NESTED DESIGN APPROACH

Abstract
A structure for a system of chip packages includes a master substrate and at least one subset substrate of the master substrate. The subset substrate includes a portion of the master substrate that has an identical pin out pattern as the portion of the master substrate. The subset substrate has identical internal net lists as the portion of the master substrate. The subset substrate is adapted to accommodate a smaller chip than the master substrate. The master substrate is the largest substrate in the system. The invention also prepares a system of chip packages. The invention selects a master substrate and then selects a subset substrate of the master substrate.
Description
BACKGROUND OF THE INVENTION

1. Field of the Invention


The present invention generally relates to chip package designs and more particularly to an improved design that pre-assigns the contact pad array pattern and associated net list for a family of chips.


2. Description of the Related Art


Semiconductor chips are generally encased in packages that are attached to a printed circuit board. The packages protect the chip from environmental degradation and form electrical power and signal connections to the printed circuit board. The chips generally perform different functions some of which are less complex than others. Therefore, the chips often have different sizes (sometimes relating to their complexity).


One problem encountered by designers is that each chip package (substrate) must be uniquely designed for each newly designed chip. This is true for even those chips that are in the same family (same family corresponding to the same technology node and device type, and sometimes extending to the same topology and/or I/O structure and pinout). The invention described below overcomes such problems by utilizing pre-assigned pin-out patterns on the chip packages.


SUMMARY OF THE INVENTION

In view of the foregoing and other problems, disadvantages, and drawbacks of the conventional chip packages and associated design methodology limitations, the present invention has been devised, and it is an object of the present invention to provide a structure and method for an improved chip package and design methodology.


There is provided, according to one aspect of the invention, a menu of chip packages that have a master substrate and at least one subset substrate of the master substrate. The subset substrate is a portion of the master substrate and has an identical pin out pattern as that portion of the master substrate. The subset substrate has identical internal net lists as that portion of the master substrate. The subset substrate is adapted to accommodate a smaller chip than the master substrate. The master substrate is the largest substrate in the menu. The invention also prepares a menu of chip packages. The invention selects a master substrate and then selects a subset substrate of the master substrate.


The master design represents the largest possible logical netlist and largest physical wiring possible for the given die and package combination. Further, the logical netlist is not changed or revised in any way except that programmatic (e.g., from the outermost I/O proceeding inboard, in series, as required) deletions are allowed. Thus, the derived subordinate netlist/package is an exact identical subset of the parent master netlist/package.


Thus, as explained above, when different sizes of chips from the same family are placed on multiple substrate sizes, the invention designs the substrate with the largest body size first. Designs for the smaller body sizes are then subsets of the initial design. The bottom surface pad assignments for each physical location are made common between each substrate size to enable this sharing of design structures to produce the cost and time savings associated with the invention.


Traditionally, each substrate design has been done from scratch, which utilizes much more design resources, and requires a long cycle time to produce each design or netlist. With the invention, the subset netlists can be delivered to the customer much more quickly, by just running through a program to obtain the subset netlist, and the smaller designs can quickly be obtained, reducing the design cycle time.




BRIEF DESCRIPTION OF THE DRAWINGS

The foregoing and other objects, aspects and advantages will be better understood from the following detailed description of preferred embodiments of the invention with reference to the drawings, in which:



FIG. 1 is a schematic diagram of different ball grid array patterns which shows how the smaller chip package array patterns are subsets of the array pattern of the master; and



FIGS. 2A and 2B are schematic diagrams of the BSM assignments of the chip package using the master netlist and the BSM assignments of the chip package using a subset netlist of the master netlist;



FIGS. 3A and 3B are schematic diagrams of the die footprint on the TSM of the master netlist and a subset netlist of the master netlist;



FIGS. 4A and 4B are schematic diagrams of the first internal layer wiring of the V2 power level connections for the chip package using the master netlist and a subset netlist of the master netlist;



FIGS. 5A and 5B are schematic diagrams of the first internal layer wiring of the ground connections for the chip package using the master netlist and a subset netlist of the master netlist;



FIGS. 6A and 6B are schematic diagrams of the first internal layer wiring of the V1 power level connections for the chip package using the master netlist and a subset netlist of the master netlist;



FIGS. 7A and 7B are schematic diagrams of the first internal layer wiring of the signal connections for the chip package using the master netlist and a subset netlist of the master netlist of an intermediate level within a chip package;



FIGS. 8A and 8B are schematic diagrams of the second internal layer wiring of the ground connections for the chip package using the master netlist and a subset netlist of the master netlist;



FIGS. 9A and 9B are schematic diagrams of the second internal layer wiring of the signal connections for the chip package using the master netlist and a subset netlist of the master netlist;



FIGS. 10A and 10B are schematic diagrams of the second internal layer wiring of the V1 power level connections for the chip package using the master netlist and a subset netlist of the master netlist;



FIGS. 11A and 11B are schematic diagrams of the third internal layer wiring of the signal connections for the chip package using the master netlist and a subset netlist of the master netlist;



FIGS. 12A and 12B are schematic diagrams of the third internal layer wiring of the ground connections for the chip package using the master netlist and a subset netlist of the master netlist;



FIGS. 13A and 134B are schematic diagrams of the second internal layer wiring of the V2 power level connections for the chip package using the master netlist and a subset netlist of the master netlist; and



FIGS. 14A and 14B are schematic diagrams of the BSM array pattern for the chip package using the master netlist and a subset netlist of the master netlist.




DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS OF THE INVENTION

As mentioned above, the invention reduces the amount of design time and effort required which allows new products to be brought to the market earlier and reduces the cost of creating such products. Generally, the invention designs the wiring connections of a package to allow that package to be used generically with a wide variety of semiconductor chips. The invention creates a master (or super-substrate) design the of the largest substrate (chip package) that will be included within a menu of substrates. Substrates that are smaller than the master substrate are subsets of the master design. Thus, the invention represents both an improved design methodology, wherein design reuse of the super-substrate is supported for all design subsets therefor, and an improved package, whereupon multiple new packages for multiple die (chips) of different sizes, pursuant to common netlist hierarchy (I/O and functional) considerations, may be extracted from the super-substrate by a simple programmatic deletion of nets.


Therefore, once the master design is completed, smaller substrates can be created by programmatically deleting net lists from the larger master substrate. With the invention, the pin-out from the top of the substrate (at the flip-chip bumps) to the bottom of the substrate (at the bottom surface metallurgy (BSM) pads) is predetermine for a menu of substrates. Thus, with the invention, the pattern of the BSM pads is pre-assigned for all substrates (packages) within a given menu and there is no need to customize the substrates for each different chip. This is possible by virtue of the improved design re-use methodology wherein the super-substrate netlist is a logical super-set of inferior (smaller, subordinate) netlists, and the physical topology of the I/O (both TSM and BSM planes) support the logical deletion of unneeded nets for subordinate packages. Such logical or programmatic deletion of unneeded nets is predicated upon a 3-D fanout of said nets through the package such that the outermost I/O at the TSM plane remains the outermost row at the BSM plane (thus, the nets do not physically cross). This procedure supports the programmatic deletion of nets in derived packages of subordinate chips in that nets corresponding to chip I/Os from the outermost periphery of the die are deleted first and deletion continues inboard to more internal I/O as required.


For example, as shown in FIG. 1, the BSM design of a master substrate is shown. In this example, the master substrate has a size of 42.5×42.5 mm. In addition, FIG. 1 illustrates a number of subsets of the master substrate (shown in different scaling) down to the smallest substrate which is 25×25 mm. As shown in FIG. 1, in order to create a subset of the master substrate, it is only necessary to remove selected portions of the master substrate. The remaining BSM pads do not need to be redesigned, nor do the electrical connections which are formed above the BSM pads need to be redesigned. The substrate could comprise ceramic, organic, plastic, semiconductor, etc.



FIGS. 2A and 2B show internal wiring net lists within chip packages (substrates). The net list in FIG. 2A is the master substrate, while the one shown in FIG. 2B is a subset of the master shown in FIG. 2A. As can be seen by comparing FIGS. 2A and 2B, the designs are identical except that the outer portion (peripheral portions) of the master design in FIG. 2A has been eliminated from the subset design in FIG. 2B. The substrate shown in FIG. 2A is substantially larger and would be useful with a substantially larger chip than the substrate shown in FIG. 2B. The package of FIG. 2B can be designed in just a few hours by simple programmatic deletion of the external most unneeded nets in 2A. A fresh design of the package shown in FIG. 2B would otherwise require about two weeks. So again, the invention provides a design methodology improvement which results from design reuse, predicated upon two features of the invention: 1) logical super-netlist employing a master or super-substrate (e.g., largest supportable netlist for the chip and package combination) method; and 2) physical super-topology employing uncrossed physical nets, deleting unneeded nets from the outermost die I/O first and proceeding inward as required.



FIGS. 3A-14B represent the various layers within two different substrates (chip packages). More specifically, the “A” figures represent the master substrate while the “B” figures represent a subset of the master within the same menu. FIGS. 3A-3B and 14A-14B represent the top and bottom of the substrate, respectively. FIGS. 4A-13B illustrate the successive layers within the substrate.


As can be seen by comparing the “A” and “B” figures within each layer of the different substrates, the electrical connections and wiring positions are identical for the master and the subset except that the subset does not include the peripheral region 30 that is included within the master substrate. As explained above, this allows the subsets of the master to be very easily and quickly designed, which produces cost savings and reduces the time it takes to bring a product to market. The substrate or package designer can leverage the invention, new design re-use methodology, to derive subordinate logical netlists and substrates/packages quickly and easily, saving time and money.


The master design represents the largest possible logical netlist and largest physical wiring possible for the given die and package combination. Further, the logical netlist is not changed or revised in any way except that programmatic (e.g., from the outermost I/O proceeding inboard, in series, as required) deletions are allowed. Thus, the derived subordinate netlist/package is an exact identical subset of the parent master netlist/package. The comparison of the A and B figures in each case show how, for each layer, the subset chip package (B) is derived simply by the deletion of wiring on each layer from the master chip package (A).


The invention discloses a hierarchical system of congruent logical and topological coherence chip package that has a master substrate and at least one subset substrate of the master substrate. The master substrate has the largest logical and physical rendition supportable by the chip packages. The subset substrate is derived by programmatic deletion of unneeded nets from the master substrate. Congruent chip packages mean that for a given menu of chip packages having at least one master chip package and a derived, smaller chip package, all of the wiring nets of the derived, smaller chip package are common to, and congruent (i.e., coincident) with, a subset of the wiring nets of the master chip package.


Thus, as explained above, when different sizes of chips from the same family are placed on multiple substrate sizes, the invention designs the substrate with the largest body size first. Designs for the smaller body sizes are then subsets of the initial design. The bottom surface pad assignments for each physical location are made common between each substrate size to enable this sharing of design structures to produce the cost and time savings associated with the invention.


Traditionally, each substrate design has been done from scratch, which utilizes much more design resources, and requires a long cycle time to produce each design or netlist. With the invention, the subset netlists can be delivered to the customer much more quickly, by just running through a program to obtain the subset netlist, and the smaller designs can quickly be obtained, reducing the design cycle time.


While the invention has been described in terms of preferred embodiments, those skilled in the art will recognize that the invention can be practiced with modification within the spirit and scope of the appended claims.

Claims
  • 1. A hierarchical system of chip packages comprising: a master substrate; and at least one subset substrate of said master substrate.
  • 2. The system in claim 1, wherein said subset substrate comprises a congruent logical and topologically coherent portion of said master substrate.
  • 3. The system in claim 2, wherein said subset substrate has a pin out pattern commensurate to said portion of said master substrate, such that said subset substrate is both a logical and physical subset of the master.
  • 4. The system in claim 2, wherein said subset substrate has identical internal net lists as said portion of said master substrate.
  • 5. The system in claim 1, wherein said subset substrate is adapted to accommodate a smaller chip than said master substrate.
  • 6. The system in claim 1, wherein said master substrate is the largest substrate in said hierarchical system.
  • 7. The system in claim 1, wherein said master substrate comprises the largest logical and physical rendition supportable by said chip packages; and said subset substrate is derived by programmatic deletion of unneeded nets from said master substrate.
  • 8. A hierarchical system of chip packages comprising: a master substrate; and at least one subset substrate of said master substrate, wherein said master substrate comprises the largest logical and physical rendition supportable by said chip packages; and said subset substrate is derived by programmatic deletion of unneeded nets from said master substrate.
  • 9. The system in claim 8, wherein said subset substrate comprises a congruent logical and topologically coherent portion of said master substrate.
  • 10. The system in claim 9, wherein said subset substrate has a pin out pattern commensurate to said portion of said master substrate, such that said subset substrate is both a logical and physical subset of the master.
  • 11. The system in claim 9, wherein said subset substrate has identical internal net lists as said portion of said master substrate.
  • 12. The system in claim 8, wherein said subset substrate is adapted to accommodate a smaller chip than said master substrate.
  • 13. The system in claim 8, wherein said master substrate is the largest substrate in said hierarchical system.
  • 14. A method of forming a hierarchical system of chip packages, said method comprising: designing a master substrate; and selecting at least one subset substrate of said master substrate.
  • 15. The method in claim 14, wherein said subset substrate comprises a congruent logical and topologically coherent portion of said master substrate.
  • 16. The method in claim 15, wherein said subset substrate has a pin out pattern commensurate to said portion of said master substrate, such that said subset substrate is both a logical and physical subset of the master.
  • 17. The method in claim 15, wherein said subset substrate has identical internal net lists as said portion of said master substrate.
  • 18. The method in claim 14, wherein said subset substrate is adapted to accommodate a smaller chip than said master substrate.
  • 19. The method in claim 14, wherein said master substrate is the largest substrate in said hierarchical system.
  • 20. The method in claim 14, wherein said master substrate comprises the largest logical and physical rendition supportable by said chip packages; and said subset substrate is derived by programmatic deletion of unneeded nets from said master substrate.
Continuations (1)
Number Date Country
Parent PCT/US02/40671 Dec 2002 US
Child 11160307 Jun 2005 US