Module assembly and method for stacked BGA packages

Abstract

Ball grid array packages that can be stacked to form highly dense components and the method for stacking ball grid arrays are disclosed. The ball grid array packages comprise flexible or rigid substrates. The ball grid array packages additionally comprise an arrangement for the substantial matching of impedance for the circuits connected to the semiconductor devices.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates to ball grid array packages that can be stacked to form highly dense components and the method for stacking ball grid arrays. The ball grid array packages may be stacked on flexible or rigid substrates.

2. State of the Art

Chip-on-board technology generally consists of three types of techniques for attaching a semiconductor device to a printed circuit board, such as flip-chip attachment, wire bonding, and tape automated bonding techniques.

Flip-chip attachment consists of attaching a semiconductor device, generally having a ball grid array (BGA), a slightly larger than integrated circuit carrier (SLICC), or a pin grid array (PGA) to a printed circuit board. With the BGA or SLICC, the solder ball arrangement on the semiconductor device must be a mirror-image of the connecting bond pads on the printed circuit board such that precise connections are made. The semiconductor device is bonded to the printed circuit board by refluxing the solder balls. With the PGA, the pin arrangement of the semiconductor device must be a mirror-image of the pin recesses on the printed circuit board. After insertion, the semiconductor device is generally bonded by soldering the pins into place. An underfill encapsulant is generally disposed between the semiconductor device and the printed circuit board to prevent contamination. A variation of the pin-in-recess PGA is a J-lead PGA, wherein the loops of the J-leads are soldered to pads on the surface of the circuit board. However, the lead and pad locations must coincide, as with the other types of flip-chip techniques.

Wire bonding and tape automated bonding (TAB) attachment generally begin with attaching a semiconductor device to the surface of a printed circuit board with an appropriate adhesive. In wire bonding, a plurality of bond wires is attached, one at a time, from each bond pad of the semiconductor device to a corresponding lead on the printed circuit board. The bond wires are generally attached through one of three industry-standard wire bonding techniques, such as ultrasonic bonding—using a combination of pressure and ultrasonic vibration bursts to form a metallurgical cold weld, thermocompression bonding—using a combination of pressure and elevated temperature to form a weld, and thermosonic bonding—using a combination of pressure, elevated temperature, and ultrasonic vibration bursts. The semiconductor device may be oriented having either the active surface up or the active surface down (with the bond pads thereon either up or down with respect to the printed circuit board) for wire bonding, although active surface up is the most common. With TAB, metal tape leads are attached between the bond pads on the semiconductor device and the leads on the printed circuit board. An encapsulant is generally used to cover the bond wires and metal tape leads to prevent contamination.

Although such methods are effective for bonding semiconductor devices to printed circuit boards, the terminal arrangements of the devices and the connection arrangements of the boards must be designed to accommodate one another. Thus, it may be impossible to electrically connect a particular semiconductor device to a printed circuit board for which the semiconductor device terminal arrangements were not designed to match the board's connection arrangement. With either wire bond or TAB attachment, the semiconductor device bond pad arrangement may not correspond to the lead ends on the circuit board, making attachment difficult due to the need for overlong wires and the potential for inter-wire contact and shorting. With flip-chip attachment, if the printed circuit board connection arrangement is not a mirror-image of the solder ball or pin arrangement of the semiconductor device, electrically connecting the flip-chip to the printed circuit board is impossible.

Ball grid array (BGA) semiconductor device packages are well known in the art. A BGA package typically comprises a substrate, such as a printed circuit board, with a semiconductor device, such as a dynamic random access memory device, mounted on the top side of the substrate. The semiconductor device has a plurality of bond pads on the active surface thereof electrically connected to a series of metal traces on the top surface or top side of the printed circuit board. The connection between the bond pads and the metal traces is provided by wire bonds electrically and mechanically connecting the semiconductor device and the printed circuit board. The series of metal traces on the printed circuit board is connected, in turn, to a second series of metal traces on the bottom surface or bottom side of the printed circuit board using a series of vias extending therethrough. The second series of metal traces each terminate with a connection contact pad where a conductive element is attached. The conductive elements can be solder balls or conductive filled epoxy. The conductive elements are arranged in an array pattern and the semiconductor device and wire bonds are encapsulated with a molding compound.

As semiconductor device and grid array densities increase, the desire in packaging semiconductor devices has been to reduce the overall height or profile of the semiconductor package. The use of BGAs has allowed for this reduction of profile, as well as increased package density. Density has been increased by using lead frames, such as lead-over-chip type lead frames, in an effort to increase the semiconductor device density, as well as allow stacking of the semiconductor devices one on top another.

One example of a lead chip design in a BGA package is shown in U.S. Pat. No. 5,668,405. A semiconductor device is disclosed having a lead frame attached to the semiconductor device. Through holes are provided that allow for solder bumps to connect via the lead frame to the semiconductor device. Such a mounting arrangement requires several steps for attaching the semiconductor device to the lead frame, then providing sealing resin, and subsequently adding a base film and forming through holes in the base film. A cover resin is added before solder bumps are added in the through holes to connect to the lead frame. This particular structure lacks the ability to stack semiconductor devices one on top another.

U.S. Pat. No. 5,677,566, commonly assigned to the assignee of the present invention, illustrates a semiconductor device package that includes discrete conductive leads with electrical contact bond pads on a semiconductor device. The lead assembly is encapsulated with a typical encapsulating material and electrode bumps are formed through the encapsulating material to contact the conductive leads. The electrode bumps protrude from the encapsulating material for connection to an external circuit. The semiconductor device has the bond pads located in the center of the active surface of the device, thus allowing the conductive leads to be more readily protected once encapsulated in the encapsulating material. However, the assembly illustrated in the '566 patent lacks the ability to stack one semiconductor device on top another.

U.S. Pat. No. 5,625,221 illustrates a semiconductor device package assembly that has recessed edge portions that extend along at least one edge portion of the assembly in an attempt to form a stacked package of semiconductor devices. An upper surface lead is exposed therefrom and a top recess portion is disposed on a top surface of the assembly. A bottom recess portion is disposed on the bottom surface of the assembly such that when the assembly is used in fabricating a three-dimensional integrated circuit module, the recess edge portion accommodates leads belonging to an upper semiconductor assembly to provide electrical interconnection therebetween. However, the assembly requires long lead wires from the semiconductor chip to the outer edges. These lead wires add harmful inductance and unnecessary signal delay and can form a weak link in the electrical interconnection between the semiconductor device and the outer edges. Further, the assembly profile is a sum of the height of the semiconductor devices, the printed circuit boards to which they are bonded, the conductive elements, such as the solder balls, and the encapsulant that must cover the semiconductor devices and any wire bonds used to connect the devices to the printed circuit boards. Reducing such a package profile is difficult because of the geometries required in having the bond pads on the semiconductor device along the outer periphery with extended lead wires reaching from the semiconductor device to the outer edges.

U.S. Pat. Nos. 5,266,912 and 5,400,003 illustrate another stacked arrangement of semiconductor devices on a substrate interconnected by pins. However, the height of the stacked package is limited by the length of the pin connections between the individual multi-chip modules or printed circuit boards.

Another problem which arises in stacking semiconductor devices mounted on printed circuit boards is that it is difficult to provide a flat, smooth surface on which to mount the printed circuit board. Accordingly, flexible boards have been developed to allow both lighter-weight structures and greater adaptability at conforming to nonuniform surfaces. However, the use of such flexible circuit boards has resulted in other problems, such as the problem in joining several flexible boards while maintaining the proper interconnection between the respective boards. Further, in some applications, such as protecting semiconductor devices mounted on a bottom surface of a flexible substrate from touching the top of another flexible circuit board, the use of a rigid member or assembly is required to support the stacked flexible circuit boards. This sacrifices the flexibility that is present in the flexible circuit boards that allows their compliance with a non-planar surface.

U.S. Pat. No. 5,440,171 illustrates semiconductor devices mounted on flexible, stackable circuit boards to form semiconductor modules. A basic structure unit is illustrated comprising a flexible circuit board made from a polyamide film with circuit lines formed on both sides, typically using copper foil. A supporting frame is provided and bonded to the flexible circuit board with a heat-resistant resin, such as a polyamide resin. Electrical connections are possible between the flexible circuit board and the support frame. Conductive through holes are provided so that electrical continuity exists between a semiconductor device mounted upon the flexible circuit board and either at least one other semiconductor device mounted on another flexible circuit board stacked within the module assembly or an outside source upon which the entire basic structure unit is mounted. The semiconductor devices are electrically connected to electrodes on the support frame. Although the semiconductor device is mounted on a flexible circuit board that is stackable in an arrangement, the support frame attaching the stackable circuit boards one to another is made from a rigid material that does not allow for any bending. One type of frame material is ceramic, such as silicon nitride. Silicon nitride is used for its high thermal conductivity for heat radiation or dissipation when the semiconductor device has a high power consumption. Since the support frame is made from rigid and non-flexible material, the semiconductor device package assembly needs to be mounted on a substantially planar surface, thereby preventing the assembly from being molded on surfaces that are not uniformly planar or smooth.

Additionally, when stacking semiconductor devices using flexible or rigid substrates, as the operation speed of the semiconductor device increases it is desirable to match the impedance of the various circuits to which the semiconductor devices are connected, to try to keep the circuit response time the same for each circuit. Since in stacked arrangements the circuit length for each semiconductor device will vary, attention must be given to keeping the circuit impedance substantially the same.

Accordingly, what is needed is a ball grid array package that allows for the stacking of packages where printed circuit board substrates or flexible substrates may be used as desired and which allows for the matching of the impedance for the different circuits as required.

BRIEF SUMMARY OF THE INVENTION

The present invention comprises ball grid array packages that can be stacked to form highly dense components and the method for stacking ball grid arrays. The ball grid array packages comprise flexible or rigid substrates. Additionally, the present invention comprises an arrangement for the substantial matching of impedance for the circuits connected to the semiconductor devices.

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

FIG. 1 is a view of a semiconductor package for use in the present invention;

FIG. 2 is a cross-sectional view of a semiconductor package for use in the present invention;

FIG. 3 is a bottom view of a semiconductor package for use in the present invention;

FIG. 4 is a side view of a first embodiment of the present invention using semiconductor packages in a stacked package arrangement on each side of a substrate;

FIG. 5 is a side view of a second embodiment of the present invention using semiconductor packages stacked in a package arrangement on one side of a substrate wherein resistors and a bus bar arrangement are used to match the impedance of the circuits;

FIG. 6 is a side view of a third embodiment of the present invention using semiconductor packages stacked in a package arrangement on both sides of a substrate wherein resistors and bus bars are used to match the circuit impedance of the stacked packages;

FIG. 7 is a top view of a fourth embodiment of the semiconductor package of the present invention using a flexible substrate for the mounting of a semiconductor device thereon;

FIG. 8A is a side view of the fourth embodiment of the semiconductor package of the present invention of FIG. 7 shown in cross-section using a flexible substrate for the mounting of a semiconductor device thereon;

FIG. 8B is an enlarged view of a portion of the flexible substrate of the semiconductor package of FIG. 8A;

FIG. 9 is a side view of a fifth embodiment of the semiconductor package of the present invention using a plurality of stacked semiconductor packages using a flexible substrate for the mounting of a semiconductor device thereon; and

FIG. 10 is a side view of a sixth embodiment of the semiconductor package of the present invention using a plurality of stacked semiconductor packages using a flexible substrate for the mounting of a semiconductor device thereon.

DETAILED DESCRIPTION OF THE INVENTION

Referring to drawing FIG. 1, a wire bond style flip-chip assembly 100 is illustrated. An adapter board 18 is shown having a semiconductor device 12 located on the upper surface 20 thereof with the lower surface of the adapter board 18 having a plurality of solder balls 22 located thereon in rows for connection purposes.

Referring to drawing FIG. 2, the wire bond style flip-chip assembly 100 is illustrated in cross-section. The semiconductor device 12 has a plurality of bond pads 38 arranged in two rows on the active surface 14 thereof. The semiconductor device 12 is secured to the adapter board 18 by a suitable adhesive 40. The adapter board 18 is formed having at least one longitudinally extending aperture 42 therethrough and a plurality of connection pads 39 located on the bottom surface thereof. A plurality of circuits or circuit traces 23 of adapter board 18 connects connection pads 39 to a desired solder ball(s) 22. Wires 34 extend between the bond pads 38 of the semiconductor device 12 and the connection pads 39 of the adapter board 18, the wires 34 being bonded to the pads 38 and 39 through the use of a suitable wire bonder well known in the industry. After the connections using wires 34 have been made through aperture 42 in adapter board 18, a suitable encapsulant material 44 is applied to the aperture 42 to cover the wires 34, the bond pads 38 on the semiconductor device 12, and the connection pads 39 on the adapter board 18.

Referring to drawing FIG. 3, an adapter board 18 is illustrated from the bottom thereof. As illustrated, the encapsulant material 44 covers the aperture 42 in the adapter board 18. The solder balls 22 are illustrated in a plurality of rows. The semiconductor device 12 is shown in dashed lines as well as adhesive 40 connecting the semiconductor device 12 to the upper surface of the adapter board 18.

Referring to drawing FIG. 4, a plurality of wire bond style flip-chip assemblies 100 is illustrated connected to a substrate 50 in a stacked arrangement, each style flip-chip assembly 100 having two rows of solder balls 22 thereon. Each style flip-chip assembly 100 is connected to another style/flip-chip assembly 100 through circuits 52 in adapter boards 18 and connected to circuits 54 in substrate 50. The substrate 50 may be any suitable substrate, such as a printed circuit board, FR-4 board, or the like, which is structurally and electrically capable of connecting a plurality of flip-chip assemblies 100 thereto. Any desired number of flip-chip assemblies 100 may be connected to the substrate 50 on both or only one side thereof. The substrate 50 may have connection pads 56 thereon connected to circuits 54 for connection to other circuits or components. Although one stack of flip-chip assemblies 100 has been illustrated on each side of the substrate 50, any number may be used on each side or one side of the substrate 50.

Referring to drawing FIG. 5, a plurality of wire bond style flip-chip assemblies 100 is illustrated installed on a substrate 60 having suitable circuits therein in a first stack 62 and a second stack 64, the flip-chip assemblies 100 being interconnected using solder balls 22. The substrate 60 may be any suitable substrate, such as a printed circuit board, FR-4 board, or the like, capable of supporting the stacks 62 and 64 of flip-chip assemblies 100. In the arrangement, the stacked style flip-chip assemblies 100 are serially connected by means of the solder balls 22 using a jumper board 70 which includes bus lines therein and acts as a heat sink for the stacks 62 and 64. The jumper board 70 may be any suitable board, such as a printed circuit board, FR-4 board, or the like. Included on the substrate 60 is a plurality of resistors 66 which is used to balance the impedance of the circuits of the serially connected style flip-chip assemblies 100 in the stacks 62 and 64. The flip-chip assemblies 100 are serially connected to a resistor 66 through the circuits in the substrate 60 and jumper board 70 as illustrated by the arrows 72. In this manner, the impedance of the various circuits in the flip-chip assemblies 100 in the stacks 62 and 64 may be matched so that the response of the stacked style flip-chip assemblies 100 will not substantially vary.

Referring to drawing FIG. 6, a plurality of wire bond style flip-chip assemblies 100 is illustrated installed on both sides of substrate 60 having suitable circuits therein in first stacks 62 and a second stacks 64, the flip-chip assemblies 100 being interconnected using solder balls 22. The substrate 60 may be any suitable substrate, such as a printed circuit board, FR-4 board, or the like, capable of supporting the stacks 62 and 64 of flip-chip assemblies 100. In the arrangement, the stacked style/flip-chip assemblies 100 are serially connected by means of the solder balls 22 using jumper boards 70 which include bus lines therein and act as heat sinks for the stacks 62 and 64. The jumper boards 70 may be any suitable board, such as a printed circuit board, FR-4 board, or the like. Included on the substrate 60 is a plurality of resistors 66 which is used to balance the impedance of the circuits of the serially connected style flip-chip assemblies 100 in the stacks 62 and 64. The flip-chip assemblies 100 are serially connected to a resistor 66 through the circuits in the substrate 60 and jumper board 70 as illustrated by the arrows 72. In this manner, the impedance of the various circuits in the flip-chip assemblies 100 in the stacks 62 and 64 may be matched so that the response of the stacked style flip-chip assemblies 100 will not substantially vary.

Referring to drawing FIG. 7, a bottom view of an assembly 200 of a flexible substrate 202 is illustrated having a plurality of conductors 204 formed thereon connected by bonds 206 through apertures 212 in substrate 202 to bond pads 208 of semiconductor device 210. The flexible substrate 202 may be any suitable type material, such as polyamide tape, and have a plurality of desired conductors 204 formed thereon, such as copper-type conductors. The substrate 202 may include alignment apertures 214 therein, if desired. The substrate 202 further includes apertures 216 therein for the connection of the conductors 204 to other conductors 204 on adjacent stacked substrates. The apertures 216 have a size sufficient to allow a solder ball having a diameter of at least twice the thickness of the substrate 202 to be used therein. A gold ball-type bond 206 may be used to connect the conductors 204 to the bond pads 208 of the semiconductor device 210.

Referring to drawing FIG. 8A, the assembly 200 is shown in cross-section, the flexible substrate 202 having the semiconductor device 210 mounted on the upper surface thereof with bonds 206 to the bond pads 208 (not shown) of the semiconductor device 210 and conductors 204 of the substrate 202.

Referring to drawing FIG. 8B, a portion of the flexible substrate 202 is illustrated having aperture 216 therein having a solder ball 220 contained therein for connection to an adjacent substrate 202.

Referring to drawing FIG. 9, a plurality of stacked assemblies 200 is illustrated being vertically stacked with the flexible substrates 202 extending therebetween and connections between the conductors 204 (not shown) of each flexible substrate 202 being made through apertures 216 in the substrates 202 by means of melted solder balls 220 extending therebetween in the apertures 216. Since polyamide tape is used as the substrate 202, the substrate is flexible and readily bends and complies to the vertically stacked arrangement of a plurality of assemblies 200. As illustrated, a stacked arrangement of assemblies 200 is made on one side of the bottom substrate 202. The assemblies 200 are aligned through the use of the alignment apertures 214 (not shown) in the flexible substrates 202 forming the stacked arrangement. If desired, a mechanical connection may be made to the melted solder balls 220 in the apertures 216 to connect the conductors 204 to the appropriate bond pad of the semiconductor device 210.

Each assembly 200 may be fabricated individually or in strip form and subsequently singulated.

Referring to drawing FIG. 10, a vertical stack of assemblies 200 is illustrated where the assemblies 200 are stacked on both sides of the bottom substrate 202 with interconnections between the conductors 204 (not shown) on the substrates 202 being made by melted solder balls 220 extending within apertures 216 of the substrates 202. The assemblies 200 are aligned through the use of the alignment apertures 214 (not shown) in the flexible substrates 202 forming the stacked arrangement. If desired, a mechanical connection may be made to the melted solder balls 220 in the apertures 216 to connect the conductors 204 to the appropriate bond pad of the semiconductor device 210.

Having thus described the invention, it will be understood that changes, revisions, additions, and deletions may be made to the invention which will come within the scope of the invention. Such may be required by the design of the semiconductor device and its attachment to the substrates and/or to adjacent assemblies of semiconductor devices.

Claims

1. A method of forming a semiconductor assembly comprising: securing a portion of an active surface of each semiconductor device of a plurality of semiconductor devices secured to a first side of a first substrate of a plurality of first substrates;extending a wire through an aperture in the first substrate of the plurality and connected to a portion of at least one circuit of a plurality of circuits on a second side of the first substrate of the plurality and a bond pad of a plurality of bond pads on the active surface of a semiconductor device of the plurality of semiconductor devices;forming a connection between a circuit on the first substrate of the plurality and a circuit of a second substrate; andforming a connection between a connection area on the first side of the first substrate of the plurality and a circuit on a bus bar.

2. A method for forming a densely stacked semiconductor device assembly having at least a semiconductor device mounted on a first substrate and at least a semiconductor device mounted on a second substrate, the assembly comprising: securing at least a portion of an active surface of the at least a semiconductor device to a first side of a first substrate of a plurality of first substrates;forming a wire and extending the wire through an aperture in the first substrate of the plurality of first substrates to connect to a portion of an at least one circuit on a second side of the first substrate of the plurality of substrates and at least one bond pad of a plurality of bond pads on an active surface of a semiconductor device of a plurality of semiconductor devices;forming at least one connection between at least one circuit on the first substrate of the plurality of first substrates and one of at least two circuits of a second substrate;forming a connection between one of a plurality of connection areas on the first side of the first substrate of the plurality of first substrates and at least one circuit on a bus bar; andforming an electrical component connected to a circuit of the at least two circuits on a first side of the second substrate.

3. The method for forming a densely stacked semiconductor device assembly of claim 2, wherein the second substrate includes at least one circuit on at least one of the first side and the second side thereof.

4. The method for forming a densely stacked semiconductor device assembly of claim 3, wherein at least one first substrate of the plurality is connected to the at least one circuit on the at least one of the first side and the second side of the second substrate.

5. The method for forming a densely stacked semiconductor device assembly of claim 2, wherein the plurality of first substrates is assembled in a first stack and a second stack on the first side of the second substrate.

6. The method for forming a densely stacked semiconductor device assembly of claim 5, further comprising: at least one connection between the at least one circuit on at least one first substrate in the first stack of first substrates and at least one of the at least two circuits of the second substrate;at least one connection between the at least one circuit on at least one first substrate in the second stack of first substrates and at least one other circuit of the at least two circuits of the second substrate; anda connection between one of the plurality of connection areas on the first side of a first substrate of the first stack of first substrates and the at least one circuit on the bus bar and a connection between one of the plurality of connection areas on the first side of a first substrate of the second stack of first substrates and the at least one circuit on the bus bar.

7. The method for forming a densely stacked semiconductor device assembly of claim 6, further comprising: providing an electrical component connected to at least two circuits on the second substrate.

8. The method for forming a densely stacked semiconductor device assembly of claim 7, wherein the electrical component includes a resistor.

9. The method for forming a densely stacked semiconductor device assembly of claim 8, wherein the resistor substantially affects the impedance of the at least one of the at least two circuits on the second substrate.

10. The method for forming a densely stacked semiconductor device assembly of claim 9, wherein the first stack of first substrates and the second stack of first substrates are included on each side of the second substrate.

11. An assembly method comprising: securing a portion of an active surface of a semiconductor device to a first side of a first substrate;extending a wire through an aperture in the first substrate connected to a portion of at least one circuit of a plurality of circuits on a second side of the first substrate and a bond pad of a plurality of bond pads on the active surface of the semiconductor device for forming a connection between a circuit on the first substrate and a circuit of a second substrate; andforming a connection between a connection area on the first side of the first substrate and a circuit on a bus bar.

12. An assembly method comprising: securing a portion of an active surface of a semiconductor device to a first side of a first substrate;extending a wire extending through an aperture in a first substrate connected to a portion of at least one circuit on a second side of the first substrate and at least one bond pad of plurality of bond pads on the active surface of a semiconductor device;forming at least one connection between at least one circuit on the first substrate and one of at least two circuits of a second substrate;forming a connection between one of a plurality of connection areas on the first side of the first substrate and the at least one circuit on a bus bar; andforming an electrical component connected to a circuit of at least two circuits on a first side of a second substrate.

13. The method of claim 12, wherein the second substrate includes at least one circuit on at least one of the first side and the second side thereof.

14. The method of claim 13, wherein the first substrate is connected to the at least one circuit on the at least one of the first side and the second side of the second substrate.

15. The method of claim 12, wherein a plurality of first substrates is assembled in a first stack and a second stack on the first side of the second substrate.

16. The method of claim 15, farther comprising: at least one connection between the at least one circuit on at least one first substrate in the first stack of first substrates and at least one of the at least two circuits of the second substrate;at least one connection between the at least one circuit on at least one first substrate in the second stack of first substrates and at least one other circuit of the at least two circuits of the second substrate; anda connection between one of the plurality of connection areas on the first side of a first substrate of the first stack of first substrates and the at least one circuit on the bus bar and a connection between one of the plurality of connection areas on the first side of a first substrate of the second stack of first substrates and the at least one circuit on the bus bar.

17. The method of claim 16, further comprising: installing an electrical component connected to at least two circuits on the second substrate.

18. The method of claim 17, wherein the electrical component includes a resistor.

19. The method of claim 18, wherein the resistor substantially affects the impedance of the one of the at least two circuits on the second substrate.

20. The method of claim 19, wherein the first stack of first substrates and the second stack of first substrates are included on each side of the second substrate.