320 to the pillars 310. The substrate 320 has similar metal layers for signal, power, and ground distribution. In a SiP embodiment, the substrate 320 may be a carrier that also supports the power amplifier die 230. Alternatively, the substrate 320 may be a circuit board in an embodiment in which the transceiver die 205 and the power amplifier die 230 are packaged separately.

[0021] The patterned metal layers forming the balun 400 are shown in a perspective view in FIG. 4A. In this simplified view, the metal layer 405 is patterned into an open-ended loop to form the first coil 410. Similarly, the metal layer 415 is patterned into another open-ended loop to form the second coil 420. In general, each coil may be distributed across two or more metal layers with multiple loops in each metal layer. Additionally, while the first coil 410 is shown as being above the second coil 420, in alternative embodiments, this order may be reversed or the coils may be interleaved. Thus, it will be appreciated that the balun 400 is shown in a simplified form for clarity. In the balun 400, the second coil 420 includes two input ports 425 for receiving differential RF signals. The first coil 410 includes an output port 430 for a single-ended RF signal, as well as an output port for ground. Each coil surrounds a central region 435, as shown in the plan view of FIG. 4B.

[0022] Referring again to FIG. 1, baluns such as balun 215 or balun 400 are generally excluded from the LBR 110. The required bump density results in signals that interfere with the operation of such baluns. For this reason, conventionally, die space beneath each LBR 110 has been used to integrate other components such as input / output (I / O) switches and attenuators, electrostatic discharge (ESD) protection circuits, and digital block transceiver control circuits. However, the use of such conventional components does not necessarily fully occupy the die space area beneath each LBR 110. For example, the common utilization factor for the LBR die space is approximately 55%. To dramatically improve die space utilization for the LBR 110 (and thus increase density), an LBR layout is provided that includes at least one balun.

[0023] Referring again to FIG. 2, it is noted that the associated active circuitry for the balun 215, such as the RF amplifier 210, is beneficially located on the active die surface directly below the balun 215. Typically, the RF differential signal that the RF amplifier 210 feeds into the balun 215 is relatively strong and therefore tends to inadvertently couple strongly to other signals in the transceiver die 205. However, this coupling is reduced by keeping the associated circuitry in close proximity to the balun 215. Referring now to FIG. 5, an exemplary LBR 500 is shown in which the balun 505 is substantially contained within the LBR 500. The location of the balun 505 within the LBR 500 allows the transceiver die 205 to locate the associated circuitry, such as the RF amplifier 215, within the LBR 500, to also significantly improve density. For example, the utilization factor for the transceiver die 205 within the LBR 500 can be increased to 80% or even 90% or more. Thus, density is significantly improved compared to a conventional utilization factor of 55%. In this embodiment, the balun 505 is completely contained within the LBR 500. However, a portion of the balun 505 may be outside the LBR 500 in alternative embodiments. To maintain a desired bump density, the pillar 510 is placed within a central region 435 (FIG. 4B) of one or more of the coils in the balun 505. Referring again to FIG. 4A, the balun 505 has multiple stacked metal layer loops, whereby the first and second coils may be magnetically coupled to each other. The pillar 510 may be placed substantially at the center of these loops. To reduce any coupling to the differential and single-ended RF signals processed by the balun 505, the pillar 510 is floating with respect to these RF signals. As used herein, a pillar is considered to be floating with respect to a signal if there is no conductive wiring or lead in the metal layers for the signal that connects to the pillar through a via. In some embodiments, the pillar 510 may be grounded or floating. In that regard, if a pillar is defined as floating, with no association with any signal, then such pillar is not conductively connected to any signal node or to any DC node, such as ground, or to a power node.It should be noted that while pillars 510 may be floating, the metal layer underneath pillars 510 may be patterned accordingly in a "dummy" manner. In other words, even if no signal is directed towards pillars 510, the metal layer is patterned to support pillars 510, e.g., to provide greater structural integrity.

[0024] A transceiver die, such as the transceiver die 205 (FIG. 2) under the LBR 500, may operate on several frequency bands. The RF signals generated for each frequency band are processed by a corresponding balun. For example, the transceiver die 205 may require a low-band balun, a mid-band balun, a high-band balun, etc. Each balun is configured to operate within its corresponding frequency regime. For example, a low-band balun is configured for a relatively low-frequency RF carrier signal. Similarly, a mid-band balun is configured for a mid-frequency RF carrier signal, which is a higher frequency than the relatively low-frequency RF carrier signal. However, it should be noted that the mid-band may partially overlap with the low-band. In general, the transceiver die 205 may have any number of such bands and corresponding baluns, the outputs of which may be directed to components, such as respective power amplifiers, configured to further process the signals within the respective bands. The pillars 510 are floating with respect to any RF signals, however, in a multi-band embodiment, the balun 505 may be a low-band (LB) balun 505 to further reduce interaction of the pillars 510 with RF signals within the LB balun 505.

[0025] LBR 500 also contains a portion of another balun, such as mid-band (MB) balun 535. It will be appreciated that the partial inclusion of such another balun is optional. In general, since the single-ended RF signals propagating to power amplifier die 230 (FIG. 2) are relatively strong, it is beneficial to have the balun located near the periphery of transceiver die 205. To reduce power losses for such signals, the routing distance between a balun output bump, such as bump 235, and power amplifier die bump 240 should be reduced, which may result in the balun being located along the periphery of transceiver die 205. In some embodiments, the carrier frequency for mid-band balun 535 may be about 3 GHz. Similarly, in some embodiments, the carrier frequency for low-band balun 505 may be about 1 GHz.

[0026] Referring again to the LBR 500, the positioning of the output pillars similar to the bumps 235 will now be described. With regard to this positioning, it should be noted that operation within any given frequency band may be further divided into frequency sub-bands or for operation having different signaling protocols. For example, the MB balun 535 may be switched between operation in a first sub-band and operation in a second sub-band. Thus, the transceiver die 205 may have a band switch (not shown) having a first configuration in which the MB balun 535 is driven by a first mid-band sub-band differential RF signal to create a first mid-band sub-band (MB1) single-ended RF signal at the output MB1 pillar 530. Similarly, there may be a second configuration in which the MB balun 535 is driven by a second mid-band sub-band differential RF signal to create a second mid-band sub-band (MB2) single-ended RF signal at the output MB2 pillar 525. The pillars 530 and 525 are located near the periphery of the LBR 500 and also near the periphery of the corresponding transceiver die 205. As is known in the semiconductor art, the periphery of the transceiver die 205 is associated with a seal ring formed in a corresponding metal layer. This seal ring is represented by dashed line 515 in FIG. 5. Because the output MB1 pillar 530 and the output MB2 pillar 525 are adjacent to the die periphery, these pillars are adjacent to the seal ring to one side of the crack stop region 105 (and therefore to one side of the apex or corner of the transceiver die 205). The edge of the LB balun 505 is also adjacent to the seal ring to the other side of the crack stop region (and therefore to the other side of the same die corner). Because the LB balun 505 is adjacent to the die periphery on one side of the die corner in the LBR 500, while the output MB1 pillar 530 and the output MB2 pillar 325 are adjacent to the die periphery on the other side of the die corner in the LBR 500, there is left room in the LBR 500 for only one output pillar for the LB balun 505 to be adjacent to the die periphery. As previously explained, there may be a minimum separation distance p between pillars in the LBR 500.As used herein, a bump is considered adjacent to the die periphery in LBR 500 if such bump is placed within a distance p to the die periphery or to the inner edge of crack stop region 105. Given this proximity, in some manufacturing and / or design processes, no other bumps may be placed between the bump adjacent to the die periphery and the die periphery.

[0027] The LB balun 505 is similar to the MB balun 535 in that it has a first low band sub-band (output LB1) pillar 545 and a second low band sub-band (output LB2) pillar 550. As mentioned above, there is room left for only one such output bump at the die periphery within the LBR 500. In one embodiment, it is the output LB2 pillar 550 adjacent the die periphery, but in alternative embodiments, it may be the output LB1 pillar 545. Note that a pillar may be considered to be within the LBR 500 even if the pillar is only partially contained within the LBR 500, as is the case for the output LB1 pillar 545 and the output LB2 pillar 550. The output LB1 pillar 545 is located along the inner periphery of the LBR 500 and is separated from the output LB2 pillar 550 by approximately the minimum separation distance P. Additionally, along the same inner circumference of the LBR 500 near the inner corners of the LBR 500, there is a ground (GND) pillar 540 that may serve as a ground pillar for the MB balun 535. A balun GND pillar 555 located between the MB balun 535 and the LB balun 505 in the LBR 500 serves as a ground pillar for the LB balun 505. Finally, there is sufficient room left between the LB balun 505 and the crack stop region 105 for a seal ring pillar 520 that covers the seal ring and may serve as either a ground pillar or a float. While the above discussion has been focused on the integration of the balun 505 into the LBR 500, it should be noted that the balun 505 may be replaced with a single coil that forms a separate inductor. The resulting single coil thus has a central area that may be advantageously occupied by a bump similar to that described for the pillar 510.

[0028] Referring again to FIG. 1, it is recalled that an LBR having a length and width three times the bump pitch p requires a minimum of six bumps to meet the minimum required bump density. It will be appreciated that if the LBR 500 has the same minimum bump density of six bumps, the LBR may be rearranged in many ways while still meeting the minimum bump density. For example, if the mid-band balun 535 is removed from the LBR 500, the pillars 530 and 525 may instead be used as output ports for the LB balun 505. Thus, in such an embodiment, the pillars 545 and 550 may be removed or utilized for different purposes. In general, it is advantageous for the output pillars for the balun to be adjacent to the balun to reduce the possibility of coupling to other signals. Note that the LB balun 505 has a bottom edge adjacent to the die periphery, thereby leaving no room for the pillars between this bottom edge and the die periphery. Thus, for such baluns, the output pillars may be placed on either side of the balun to reduce the routing distance between the balun and the output pillars. In the LB balun 505, the output pillars 545 and 550 are on the left side of the balun 505, but in alternative embodiments, they may be replaced on the right side. In contrast, replacing the ground pillar 555 with an output pillar is of little benefit since the pillar 555 is relatively moved further away from the die periphery. Thus, there are many ways in which the pillars may be arranged within an LBR including a pillar-containing balun according to the concepts disclosed herein. For example, in alternative embodiments, the LBR 500 may have six pillars, seven pillars, or more than eight pillars.

[0029] A method of RF signaling will now be described with respect to the flow chart shown in FIG. 6. The method includes act 600 of injecting a differential RF signal into a first coil of a balun located in a local bump area abutting a corner area for an RF flip chip die, the first coil being formed around a central balun area including a floating copper pillar for the local bump area. An example of act 600 is driving a LB balun 505 with a differential RF signal. The method further includes act 605 of creating a single-ended RF signal in a second coil of the balun in response to driving the differential RF signal. An example of act 605 is creating a single-ended RF signal at one of the output pillars for the LB balun 505.

[0030] The method of receiving an RF signal is similar. For example, such a method may include injecting a single-ended RF signal into a first coil of a balun located in a local bump area abutting a corner area for an RF flip chip die, the first coil being formed around a central balun area including a floating copper pillar for the local bump area. One example of such an act is injecting the received RF signal into the LB balun 505. In addition, the method of receiving an RF signal may include creating a differential RF signal in a second coil of the balun in response to driving the single-ended RF signal. One example of such an act is creating a received RF differential signal in the LB balun 505.

[0031] The RF flip chips disclosed herein may be advantageously incorporated into any suitable mobile device or electronic system. For example, as shown in Figure 7, a mobile phone 700, a laptop computer 705, and a tablet PC 710 may all include an RF flip chip according to the present disclosure. Other exemplary electronic systems, such as communication devices or personal computers, may also be configured with an RF flip chip constructed according to the present disclosure.

[0032] It will be appreciated that numerous modifications, substitutions, and variations may be made in or to the materials, apparatus, construction, and methods of use of the devices of the present disclosure without departing from the scope thereof. In light of this, the specific embodiments shown and described herein are merely some examples thereof, and the scope of the present disclosure should not be limited to the scope of such specific embodiments, but rather should be entirely within the scope of the following appended claims and their functional equivalents. [Explanation of symbols]

[0033] 100 Die 105 Crackstop Area 110 Local Bump Region (LBR) 115 Bump 200 RF Front-End System 205 Transceiver die 210 RF Drivers, RF Amplifiers 215 Balan 220 Power Amplifier 225 Antenna 230 Power Amplifier Die 235 Bump 240 Bump 245 First Coil 250 Second coil 255 Bump 300 Flip Chip Dies 305 Active Surface 310 Pillar 315 Under Bump Metallization (UMB) Area 320 Substrate 400 Balun 405 Metal layer 410 First coil 415 Metal layer 420 Second coil 425 input port 430 output port 435 Central area 500 LBR 505 Balun 510 Pillar 515 dashed line 520 Seal Ring Pillar 525 Output second mid-band sub-band (MB2) pillar, output MB2 pillar 530 output first mid-band sub-band (MB1) pillar, output MB1 pillar 535 MB Balun 540 Ground (GND) pillar 545 1st low band sub-band (output LB1) pillar 550 Second Low Band Sub-Band (Output LB2) Pillar 555 Balun GND pillar 700 Mobile Phones 705 Laptop Computer 710 Tablet PC

Claims

1. a crack stop region including a corner of the die; a local bump region abutting the crack stop region, the local bump region comprising: a first balun formed by coils arranged around a central region; and a die interconnect located in the central region.

2. The RF flip chip die of claim 1 , wherein the die interconnects are solder balls.

3. The RF flip chip die of claim 1 , wherein the die interconnects are metal pillars, preferably the metal pillars are copper pillars.

4. The RF flip chip die of claim 3 , wherein the local bump area further comprises a portion of a second balun.

5. The RF flip chip die of claim 4 , wherein the copper pillar is configured to float with respect to an RF signal carried by the first balun.

6. The RF flip chip die of claim 4 , wherein the pillars are grounded.

7. 6. The RF flip chip die of claim 5, wherein the first balun is configured to process a low band differential carrier RF signal, and the second balun is configured to process a mid band differential carrier RF signal having a higher frequency than the low band differential carrier RF signal.

8. 6. The RF flip chip die of claim 5, wherein an edge of the first balun is within a minimum bump separation distance to a periphery of the die on a first side of the crack stop region, and the local bump area includes a first output pillar for the second balun adjacent to a periphery of the die on a second side of the crack stop region.

9. the local bump area includes a first output pillar for the first balun adjacent an inner edge of the local bump area; The local bump region is a second output pillar for the first balun; The RF flip chip die of claim 8 further comprising a second output pillar for the second balun.

10. The RF flip chip die of claim 8 further comprising a seal ring pillar located in a portion of the local bump area between the crack stop area and the first balun.

11. 6. The RF flip chip die of claim 5, wherein the RF flip chip die is attached to a substrate for an RF system-in-package through the copper pillar, the RF system-in-package further comprising a power amplifier configured to amplify a single-ended RF signal from the first balun.

12. The RF flip chip die of claim 1 , wherein the RF flip chip die is integrated into a mobile phone.

13. The RF flip chip die of claim 5 , wherein the width and height of the local bump area is three times the minimum bump separation distance.

14. 1. A method of RF signaling comprising: injecting a differential RF signal into a first coil of a balun located within a local bump area for an RF flip chip die, the first coil being formed around a central balun area that includes a floating copper pillar for the local bump area; and producing a single-ended RF signal in a second coil of the balun in response to the step of injecting the differential RF signal.

15. amplifying the single-ended RF signal in a power amplifier to create an amplified single-ended RF signal; and driving an antenna with the amplified single-ended RF signal.

15. The method of claim 14, wherein amplifying the single-ended RF signal in the power amplifier comprises amplifying the single-ended RF signal in an RF system-in-package.

Images