Packaging structure and semiconductor structure

By adopting specific arrangement methods and symmetrical designs in semiconductor chip ball grid arrays, the complexity and cost of chip designs of different specifications are solved, and compatibility and performance improvements are achieved.

WO2024244930A1PCT designated stage expired Publication Date: 2024-12-05RUILI INTEGRATED CIRCUIT CO LTD
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
PCT/CN2024/091816
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-29
Filing Date
2024-05-09
Publication Date
2024-12-05

AI Technical Summary

Technical Problem

The ball grid array design of existing semiconductor chips is affected due to different signal types and level variation patterns, and chips of different specifications require different substrate designs, resulting in complex designs and high cost.

Method used

Design a package structure in which the ball grid array is arranged in a specific manner in row and column directions, allowing flexible distribution of data ball grids and command address ball grids, supporting 4-bit, 8-bit or 16-bit data bit width, through symmetry The design of the isolation area reduces the number and area of ​​the ball grid, and is compatible with chips of different specifications.

Benefits of technology

Compatibility with chips of different specifications is achieved, substrate area is reduced, signal transmission quality and signal density is improved, design and production costs are reduced, and solder joint life is extended.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2024091816_05122024_PF_FP_ABST
    Figure CN2024091816_05122024_PF_FP_ABST
Patent Text Reader

Abstract

A packaging structure and a semiconductor structure. A ball grid array is provided on the surface of a packaging substrate in the packaging structure, wherein the ball grid array comprises a plurality of data ball grids, at most two data ball grids are allowed to be consecutively arranged in a second direction, and the second direction is a row extension direction of the ball grid array. The packaging structure has a small size.
Need to check novelty before this filing date? Find Prior Art

Description

Packaging structure and semiconductor structure

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on May 29, 2023, with application number 202310625033.4 and application name “A Packaging Structure and Semiconductor Structure”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The embodiments of the present disclosure relate to the field of semiconductor technology, and in particular to a packaging structure and a semiconductor structure. Background Art

[0003] Currently, semiconductor chips (such as dynamic random access memory (DRAM)) involve various types of input / output signals (such as command address signals, data signals, and data strobe signals), which need to be packaged using ball grids. However, due to the different number, level variation patterns, and functions of different types of signals, the arrangement of their corresponding ball grids affects chip performance.

[0004] Summary of the Invention

[0005] According to a first aspect of an embodiment of the present disclosure, there is provided a packaging structure, including:

[0006] A packaging substrate, wherein a ball grid array is provided on the surface of the packaging substrate, wherein the ball grid array includes a plurality of data balls, and the data balls are used to transmit data signals; in a second direction, at most two of the data balls are allowed to be arranged continuously; the second direction refers to the row extension direction of the ball grid array.

[0007] In some embodiments, the data balls of the first portion are located in edge rows and non-edge columns of the ball grid array, and any data ball grid of the first portion is not linearly adjacent to other data balls; the data balls of the second portion are located in non-edge rows and non-edge columns of the ball grid array, and every two data balls of the second portion are adjacent along the second direction; wherein, among any two adjacent rows of the ball grid array, only one row has the data balls distributed therein.

[0008] In some embodiments, the ball grid array further includes a plurality of command address balls, which are used to transmit command / address signals. The command address balls in the first portion are located in edge rows and non-edge columns of the ball grid array, and no command address ball in the first portion is linearly adjacent to any other command address ball. The command address balls in the second portion are located in non-edge rows and non-edge columns of the ball grid array, and at most two command address balls in the second portion are allowed to be arranged continuously in the second direction, and at most three command address balls in the second portion are allowed to be arranged continuously in the first direction. The first direction refers to the column extension direction of the ball grid array.

[0009] In some embodiments, the ball grid array has a total of N rows, the data ball grid of the first part is located in the 1st row, the data ball grids of the second part are all located in rows 2 to A, the command address ball grids of the second part are all located in rows B to (N-1), and the command address ball grid of the first part is located in the Nth row; A, B and N are all positive integers, A is less than B, and B is less than N-1.

[0010] In some embodiments, the data ball grids located in the 1st to (C-1)th rows are used to transmit high-bit data signals, and the data ball grids located in the (C+1)th to Ath rows are used to transmit low-bit data signals; or, the data ball grids located in the 1st to (C-1)th rows are used to transmit low-bit data signals, and the data ball grids located in the (C+1)th to Ath rows are used to transmit high-bit data signals; the ball grids located in the Cth row are all ground ball grids or power ball grids, and the ground ball grids are used to transmit ground signals, and the power ball grids are used to transmit power signals; wherein, C = (A+1) / 2.

[0011] In some embodiments, the ball grid array is divided into a first array, a central array and a second array. The first array, the central array and the second array are arranged sequentially along the second direction. The number of ball grids in the first array is the same as the number of ball grids in the second array, and all ball grids in the central array are removed.

[0012] The data ball grid located in the first array and the data ball grid located in the second array are symmetrical about the central array, and the command address ball grid located in the first array and the command address ball grid located in the second array are symmetrical about the central array.

[0013] In some embodiments, the ball grid array further includes a plurality of data strobe balls configured to transmit data strobe signals; each of the data strobe balls is adjacent to the center array in the second direction, and the data strobe balls are allowed to be adjacent to the data balls in the first direction and in the second direction. The ball grid array further includes a plurality of mask control balls configured to transmit mask flip control signals or terminal data strobe signals; the mask control balls are symmetrical to the data strobe balls about the center array.

[0014] In some embodiments, the ball grid array further includes a local data strobe ball grid, a local data ball grid, and a calibration indicator ball grid; the local data strobe ball grid is located in an edge column and is adjacent to the data ball grid along a second direction; the local data strobe ball grid and the local data ball grid are symmetrical about the center array; the calibration indicator ball grid is adjacent to the local data strobe ball grid along a first direction; wherein the local data ball grid is used to transmit a local data signal, the local data strobe ball grid is used to transmit a local data strobe signal, and the calibration indicator ball grid is used to transmit an impedance calibration signal.

[0015] In some embodiments, the ball grid array also includes a plurality of chip select ball grids adjacent along a first direction, and the chip select ball grids are used to transmit chip select signals; a portion of the chip select ball grids is adjacent to one of the data ball grids along the first direction; another portion of the chip select ball grids is adjacent to one of the command address ball grids along the second direction, and is adjacent to another one of the command address ball grids along the first direction.

[0016] In some embodiments, the ball grid array also includes a plurality of clock ball grids adjacent along a first direction, and the clock ball grids are used to transmit clock signals; a portion of the clock ball grids is not linearly adjacent to any of the command address ball grids, and is not linearly adjacent to any of the data ball grids; another portion of the clock ball grids is adjacent to one of the command address ball grids along a second direction, and is adjacent to another one of the command address ball grids along the first direction.

[0017] In some embodiments, multiple chip select balls are located in the first array and adjacent to the edge columns of the ball grid array; multiple clock balls are located in the second array and adjacent to the center array; and one chip select ball and one clock ball are located in the same row.

[0018] In some embodiments, the ball grid array further includes a parity error indication ball grid and a mirror mode ball grid, and both the parity error indication ball grid and the mirror mode ball grid are located in edge columns of the ball grid array; the mirror mode ball grid is adjacent to one of the chip select ball grids along the second direction; the error indication ball grid is adjacent to another of the chip select ball grids along the second direction; wherein the mirror mode ball grid is used to transmit at least a mirror mode enable signal, and the error indication ball grid is used to transmit a parity error indication signal.

[0019] In some embodiments, the ball grid array further includes a plurality of reserved balls, one test mode ball, and two protocol balls. The test mode ball is used to transmit a test mode enable signal, and the protocol ball is used to transmit a protocol control signal. The plurality of reserved balls and the plurality of chip select balls are symmetrical about the central array. The second protocol ball is symmetrical about the central array with the mirror mode ball, and the test mode ball is symmetrical about the central array with the error indication ball. The first protocol ball, the second protocol ball, and the test mode ball are arranged consecutively along a first direction.

[0020] In some embodiments, the ball grid array further includes a termination ball grid and a reset ball grid, wherein the termination ball grid and the reset ball grid are located in edge columns of the ball grid array; the termination ball grid and the reset ball grid are symmetrical about the center array; linearly adjacent ball grids of the termination ball grid are all ground ball grids and / or power ball grids; the termination ball grid and at least one of the command and address ball grids are located in the same row, and the reset ball grid and at least one of the command and address ball grids are located in the same row; wherein the termination ball grid is used to transmit on-chip termination commands, and the reset ball grid is used to transmit reset signals.

[0021] In a second aspect, an embodiment of the present disclosure provides a semiconductor structure, comprising a packaging structure and a chip as in the first aspect; wherein the data bit width of the chip is 4 bits, 8 bits or 16 bits.

[0022] The embodiments of the present disclosure provide a packaging structure and a semiconductor structure that are compatible with chips of different specifications, have strong adaptability, and have a small overall area. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] FIG1 is a schematic diagram of a packaging structure provided by an embodiment of the present disclosure;

[0024] FIG2 is a schematic diagram of a ball grid array provided by an embodiment of the present disclosure;

[0025] FIG3 is a first detailed structural diagram of a ball grid array provided by an embodiment of the present disclosure;

[0026] FIG4 is a schematic diagram of another ball grid array provided by an embodiment of the present disclosure;

[0027] FIG5 is a second detailed structural diagram of a ball grid array provided by an embodiment of the present disclosure;

[0028] FIG6 is a schematic structural diagram of a semiconductor structure provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0029] The technical solutions of the present disclosure will be further described in detail below with reference to the accompanying drawings and examples. Although the accompanying drawings illustrate exemplary implementations of the present disclosure, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0030] The following paragraphs describe the present disclosure in more detail by way of example with reference to the accompanying drawings. The advantages and features of the present disclosure will become more apparent from the following description and claims. It should be noted that the drawings are highly simplified and not to exact scale, and are intended solely to facilitate and clearly illustrate the embodiments of the present disclosure.

[0031] It will be understood that the meanings of “on,” “over,” and “over” throughout this disclosure should be interpreted in the broadest manner, such that “on” not only means being “on” something with no intervening features or layers (i.e., directly on something), but also includes being “on” something with intervening features or layers.

[0032] In the embodiments of the present disclosure, the terms "first," "second," "third," etc. are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0033] In the embodiments of the present disclosure, the term "layer" refers to a portion of a material including an area having a thickness. A layer may extend over the entirety of a lower or upper structure, or may have an extent that is smaller than the extent of the lower or upper structure. In addition, a layer may be an area of ​​a homogeneous or inhomogeneous continuous structure having a thickness that is smaller than the thickness of the continuous structure. For example, a layer may be located between the top and bottom surfaces of a continuous structure, or a layer may be between any horizontal faces at the top and bottom surfaces of a continuous structure. A layer may extend horizontally, vertically, and / or along an inclined surface. A layer may include multiple sublayers.

[0034] It should be noted that the technical solutions described in the embodiments of the present disclosure can be arbitrarily combined without conflict.

[0035] Static Random-Access Memory (SRAM) semiconductor chip;

[0036] Synchronous Dynamic Random Access Memory (SDRAM) semiconductor chip;

[0037] Double Data Rate SDRAM (DDR);

[0038] Fifth-generation DDR (DDR5);

[0039] Sixth-generation DDR (DDR6);

[0040] Two-wire bidirectional serial bus protocol (I3C);

[0041] Printed Circuit Board (PCB).

[0042] Before introducing the embodiments of the present disclosure, the three directions that may be used to describe the three-dimensional structure of the plane involved in the following embodiments are defined. Taking the Cartesian coordinate system as an example, the three directions may include the X-axis, Y-axis and Z-axis directions (not involved in the embodiments of the present disclosure). The packaging structure may include a top surface on the front side and a bottom surface on the back side opposite to the front side; ignoring the flatness of the top surface and the bottom surface, the direction that intersects (for example, is perpendicular to) the top surface and the bottom surface of the packaging structure is defined as the third direction. In the directions of the top surface and the bottom surface of the packaging structure (that is, the plane where the packaging structure is located), two directions that intersect with each other are defined. For example, the column extension direction of the ball grid array can be defined as the first direction, and the row extension direction of the ball grid array can be defined as the second direction. The plane direction of the packaging structure can be determined based on the first direction and the second direction. In the embodiments of the present disclosure, the first direction and the second direction may be perpendicular to each other. In other embodiments, the first direction and the second direction may not be perpendicular.

[0043] In addition, the term "linearly adjacent" as used herein in conjunction with a ball grid array means and includes balls located directly above, below, to the left, and to the right of a given ball grid when the ball grid array is parallel to the plane of the drawing; the term "diagonally arranged" as used herein in conjunction with a ball grid array means and includes balls located to the upper right, lower right, upper left, and lower left of a given ball grid when the ball grid array is parallel to the plane of the drawing; the term "surrounding the ball grid" as used herein in conjunction with a ball grid array means and includes balls located directly above, below, to the left, right, upper right, lower right, upper left, and lower left of a given ball grid when the ball grid array is parallel to the plane of the drawing. The term "ball grid A is adjacent to ball grid B along a first direction" as used herein in conjunction with a ball grid array only includes balls A and B being adjacent when the ball grid array is parallel to the plane of the drawing, and ball grid A being located directly above or below ball grid B. The term "ball grid A is adjacent to ball grid B along the second direction" used herein in conjunction with the ball grid array only includes that ball grid A and ball grid B are adjacent when the ball grid array is parallel to the plane of the drawing, and ball grid A is located to the left or right of ball grid B.

[0044] In particular, the illustrations presented in this disclosure are not meant to be actual views of any particular microelectronic device package, ball grid array, or components thereof, but are merely idealized representations used to describe illustrative embodiments.

[0045] Since the subsequent description of the embodiments of the present disclosure involves DDR6 chips with 4-bit data width (X4), 8-bit data width (X8), and 16-bit data width (X16), Table 1 is provided to illustrate the relevant signals and their number involved in the packaging of the above chips.

[0046] Table 1

[0047] In particular, for X8 chips, terminal data strobe signals tdqs_c and tdqs_t are required. X4 and X16 chips do not require them, but mask flip control signals dmu_n and dml_n are required. In particular, for the same design specifications, the terminal data strobe signal can be multiplexed with the mask flip control signal on the same ball grid, that is, tdqs_c and dmu_n are multiplexed on the same ball grid, and tdqs_t and dml_n are multiplexed on the same ball grid.

[0048] The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0049] In one embodiment of the present disclosure, referring to Figure 1, a schematic diagram of a package structure 10 provided by the present disclosure is shown. As shown in Figure 1, the package structure 10 includes a package substrate 11, and a ball grid array 20 is provided on the surface of the package substrate 11.

[0050] It should be noted that the packaging structure 10 provided in the embodiment of the present disclosure can be applied to at least DDR5 or DDR6, and can simultaneously support DDR5 and DDR6 chips with data bit widths of 4 bits, 8 bits, and 16 bits. Specifically, the DDR chip includes a device substrate, which includes a semiconductor material and a bonding pad coupled to the active surface of the device substrate. The packaging substrate is fixed to the device substrate, and the aforementioned packaging substrate 11 is configured to route signals to / from the bonding pad, while the ball grid array 20 is supported on the packaging substrate 11 and electrically connected thereto. In particular, the distribution position of the ball grid array 20 on the packaging substrate 11 is only for illustration and does not constitute a limitation. In fact, the surface of the packaging substrate 11 is not necessarily a completely regular rectangle, and the ball grid array 20 can be distributed in any area of ​​the surface of the packaging substrate 11.

[0051] In the following description, the first direction is the column extending direction of the BGA 20 , and the second direction is the row extending direction of the BGA 20 .

[0052] As shown in FIG1 , the ball grid array 20 includes a plurality of data balls (only a portion of which, namely, 201 to 204, is shown in FIG1 ), which are used to transmit data signals (e.g., dqu[7:0], dql[7:0] in Table 1); in the second direction, at most two data balls are allowed to be arranged continuously.

[0053] Thus, in the disclosed embodiment, two data balls can be arranged adjacently, reducing the number of rows in the data ball array and, consequently, the package size. This also helps reduce the length of the wiring from the data pins to the chip's I / O terminals, benefiting high-speed storage products. However, arranging three or more data balls adjacently can generate significant crosstalk between signals, causing signal distortion.

[0054] In some embodiments, only one row of any two adjacent rows of the ball grid array 20 is distributed with data balls, that is, no other data balls are distributed in the row before or after each data ball. For example, see FIG2 , which shows a schematic diagram of another ball grid array 20 provided by an embodiment of the present disclosure. As shown in FIG2 , the data balls are distributed in rows 1, 3, 5, and 7. In this way, there is at most one other data ball around each data ball (directly above, directly below, directly to the left, directly to the right, upper left, lower left, upper right, and lower right), thereby improving the signal distortion problem caused by magnetic field superposition.

[0055] In some embodiments, referring to FIG. 2 , the data sphere grid can be divided into two parts according to the characteristics of the distribution position:

[0056] (1) For the data ball grids of the first part (207 and 208 in FIG2 ), they are located in the edge rows and non-edge columns of the ball grid array 20, and any data ball grid of the first part is not linearly adjacent to other data ball grids, that is, data ball grid 205 (or 206) is not linearly adjacent to any data ball grid (whether in the first part or the second part).

[0057] (2) The data balls in the second part (201 to 206 in FIG. 2 ) are located in non-edge rows and non-edge columns of the ball grid array 20, and every two data balls in the second part are adjacent to each other along the second direction, that is, data ball grid 201 is adjacent to data ball grid 202 along the second direction, data ball grid 203 is adjacent to data ball grid 204, and data ball grid 205 is adjacent to data ball grid 206.

[0058] It should be noted that FIG. 2 is only one possible distribution form of the data spheres, and adaptive adjustments can be made without violating the text description, for example, the data spheres 201 to 206 are shifted rightward in whole or in part by one position.

[0059] In this way, in the disclosed embodiment, the data ball grids are allowed to be distributed in the edge rows, which can reduce the number of ball grid rows of the ball grid array 20 as a whole and reduce the substrate area occupied; at the same time, the data ball grids are not allowed to be distributed in the edge columns, which can minimize the distribution width of the data ball grids along the second direction, shorten the transmission distance of the data signal, and better ensure the quality of the data signal.

[0060] In addition, as mentioned above, the ball grid array 20 is compatible with X4, X8 and X16 chips, which have different numbers of data signals. Therefore, some or all of the data balls can be selectively enabled according to actual application scenarios.

[0061] In some embodiments, as shown in FIG. 2 , the BGA 20 further includes a plurality of command and address balls ( 301 ˜ 307 and 311 ˜ 317 ), which are used to transmit command / address signals ca.

[0062] (1) For the command address ball grid (301) of the first part, it is located in an edge row and non-edge column of the ball grid array 20, and any command address ball grid of the first part is not linearly adjacent to other command address ball grids; as shown in Figure 2, the command address ball grid 301 (or 307) is not linearly adjacent to any command address ball grid (whether in the first part or the second part).

[0063] (2) For the command address balls (302 to 306) of the second part, they are located in non-edge rows and non-edge columns of the ball grid array 20, and at most two of the command address balls of the second part are allowed to be arranged continuously and adjacently in the second direction, and at most three of the command address balls of the second part are allowed to be arranged continuously and adjacently in the first direction; as shown in FIG2 , command address balls 302, command address balls 303, and command address balls 304 are arranged continuously along the first direction, and command address balls 303 and command address balls 305 are arranged continuously along the second direction.

[0064] It should be noted that Figure 2 is only one possible distribution form of the command address ball grid, and adaptive adjustments can be made without violating the text description, for example, the command address ball grids 305 to 307 are shifted downward in whole or in part by one position and the command address ball grid 301 is shifted to the left by one position, or the command address ball grid 305 and the command address ball grid 306 are shifted to the left by one position and the command address ball grid 307 is shifted to the right by one position.

[0065] Thus, in the disclosed embodiment, command address balls are permitted to be distributed in edge rows, reducing the area of ​​BGA 20 and the number of rows occupied by command address balls in BGA 20, further saving space while also better matching the timing between clock and address signals. Furthermore, command address balls are not permitted to be distributed in edge columns, thereby minimizing the extent of the command address balls along the second direction, shortening the transmission path of the command address signals, and improving signal transmission quality.

[0066] In some embodiments, the ball grid array 20 has a total of N rows, the data ball grid of the first part is located in the 1st row, the data ball grids of the second part are all located in rows 2 to A, the command address ball grids of the second part are all located in rows B to (N-1), and the command address ball grid of the first part is located in row N; A, B and N are all positive integers, A is less than B, and B is less than N-1.

[0067] Taking Figure 2 as an example, N = 13, A = 7, and B = 9, but these can be adjusted in actual application scenarios, for example, N can be adjusted to 14.

[0068] In a specific embodiment, B>A+1. As shown in FIG2 , the ball grid array 20 can be divided into a data transmission area, an isolation area, and a command transmission area along a first direction. The data balls are distributed in the data transmission area, and the command and address balls are distributed in the control transmission area. The data transmission area and the control transmission area are separated by the isolation area.

[0069] For the convenience of explanation, please refer to Figure 3, which is a schematic diagram of the arrangement of the ball grid array 20 for the specific application scenario of the DDR6 chip. For Figure 3 only, N=13, A=7, B=9, but this does not constitute a corresponding limitation. The ball grid array 20 has 13 rows × 11 columns, including 16 data balls DQ and 14 command address balls CA. It should be understood that Figure 3 is only a specific embodiment of the present disclosure. Without violating the following text description, the scale of the ball grid array 20, the number of different types of signals, and the detailed arrangement of each signal can be adaptively adjusted.

[0070] Referring to FIG3 , the balls in row C are all ground balls VSS or power balls. The ground ball VSS is used to transmit the ground signal vss, and the power balls (VDD, VDDQ) are used to transmit the power signals (vdd, vddq). When A is an odd number, C = (A+1) / 2, i.e., C = 4 in FIG4 .

[0071] In one embodiment, the data balls DQs located in rows 1 to (C-1) are used to transmit high-order data signals (e.g., dqu[7:0] in Table 1), and the data balls DQs located in rows (C+1) to A are used to transmit low-order data signals (e.g., dql[7:0] in Table 1). Specifically, for an X4 chip, only the four data balls DQs located in rows (C+1) to A are enabled; for an X4 chip, only the eight data balls DQs located in rows (C+1) to A are enabled; and for an X16 chip, only all the data balls DQs are enabled.

[0072] In another possibility, the data balls DQs located in rows 1 to (C-1) are used to transmit low-order data signals (e.g., dql[7:0] in Table 1), and the data balls DQs located in rows (C+1) to A are used to transmit high-order data signals (e.g., dqu[7:0] in Table 1). Specifically, for an X4 chip, only the four data balls DQs located in rows 1 to (C-1) are enabled; for an X4 chip, only the eight data balls DQs located in rows 1 to (C-1) are enabled; and for an X16 chip, only all the data balls DQs are enabled.

[0073] In this way, the low-order data ball grid and the high-order data ball grid are independent of each other, and the two areas are isolated by the C-th row of ball grids. The arrangement is reasonable and can improve signal distortion caused by magnetic field superposition.

[0074] As shown in FIG3 , the ball grid array 20 is divided into a first array, a center array, and a second array. The first array, the center array, and the second array are arranged along the second direction. The number of balls in the first array is the same as the number of balls in the second array, and all the balls in the center array are removed. In other words, after the ball grid array 20 is completed, all the balls in the center column need to be removed to facilitate PCB or substrate routing. The center column in FIG3 contains three columns, but this does not constitute a relevant limitation. The number of columns contained in the center column can be more or less. In particular, in FIG3 , the length of each column in the center column (i.e., the 5th / 6th / 7th column) along the second direction is shorter, but this is only to save page space. In fact, the length of each column in the ball grid array 20 along the second direction is basically the same.

[0075] In some embodiments, see Figure 3 , the data balls DQs in the first array are symmetrically arranged with respect to the center array with respect to the data balls DQs in the second array. For example, even-numbered data balls DQs are distributed in columns 2, 3, and 4, while odd-numbered data balls DQs are distributed in columns 8, 9, and 10. Alternatively, even-numbered data balls DQs are distributed in columns 8, 9, and 10, while odd-numbered data balls DQs are distributed in columns 2, 3, and 4. In this way, the data balls DQs in the first and second arrays are mirror images of each other, allowing for shared vias when designing for both front- and back-end mounting on a PCB.

[0076] It should be noted that the ball grid array 20 of FIG3 includes a total of 16 data balls DQ, which can support chips with 4-bit, 8-bit, and 16-bit data widths. Simply put, if the package structure 10 is applied to a chip with a 4-bit data width (X4), only 4 of the data balls are enabled, and the remaining 12 data balls are disabled. If the package structure 10 is applied to a chip with an 8-bit data width (X8), only 8 of the data balls are enabled, and the remaining 8 data balls are disabled. If the package structure 10 is applied to a chip with a 16-bit data width (X16), all the data balls are enabled.

[0077] In some embodiments, the command address balls CA in the first array are symmetrical with the command address balls CA in the second array about the center array. For example, the even-numbered command address balls CA are distributed in columns 2, 3, and 4, and the odd-numbered command address balls CA are distributed in columns 8, 9, and 10. Alternatively, the even-numbered command address balls CA are distributed in columns 8, 9, and 10, and the odd-numbered command address balls CA are distributed in columns 2, 3, and 4. In this way, the command address balls CA in the first and second arrays are mirror images of each other, allowing for shared vias when designing for front- and back-side mounting on a PCB.

[0078] In some embodiments, the ball grid array 20 further includes a plurality of data strobe balls (e.g., 2 DQS_T and 2 DQS_C in FIG. 3 ), which are used to transmit data strobe signals (e.g., dqsl_t, dqsl_c, dqsu_t, and dqsu_c in Table 1); in the second direction, each data strobe ball is adjacent to the center array, and the data strobe ball is allowed to be adjacent to the data ball DQ along the first direction, and the data strobe ball is allowed to be adjacent to the data ball DQ along the second direction, thereby reducing the area of ​​the ball grid array 20.

[0079] It should be noted that, although four data strobe balls are shown in FIG3 , the number of data strobe balls may be more or less in actual scenarios.

[0080] It should also be noted that one data strobe ball DQS_T and one data strobe ball DQS_C are adjacent along the first direction and are used to transmit a pair of differential signals (dqsl_t and dqsl_c are a pair of differential signals, and dqsu_t and dqsu_c are another pair of differential signals). They can be called a group of data strobe balls. Different groups of data strobe balls are not adjacent in the first direction to reduce signal interference. At the same time, in Figure 3, all data strobe balls are in the same column, but this is not mandatory. Different groups of data strobe balls can be located in different columns (see Figure 5), but they must be adjacent to the center column to facilitate PCB or substrate wiring.

[0081] In some embodiments, the ball grid array 20 further includes a plurality of mask control balls DM, which are used to transmit mask flip control signals (eg, dmu, dml in Table 1) or terminal data selection signals (eg, tdqs_t, tdqs_c in Table 1).

[0082] It should be noted that while Figure 3 shows two mask control balls, the number of mask control balls can be greater or less. However, the number of mask control balls DM is less than or equal to the number of data strobe balls, and one mask control ball DM and one data strobe ball are symmetrical about the center array. Similarly, while the mask control balls DM are located on the same side of the center column in Figure 3, they can also be located on different sides of the center array (see Figure 5).

[0083] As shown in FIG3 , each mask control sphere DM is “adjacent to and only adjacent to” one data sphere DQ in a straight line. Taking FIG3 as an example, one mask control sphere DM is adjacent to one data sphere DQ along a first direction, and another mask control sphere DM is adjacent to one data sphere DQ along a second direction, thereby minimizing magnetic field overlap.

[0084] In some embodiments, as shown in Figure 3, the ball grid array 20 also includes a local data enable ball grid LBDQS, a local data ball grid LBDQ and a calibration indication ball grid ZQ; the local data enable ball grid LBDQS is located in the edge column, and the local data enable ball grid LBDQS is adjacent to the data ball grid DQ along the second direction; the local data ball grid LBDQ and the local data enable ball grid LBDQ are symmetrical about the central array, that is, the local data ball grid LBDQ is also located in the edge column, and is only adjacent to one data ball grid local data ball grid LBDQ along the second direction.

[0085] Here, the local data ball LBDQ is used to transmit the local data signal lbdq, and the local data strobe ball LBDQS is used to transmit the local data strobe signal lbdqs.

[0086] As shown in FIG3 , in the second direction, the local data ball grid LBDQ, one data ball grid DQ, and another data ball grid DQ are continuously and adjacently arranged; one data ball grid DQ, another data ball grid DQ, and the local data strobe ball grid LBDQS are continuously and adjacently arranged.

[0087] In some embodiments, the calibration indicator ball grid ZQ is adjacent to the local data enable ball grid LBDQS along a first direction; in Figure 3, the calibration indicator ball grid ZQ is located above the local data enable ball grid LBDQS, but the calibration indicator ball grid ZQ can also be located below the local data enable ball grid LBDQS, and the calibration indicator ball grid ZQ is used to transmit the impedance calibration signal zq.

[0088] In some other embodiments, the calibration indicator ball ZQ is adjacent to the local data ball LBDQ along the first direction, that is, the calibration indicator ball ZQ and the power ball VPP in the first column in FIG. 3 are swapped.

[0089] As shown in Figure 3, for balls 1 through A (A=7 in Figure 3), all other balls not mentioned are power balls or ground balls VSS. Power balls are labeled VDD, VDDQ, and VPP, depending on their function and / or voltage value. Thus, each data ball DQ is surrounded by at least one ground ball VSS and one power ball. Here, "surrounding a data ball DQ" includes the balls immediately above, below, to the left, to the right, and to the upper left, lower left, upper right, and lower right of the data ball DQ. This not only provides isolation but also facilitates data signal generation.

[0090] In some embodiments, referring to FIG4 , the ball grid array 20 further includes a plurality of chip select ball grids CS adjacent to each other along a first direction, the chip select ball grid CS being used to transmit a chip select signal (e.g., cs_n[1:0] in Table 1); a portion of the chip select ball grid CS is adjacent to a data ball grid DQ along the first direction; another portion of the chip select ball grid CS is adjacent to a command address ball grid CA along the second direction, and is adjacent to another command address ball grid CA along the first direction.

[0091] It should be noted that while Figure 4 shows two chip select balls CS, actual implementations may require more or fewer. Thus, the chip select ball CS is allowed to be adjacent to either the data ball DQ or the command address ball CA, and can be adjacent to at most two command address balls CA. This allows for more centralized signal arrangement, further reducing the size of the ball grid array 20 and occupying less substrate area.

[0092] In some embodiments, the ball grid array 20 further includes a plurality of clock balls CK adjacent to each other along a first direction, and the clock ball CK is used to transmit a clock signal (e.g., ck_c or ck_n in Table 1); a portion of the clock ball CK is not linearly adjacent to any command address ball CA, and is not linearly adjacent to any data ball DQ; another portion of the clock ball CK is adjacent to one command address ball CA along a second direction, and is adjacent to another command address ball CA along the first direction.

[0093] It should be noted that Figure 4 shows two clock balls (CK), but in actual scenarios, there can be more or fewer. This closer proximity of the clock balls to the command and address balls allows for better matching of clock and address delays, shortening the overall transmission path, improving signal integrity, and boosting overall product competitiveness.

[0094] In some embodiments, multiple chip select balls CS are located in a first array and adjacent to edge columns of the ball grid array 20; multiple clock balls CK are located in a second array and adjacent to the center array, and one chip select ball CS and one clock ball CK are located in the same row.

[0095] For example, referring to Figure 3, the two chip select balls can be represented as CS_N[0] and CS_N[1], respectively, for transmitting cs_n[0] and cs_n[1], with at most one of cs_n[0] and cs_n[1] being valid. The two clock balls can be represented as CK_C and CK_T, respectively, for transmitting a pair of differential signals, ck_c and ck_t. Furthermore, the chip select ball CS_N[0] and the clock ball CK_T are located in the same row, and the chip select ball CS_N[1] and the clock ball CK_C are located in the same row.

[0096] It should be noted that in Figure 3, the clock ball CK is located in the 8th column, but the clock ball CK can also be located in the 4th column (see Figure 5), as long as the clock ball CK is adjacent to the center array, which makes it easier to implement PCB or substrate routing.

[0097] In some embodiments, referring to FIG3 , the ball grid array 20 further includes a check error indication ball grid ALERT_N and a mirror mode ball grid MIR, and the check error indication ball grid ALERT_N and the mirror mode ball grid MIR are both located in the edge columns of the ball grid array 20; the mirror mode ball grid MIR is adjacent to a chip select ball grid CS_N[0] along the second direction; the error indication ball grid ALERT_N is adjacent to another chip select ball grid CS_N[1] along the second direction; wherein, the mirror mode ball grid MIR is at least used to transmit the mirror mode enable signal mir, and the error indication ball grid ALERT_N is used to transmit the check error indication signal alert_n.

[0098] In other embodiments, the positions of the check error indication ball raster ALERT_N and the error indication ball raster ALERT_N may be interchanged.

[0099] In some embodiments, the ball grid array 20 also includes multiple reserved ball grids RFU, one test mode ball grid TEN and two protocol ball grids (i.e., MSCL and MSDA), the test mode ball grid TEN is used to transmit the test mode enable signal ten, and the two protocol ball grids are respectively used to transmit the protocol control signals mscl and msda (related to the I3C protocol); the multiple reserved ball grids RFU and the multiple chip select ball grids CS are symmetrical about the central array; the second protocol ball grid and the mirror mode ball grid MIR are symmetrical about the central array, the test mode ball grid TEN and the error indication ball grid ALERT_N are symmetrical about the central array, and the first protocol ball grid MSDA, the second protocol ball grid MSCL and the test mode ball grid TEN are arranged continuously along the first direction.

[0100] In other embodiments, the positions of multiple reserved balls RFU, one test mode ball TEN and two protocol balls (ie, MSCL and MSDA) can be flexibly adjusted, mainly depending on the arrangement of the internal circuits, as long as the signal input / output is convenient.

[0101] In some embodiments, the ball grid array 20 also includes a termination ball grid CA_ODT and a reset ball grid RESET, and the termination ball grid CA_ODT and the reset ball grid RESET are located in the edge columns of the ball grid array 20; the termination ball grid CA_ODT and the reset ball grid RESET are symmetrical about the central array; the linearly adjacent balls of the termination ball grid CA_ODT are all grounded ball grids VSS and / or power ball grids, and the termination ball grid CA_ODT is in the same row as at least one command address ball grid CA; the linearly adjacent balls of the reset ball grid RESET are all grounded ball grids VSS and / or power ball grids, and the reset ball grid RESET is in the same row as at least one command address ball grid CA; wherein, the termination ball grid CA_ODT is used to transmit on-chip termination commands, and the reset ball grid RESET is used to transmit reset signals.

[0102] Similarly, in other embodiments, the positions of the termination ball grid CA_ODT and the reset ball grid RESET can be flexibly adjusted.

[0103] In addition to the above-mentioned signals, the other balls in the BGA 20 are power balls or ground balls VSS.

[0104] In other embodiments, the positions of the following balls can be flexibly swapped: mask control ball DM, test mode enable ball TEN, reserved ball RFU, mirror mode enable ball MIR (used to control whether the chip enters mirror mode), error indication ball ALERT_N (used to indicate errors detected during cyclic redundancy check operations), protocol balls MSDA / MSCL, termination ball CA_ODT, reset ball RESET, and calibration ball ZQ. See Figure 5 , which provides another possible ball arrangement.

[0105] In some embodiments, referring to FIG. 3 , the ball grid array 30 has 13 rows, the first array has 4 columns, and the second array has 4 columns, providing a total of 104 ball grids, capable of supporting signal packaging of X4, X8, and X16 chips.

[0106] In some embodiments, the distance between every two ball grid center points along the first direction is 800 microns, and the distance between every two ball grid center points along the second direction is also 800 microns. The overall size of the ball grid array is 10 mm × 11 mm, which is relatively small overall. More packaging structures can be produced with the same substrate.

[0107] As can be seen from the above, the packaging structure provided by the embodiments of the present disclosure allows X4, X8, and X16 chips to share a common substrate. Furthermore, the overall size of the ball grid array is small, allowing more substrate units to be cut from a single substrate, thereby forming more packaging structures, thereby reducing costs. This also reduces the design cycle, eliminating the need to design separate substrates for chips with different bit widths. Furthermore, testing has shown that the ball grid array 20 provided by the present disclosure has a longer solder joint life, less plastic strain, and a longer fatigue life.

[0108] In another embodiment of the present disclosure, referring to FIG6 , a schematic diagram of the composition of a semiconductor structure 60 provided in an embodiment of the present disclosure is shown. As shown in FIG6 , the semiconductor structure 60 includes at least the aforementioned package structure 10 and a chip 70, and the data bit width of the chip 70 is 4 bits, 8 bits, or 16 bits. The chip 70 is, for example, a DDR6.

[0109] Here, the chip 70 may be placed on the package structure 10 , and the chip 70 may be electrically connected to the ball array on the package structure 10 , so that various signals related to the chip are input / output through the package structure 10 .

[0110] In summary, the packaging structure 10 provided in the embodiment of the present disclosure is compatible with chips of X4, X8, and X16 specifications, so that chips of different bit widths can share the same substrate; and the substrate size is smaller, thereby controlling costs, and the overall performance is better.

[0111] The above description is merely a specific embodiment of the present disclosure, but the scope of protection of the present disclosure is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this disclosure should be included in the scope of protection of the present disclosure. Therefore, the scope of protection of the present disclosure should be based on the scope of protection of the claims.

Claims

1. A packaging structure (10), comprising: A packaging substrate (11), wherein a ball grid array (20) is provided on the surface of the packaging substrate (11), wherein the ball grid array comprises a plurality of data ball grids (201-204), and the data ball grids are used to transmit data signals; In the second direction, at most two of the data ball grids are allowed to be arranged continuously; the second direction refers to the row extension direction of the ball grid array.

2. The packaging structure according to claim 1, wherein: The data ball grids of the first part are located in edge rows and non-edge columns of the ball grid array, and any of the data ball grids of the first part is not linearly adjacent to other data ball grids; The data ball grids of the second part are located in non-edge rows and non-edge columns of the ball grid array, and every two data ball grids of the second part are adjacent along the second direction; Among any two adjacent rows of the ball grid array, only one row has the data ball grid distributed therein.

3. The packaging structure according to claim 1 or 2, wherein: The ball grid array further comprises a plurality of command address ball grids, wherein the command address ball grids are used to transmit command / address signals; The command address ball grids of the first part are located in edge rows and non-edge columns of the ball grid array, and any command address ball grid of the first part is not linearly adjacent to other command address ball grids; For the command address ball grid of the second part, it is located in the non-edge row and non-edge column of the ball grid array, and at most 2 command address ball grids of the second part are allowed to be arranged continuously in the second direction, and at most 3 command address ball grids of the second part are allowed to be arranged continuously in the first direction, and the first direction refers to the column extension direction of the ball grid array.

4. The packaging structure according to claim 3, wherein: The ball grid array has a total of N rows, the data ball grid of the first part is located in the 1st row, the data ball grids of the second part are all located in the 2nd to A rows, the command address ball grids of the second part are all located in the B to (N-1)th rows, and the command address ball grid of the first part is located in the Nth row; A, B and N are all positive integers, A is less than B, and B is less than N-1.

5. The packaging structure according to claim 4, wherein: The data ball grids located in the 1st to (C-1)th rows are used to transmit high-order data signals, and the data ball grids located in the (C+1)th to Ath rows are used to transmit low-order data signals; or, the data ball grids located in the 1st to (C-1)th rows are used to transmit low-order data signals, and the data ball grids located in the (C+1)th to Ath rows are used to transmit high-order data signals; The ball grids located in the Cth row are all ground ball grids or power ball grids, and the ground ball grids are used to transmit ground signals, and the power ball grids are used to transmit power signals; wherein, C=(A+1) / 2.

6. The packaging structure according to any one of claims 1 to 5, wherein: The ball grid array is divided into a first array, a central array and a second array. The first array, the central array and the second array are arranged in sequence along a second direction. The number of ball grids in the first array is the same as the number of ball grids in the second array, and all ball grids in the central array are removed.

7. The packaging structure according to claim 6, wherein: The data ball grid located in the first array and the data ball grid located in the second array are symmetrical about the central array, and the command address ball grid located in the first array and the command address ball grid located in the second array are symmetrical about the central array.

8. The packaging structure according to claim 6 or 7, wherein: The ball grid array further comprises a plurality of data strobe ball grids, wherein the data strobe ball grids are used to transmit data strobe signals; In the second direction, each of the data strobe balls is adjacent to the central array, and the data strobe balls are allowed to be adjacent to the data balls along the first direction, and the data strobe balls are allowed to be adjacent to the data balls along the second direction.

9. The packaging structure according to claim 8, wherein: The ball grid array further comprises a plurality of mask control ball grids, wherein the mask control ball grids are used to transmit mask flip control signals or terminal data strobe signals; The mask control ball grid and the data strobe ball grid are symmetrical about the central array.

10. The packaging structure according to any one of claims 6 to 9, wherein: The ball grid array also includes a local data strobe ball grid, a local data ball grid and a calibration indicator ball grid; The local data strobe ball grid is located in an edge column, and the local data strobe ball grid is adjacent to the data ball grid along a second direction; The local data strobe ball grid is symmetrical with the local data ball grid about the central array; The calibration indication ball grid is adjacent to the local data strobe ball grid along a first direction; The local data ball grid is used to transmit a local data signal, the local data selection ball grid is used to transmit a local data selection signal, and the calibration indication ball grid is used to transmit an impedance calibration signal.

11. The packaging structure according to any one of claims 3 to 10, wherein: The ball grid array further comprises a plurality of chip select ball grids adjacent to each other along a first direction, wherein the chip select ball grids are used to transmit chip select signals; A portion of the chip select ball grid is adjacent to one of the data ball grids along a first direction; Another portion of the chip select ball grid is adjacent to one of the command address ball grids along the second direction, and is adjacent to another of the command address ball grids along the first direction.

12. The packaging structure according to any one of claims 1 to 11, wherein: The ball grid array further includes a plurality of clock ball grids adjacent to each other along a first direction, and the clock ball grids are used to transmit clock signals.

13. The packaging structure according to claim 12, wherein: A portion of the clock ball grid is not directly adjacent to any of the command address ball grids, and is not directly adjacent to any of the data ball grids; Another portion of the clock ball grid is adjacent to one of the command address ball grids along the second direction, and is adjacent to another of the command address ball grids along the first direction.

14. The packaging structure according to claim 12 or 13, wherein: A plurality of the chip select ball grids are located in the first array and are adjacent to edge columns of the ball grid array; A plurality of the clock ball grids are located in the second array and adjacent to the central array; One of the chip select ball grids and one of the clock ball grids are located in the same row.

15. The packaging structure according to any one of claims 1 to 14, wherein: The ball grid array further includes a check error indication ball grid and a mirror mode ball grid, and the check error indication ball grid and the mirror mode ball grid are both located at the edge row of the ball grid array.

16. The packaging structure according to claim 15, wherein: The mirror mode ball grid is adjacent to one of the chip selection ball grids along a second direction; The error indication ball grid is adjacent to another chip selection ball grid along a second direction; The mirror mode ball grid is used to transmit at least a mirror mode enable signal, and the error indication ball grid is used to transmit a check signal. Error indication signal.

17. The packaging structure according to any one of claims 11 to 16, wherein: The ball grid array further includes a plurality of reserved ball grids, a test mode ball grid and two protocol ball grids, wherein the test mode ball grid is used to transmit a test mode enable signal, and the protocol ball grid is used to transmit a protocol control signal; The plurality of reserved ball grids and the plurality of chip selection ball grids are symmetrical about the central array; The second protocol ball grid and the mirror mode ball grid are symmetrical about the central array, the test mode ball grid and the error indication ball grid are symmetrical about the central array, and the first protocol ball grid, the second protocol ball grid and the test mode ball grid are arranged continuously along a first direction.

18. The packaging structure according to any one of claims 6 to 17, wherein: The ball grid array further comprises a termination ball grid and a reset ball grid, wherein the termination ball grid and the reset ball grid are located at an edge column of the ball grid array; The termination ball grid and the reset ball grid are symmetrical about the central array.

19. The package structure according to claim 18, wherein: The linearly adjacent ball grids of the termination ball grid are all ground ball grids and / or power ball grids, the termination ball grid is in the same row with at least one of the command address ball grids, and the reset ball grid is in the same row with at least one of the command address ball grids; The termination ball grid is used to transmit an on-chip termination command, and the reset ball grid is used to transmit a reset signal.

20. A semiconductor structure (60), wherein: It comprises a packaging structure (10) and a chip (70) as claimed in any one of claims 1 to 19; wherein the data bit width of the chip is 4 bits, 8 bits or 16 bits.

Citation Information

Patent Citations

  • Packaging structure and semiconductor structure

    CN119092478A

  • Ball grid array and associated device and system

    CN113871367A

  • Apparatus and system with ball grid array and related microelectronic devices and device packages

    CN114121873A

  • Ball Grid Array package being capable of mirrioringand memory module having the same

    KR1020060019064A

  • A method of manufacturing a semiconductor device, a seconductor device manufactured by the same method

    KR1020220040124A