Semiconductor device comprising power elements in juxtaposition order
09607945 ยท 2017-03-28
Assignee
Inventors
Cpc classification
H01L2924/00014
ELECTRICITY
H01L2924/00
ELECTRICITY
H01L23/5226
ELECTRICITY
H01L2224/05548
ELECTRICITY
H10D84/00
ELECTRICITY
H01L2924/00014
ELECTRICITY
H01L2924/00
ELECTRICITY
International classification
H01L27/06
ELECTRICITY
H01L23/538
ELECTRICITY
Abstract
A semiconductor device including a multiplicity of large current power elements with each power element divided into a multiplicity of divisional elements and arranged such that the power elements belonging to different power elements are arranged in a repetitive sequential order. The IC chip of the semiconductor device is formed to have output wires extending from the respective divisional elements connected to corresponding output pads without crossing other output wires. Arranged on the IC chip are output bumps in association with the respective output pads. A rewiring layer is provided having output coupling wires for connecting together the bumps that belong to the same power element and connecting them further to an external output electrode.
Claims
1. A semiconductor device comprising: N (N2) semiconductor power elements each comprising M (M2) divisional elements, said semiconductor device thus comprising a total of NM divisional elements, which are arranged such that divisional elements belonging to different ones of said N semiconductor power elements are arranged in juxtaposition and in a repetitive sequential order; N control circuits each being common to said M divisional elements of said respective N semiconductor power elements; NM output pads associated with said NM divisional elements; NM output wires for connecting each of said NM divisional elements to associated one of said NM output pads; NM power supply pads associated with said NM divisional elements; and NM power supply wires for connecting each of said NM divisional elements to associated one of said NM power supply pads, wherein said NM divisional elements are arranged along a first direction, said NM output pads and said NM power supply pads are arranged along a second direction perpendicular to the first direction such that said output pads and said power supply pads are located at opposite sides of corresponding ones of said divisional elements respectively, and wherein a first side of one divisional element of two neighboring divisional elements among said NM divisional elements is connected to a corresponding one of said output pads and a second side of the one divisional element is connected to a corresponding one of said power supply pads such that the first and second sides being opposite sides of the one divisional element along the second direction, and a first side of another divisional element of the two neighboring divisional elements among said NM divisional elements is connected to a corresponding one of said power supply pads and a second side of the another divisional element is connected to a corresponding one of said output pads such that the first and second sides being opposite sides of the another divisional element along the second direction.
2. The semiconductor device according to claim 1 further comprising: a semiconductor integrated circuit body having said NM divisional elements, said NM output pads, said NM output wires, said NM power supply pads, and said NM power supply wires; and an insulating layer formed on said semiconductor integrated circuit body.
3. The semiconductor device according to claim 2 further comprising: NM output electroconductive parts each formed on said NM output pads to pass through said insulating layer from an one side to another side, respectively; NM power supply electroconductive parts each formed on said NM power supply pads to pass through said insulating layer from the one side to the other side, respectively; and a rewiring layer formed on the other side of said insulating layer.
4. The semiconductor device according to claim 3, wherein said NM output electroconductive parts and said NM power supply electroconductive parts are bumps or posts.
5. The semiconductor device according to claim 3, wherein the rewiring layer includes: N output coupling lead wires connecting together said NM output electroconductive parts such that said NM output electroconductive parts are divided into groups corresponding to said semiconductor power elements respectively; and a single power supply coupling lead wire commonly connecting together said NM power supply electroconductive parts.
6. The semiconductor device according to claim 5 further comprising: N external output electrodes connected to said N output coupling lead wires respectively; and an external power supply electrode connected to said power supply coupling lead wire.
7. The semiconductor device according to claim 6, wherein said N external output electrodes are provided over some of said NM output electroconductive parts.
8. The semiconductor device according to claim 6, wherein said external power supply electrode is provided over one of said NM said power supply electroconductive parts.
9. The semiconductor device according to claim 6, wherein said N external output electrodes and said external power supply electrode are ball electrodes.
10. The semiconductor device according to claim 6, wherein said external power supply electrode receives a power supply voltage or a ground voltage.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
(13) The inventive semiconductor device will now be described in detail by way of example with reference to the accompanying drawings.
(14)
(15)
(16) The two divisional elements 1A-1 and 1A-2 are connected together by a signal wire 3A and an output wire 4A to form the power transistor 1A. The two divisional elements 1B-1 and 1B-2 are connected together by a signal wire 3B and an output wire 4B to form the power transistor 1B. Power supply wires 6, shown by phantom lines, are formed using a wiring layer different from the wiring layer for the output wires 4A and 4B, and connected to all the divisional elements 1A-1-1B-2. The output wires 4A and 4B are respectively connected to output pads 5A and 5B, and the power supply wires 6 are connected to a power supply pad 7. Other features of the IC chip of
(17) In the IC chip 100, the divisional element 1A-1 or 1A-2 belonging to the power transistor 1A and the divisional element 1B-1 or 1B-2 belonging to the power transistor 1B are arranged in juxtaposition. As a result, the distance between two corresponding portions (e.g. between portions Xa-1 and Xb-1, and between Xa-2 and Xb-2 of
(18) In the first embodiment, however, although the variations in the characteristics are improved, improvement is not satisfactory regarding the following points. In the first embodiment, the output wires 4A and 4B extending from the divisional elements 1A-1-1B-2 are preferably connected to the output pads 5A and 5B with as small variation in resistance as possible. In doing so, if the output pads arranged in the same angular position relative to the respective divisional elements, the output wires 4A and 4B connected to the output pads will cross each other if the same wiring layer is used. Moreover, if the power supply wires 6 are provided using the same wiring layer, they will cross the output wires 4A and 4B. Furthermore, the output wires 4A and 4B and the power supply wires 6 are required to have sufficient widths in order to suppress on-resistances of the power elements. However, since the wiring distances of the wires increase when such crossing of wires takes place, the widths of the lead wires must be increased to keep the on-resistances suppressed. Hence, when the output wires 4A and 4B and the power supply wires 6 are arranged in the regions between the divisional elements 1A-1-1B-2 and the output pads 5A and 5B, or between the divisional elements 1A-1-1B-2 and the power supply pad 7 as shown in
(19)
(20)
(21) In the example shown in
(22) The output wires 14A-1, 14B-1, 14A-2, and 14B-2 of the divisional elements 11A-1 1-B-2 are respectively connected to output pad 15A-1, 15B-1, 15A-2, and 15B-2.
(23) Moreover, power supply wires 16 extending from the divisional elements 11A-1-11B-2 are connected to a common power supply pad 17 using a wiring layer different from the wiring layer for the output wires 14A-1-14B-2. Incidentally, the power supply can alternatively be replaced by the ground. In this case, the power supply pad 17 is grounded, and the power supply wire 16 is rephrased as the grounding wire 16 and the power supply pad 17 as the grounding pad 17. This applies to other embodiments of the invention.
(24) In the IC chip body 10, the divisional elements 11A-1 and 11A-2 belonging to the power transistor 11A and the divisional elements 11B-1 and 11B-2 belonging to the power transistor 11B are arranged in close proximity. The corresponding portions (as marked as Xa-1 and Xb-1, and Xa-2 and Xb-2 in
(25) In the IC chip 10, output wires 14A-1-14B-2 extending from the divisional elements 11A-1-11B-2 are directly connected to the output pads 15A-1-15B-2. That is, the output wires 14A-1-14B-2 do not cross each other. Thus, on-resistances of the power transistors 11A and 11B including resistances of the wiring resistances can be minimized.
(26) It is noted that the output wires 14A-1-14B-2 cross the power supply wires 6. However, in the IC chip body 10, crossing does not matter, since the electric conduction layer for the output wires and that for the power supply wire are formed using different wiring layers.
(27) In a rewiring layer 20 formed on the IC chip body shown in
(28) The output bumps 21A-1-21A-2 associated with the power transistor 11A are connected together by an output coupling wire 22A, which is extended to a position where it is connected to an external output electrode 24A. The bumps 21B-1 and 21B-2 associated with the power transistor 11B are connected together by an output coupling wire 22B, which is extended to a position where it is connected to an external output electrode 24B. The power supply bump 23 is connected to a power supply bump electrode 25. It is noted that in this rewiring layer 20 the output coupling wires 22A and 22B do not cross each other or cross any other lead wires. Therefore, the output coupling wires can be formed with sufficient widths in one layer to minimize their resistances.
(29) It is noted that the external output electrodes 24A and 24B can alternatively be provided directly on the respective bumps 21A-1 and 21A-2 or on the respective bumps 21B-1 and 21B-2.
(30) The output coupling wires 22A and 22B are formed after the output bumps 21A-21B and the insulating layer are formed. The output coupling wires 22A and 22B are preferably formed of the same material, and formed to have the same thickness and the same length as the bumps. The external output electrodes 24A and 24B and the power supply bump electrode 25 may be provided in the form of, for example, ball electrodes and a bump electrode, respectively.
(31)
(32) As shown in
(33) In accordance with the second embodiment, each of the multiple power elements 11A and 11B is constituted of a multiplicity of divisional elements 11A-1-11B-2. The divisional elements belonging to different power elements are sequentially arranged in juxtaposition to thereby reduce relative variations in the characteristics of the power elements. Moreover, crossing of the output wires 14A-1-42B-2 is eliminated to suppress their layout area. In addition, only one rewiring layer 20 is used to provide the non-crossing output coupling wires 22A and 22B. Furthermore, since the rewiring layer 20 is formed to have the output coupling wires 22A and 22B that connect together the divisional elements associated with the same power element 11A or 11B, the semiconductor device of the invention can be used as an ordinary IC chip.
(34)
(35) In the IC chip body 10 of
(36) Formed on the IC chip body 10 is the same rewiring layer 20 as shown in
(37) In accordance with the third embodiment, although the length of the power supply wire 16 becomes larger as compared with that of the second embodiment, the power supply wire 16 can be formed together with the output wires 14A-1-14B-2 using the same wiring layer. The third embodiment can provide the same results as the second embodiment.
(38)
(39)
(40) The output pad 35A-1 and 35A-2 associated with the divisional elements 31A-1 and 31A-2 of the power transistor 31A are provided in proximity to the upper ends of the respective divisional elements, as shown. Power supply pads 37-1 and 37-3 are provided in proximity to the lower ends of the respective divisional elements, as shown. The output pads 35B-1 and 35B-2 associated with the divisional elements 31B-1 and 31B-2 are provided in proximity to the lower ends of the respective divisional elements, as shown. The power supply pads 37-2 and 374 are provided in proximity to the upper ends of the respective divisional elements, as shown. Two sets of the divisional elements 31A-1 and 31A-2 and of divisional elements 31B-1 and 31B-2 are respectively supplied with control signals from control circuits 32A and 32B via signal wires 33A and 33B. Thus, the divisional elements 31A-1 and 31A-2 are driven together as one power transistor 31A, and so are the divisional elements 31B-1 and 31B-2 driven as one transistor 31B.
(41) In this manner, the four (NM) divisional elements 35A-1-35B-2 are provided with respective power supply pads 37-1-374 and output pads 35A-1-35B-2. These power supply pads 37-1-37-4 and output pads 35A-1-35B-2 are connected to the divisional elements by the respective power supply wires and by the respective output wire (reference number omitted in
(42) The four (NM) output pads 35A-1-35B-2 are arranged in such a way that two of them associated with the divisional elements 31A-1 and 31A-2 belonging to the same power element 31A are each arranged at an angular position (e.g. upper position) relative to the associated divisional element as shown, while two of them associated with the divisional elements 31B-1 and 31B-2 belonging to the same power element 31B are each arranged at another angular position (e.g. lower position) relative to the associated divisional element as shown.
(43) The four (NM) power supply pads 37-1-37-4 are arranged in such a way that two of them associated with the divisional elements 31A-1 and 31A-2 belonging to the same power element 31A are each arranged at an angular position (e.g. lower position) relative to the associated divisional element as shown, while two of them associated with the divisional elements 31B-1 and 31B-2 belonging to the same power element 31B are each arranged at another angular position (e.g. upper position) relative to the associated divisional element.
(44) Thus, the output wires and the power supply wires can be arranged in considerably short length without crossing one another by arranging the output pads 35A-1-35B-2 and the power supply pads 37-1-37-4 in the configuration as described above.
(45) In the rewiring layer 40 of
(46) The output bumps 41A-1 and 41A-2 associated with the power transistor 31A are connected together by an output coupling lead wire 42A, which is extended to a point where it is connected to an external output electrode 44A. The bumps 41B-1 and 41B-2 associated with the power transistor 31B are connected together by an output coupling wire 42B, which is extended to a point where it is connected to an external output electrode 44B. The power supply bumps 43-1-43-4 are connected together by a power coupling wire 46, which is extended to a point where it is connected to an external device.
(47) Power supply electrodes 45-1 and 45-2 are connected to the power coupling wire 46 at two points, which are, in the example shown herein, the power supply bumps 43-1 and 43-4. In this rewiring layer 40, there is no crossing between the output coupling wires 42A and 42B nor between the output coupling wires and the power coupling wire 46. Therefore, the output coupling wires 42A and 42B and the power coupling wire 46 can be implemented by one electric conduction layer. Other features of this semiconductor device are the same as those of the second and the third embodiments described above.
(48) It is noted that in the fourth embodiment output wires and power supply wires never cross one another, that their lengths can be very short, and that the same results can be attained as in the second embodiment.
(49)
(50) The IC chip body 50 shown in
(51) More particularly, symbols 51A-1-51B-3 indicate the respective divisional elements of the power transistors 51A and 51B; symbols 52A and 52B, control circuits; symbols 53A and 53B, signal wires; symbols 55A-1-55B-3, output pads; and symbols 57-1-57-6, power supply pads. Further symbols 61A-1-61B-3 indicate output bumps; symbols 62A and 62B, output coupling wires; symbols 63-1-63-6, power supply bumps; symbols 64A and 64B, external output electrodes; symbols 65-1 and 65-2, external power supply electrodes; and symbol 66, power supply coupling wires.
(52) In the fifth embodiment, multiplicity M of divisional elements per power element can be increased while attaining the same results as the fourth embodiment.
(53)
(54) In the example shown in
(55) More particularly, symbols 71A-1-71C-2 denote the respective divisional elements of the power transistors 71A, 71B, and 71C; symbols 72A, 72B, and 72C, control circuits; symbols 73A, 73B, and 73C, signal wires; symbols 74A-1-74C-2, output wires; symbols 75A-1-75C-2, output pads; symbols 76, power supply wires; and symbol 77, power supply pad. Further symbols 81A-1-81C-2 denote output bumps; symbols 82A, 82B, and 82C, output coupling wires; symbols 83, a power supply bump; symbols 84A, 84B, and 84C, external output electrodes; and symbols 85, an external power supply electrode.
(56) In this sixth embodiment, output wires can connect the respective blocks on the same plane without crossing one another if the number N of power elements is increased, thereby providing the same results as the preceding embodiments.
(57) Although the invention has been described above only for the cases with M3 or N3, it will be apparent that the invention can be extended to cases with M>3 and N>3, facilitating minimization of wiring resistance and a wiring space in a semiconductor device.