AI acceleration chip

By adopting a detachable core plate and jacket structure on the AI ​​acceleration chip, it uses copper rods, side rods and other components to achieve rapid separation, which solves the problem of damage to the connection of the existing chip when it is impacted, and reduces CPU damage and replacement costs.

CN222259943UActive Publication Date: 2024-12-27SHENZHEN CHUANGXINKE TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202421059652.8
Authority / Receiving Office
CN · China
Patent Type
Utility models(China)
Current Assignee / Owner
Filing Date
2024-05-15
Publication Date
2024-12-27
Estimated Expiration
2034-05-15

AI Technical Summary

Technical Problem

Existing chips are prone to damage when the equipment is shaken, impacted or squeezed by the circuit board, resulting in increased repair and replacement costs.

Method used

An AI acceleration chip was designed, adopting a detachable core plate and jacket structure. The core plate and the core plate are realized through the copper rod, side rod, key hole, copper bottom groove, side groove, guard plate and spring components. The rapid separation of the jacket cover protects the CPU from direct damage.

Benefits of technology

In case of accidental damage or impact, the core plate can be quickly separated from the jacket shell, reducing the possibility of CPU damage, ensuring the integrity of important parts, and reducing replacement costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN222259943U_ABST
    Figure CN222259943U_ABST
Patent Text Reader

Abstract

The utility model belongs to the technical field of semiconductors, and particularly relates to an AI acceleration chip which comprises an outer sleeve shell and a core plate, the core plate is detachably assembled in the outer sleeve shell, and a CPU is fixedly installed on the surface of the core plate. Copper rods are fixedly arranged on the inner walls of the two sides of the outer sleeve shell at equal intervals, edge clamping rods are fixedly connected to the inner walls of the two sides of the outer sleeve shell and located at the two ends of the corresponding set of copper rods, and lock holes are formed in the inner walls of the other two sides of the outer sleeve shell; copper bottom grooves are formed in the outer walls of the two sides of the core plate at equal intervals, edge clamping grooves are formed in the outer walls of the two sides of the core plate and located at the two ends of the corresponding set of copper bottom grooves, protection plates used for protecting a CPU are movably assembled on the surface of the core plate, and clamping plates used for being temporarily clamped into lock holes are fixedly arranged on the outer walls of the sleeves. According to the utility model, the traditional integrated chip structure is changed, and the core plate which can be separated from the outer sleeve shell can directly carry the CPU to be quickly separated when being extruded.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The utility model belongs to the field of semiconductor technology, and specifically relates to an AI acceleration chip. Background Art

[0002] Integrated circuit in English: integrated circuit, abbreviated as IC; also known as microcircuit, microchip, wafer / chip. In electronics, it is a way to miniaturize circuits (mainly semiconductor devices, but also passive components, etc.) and is often manufactured on the surface of semiconductor wafers.

[0003] Problems with existing technologies:

[0004] Most of the existing chips are installed on the circuit board in a fixed manner. The connection between the chip and the circuit board is a place with relatively weak structural strength. Therefore, when the equipment is shaken, collided or even the circuit board is directly squeezed, the connection between the chip and the circuit board or the chip itself may be directly damaged. When the entire circuit board is damaged, the damage to important components will double the cost of repair and replacement. Utility Model Content

[0005] The purpose of the utility model is to provide an AI acceleration chip, which can change the traditional integrated chip structure. The core board that can be separated from the outer shell can directly carry the CPU and quickly detach when it is squeezed.

[0006] The technical solution adopted by the utility model is as follows:

[0007] An AI acceleration chip, comprising an outer shell and a core board, wherein the core board is detachably assembled inside the outer shell, and a CPU is fixedly mounted on the surface of the core board;

[0008] The inner walls on both sides of the outer shell are equidistantly fixed with copper rods, and the inner walls on both sides of the outer shell are located at both ends of a corresponding group of copper rods and are fixedly connected with side clamp rods, and the inner walls on the other two sides of the outer shell are provided with lock holes;

[0009] The outer walls on both sides of the core board are equidistantly provided with copper bottom grooves, and the copper rods are embedded in the corresponding copper bottom grooves. The outer walls on both sides of the core board are located at both ends of a corresponding group of copper bottom grooves and are provided with side card grooves, and the side card rods are embedded in the corresponding side card grooves.

[0010] The surfaces of the other two sides of the core plate are obliquely symmetrically provided with side grooves, and one end of the two side grooves is fixedly connected with a through rod, the surface of the core plate is movably assembled with a guard plate for protecting the CPU, and the two ends of the guard plate diagonally opposite are fixedly connected with end tubes, the through rod is movably inserted into the corresponding end tube, and a spring is connected between the top of the through rod and the inner wall of the end tube;

[0011] The outer wall of the end tube is symmetrically fixed with spherical protrusions, the outer surface of the end tube is movably sleeved with a sleeve, both sides of the inner wall of the sleeve are provided with spiral guide grooves, the spherical protrusions are embedded and slidably installed in the corresponding spiral guide grooves, and the outer wall of the sleeve is fixedly provided with a clamping plate for temporarily inserting into the lock hole.

[0012] The edge of the inner side surface of the outer shell is fixedly provided with height-increasing pins in an equidistant array.

[0013] The inside of the side card slot is fixedly installed with a spring sheet for ejecting the side card rod into the side card slot.

[0014] The two clamping plates are arranged in an oblique symmetric manner.

[0015] The technical effects achieved by the utility model are:

[0016] (1) The utility model changes the traditional integrated fixed chip structure. The core board that can be separated from the outer shell can carry the CPU and quickly detach. In the event of unexpected events such as accidental damage, a mechanism can be provided to protect the CPU, thereby reducing the possibility of direct damage to the CPU.

[0017] (2) The utility model can quickly separate the core board carrying the CPU from the outer shell when the chip is hit, so that the core board is in a flexible and free state, the CPU is as far away from the impact as possible, and the integrity of important parts is guaranteed as much as possible, which indirectly reduces the cost of parts replacement. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is an integrated structure diagram of a chip provided by an embodiment of the utility model;

[0019] Figure 2 This is a disassembled diagram of the outer shell and the core plate provided in the embodiment of the utility model;

[0020] Figure 3 This is a disassembled diagram of a core plate and a guard plate provided in an embodiment of the utility model;

[0021] Figure 4 It is a structural diagram of a card board provided in an embodiment of the utility model.

[0022] In the accompanying drawings, the components represented by the reference numerals are listed as follows:

[0023] 1. Outer shell; 101. Copper rod; 102. Side clamp rod; 103. Keyhole; 104. Heightening pin; 2. Core board; 201. Copper bottom groove; 202. Side clamp groove; 203. Spring sheet; 204. Side groove; 205. Through rod; 206. Guard plate; 207. End tube; 208. Spring; 209. Spherical protrusion; 210. Sleeve; 211. Clamp board; 212. Spiral guide groove; 3. CPU. DETAILED DESCRIPTION

[0024] In order to make the purpose and advantages of the utility model more clear, the utility model is specifically described in combination with the following embodiments. It should be understood that the following text is only used to describe one or several specific implementations of the utility model, and does not strictly limit the protection scope of the specific request of the utility model.

[0025] like Figure 1-4 As shown, an AI acceleration chip includes an outer shell 1 and a core board 2. The core board 2 can be detachably assembled inside the outer shell 1, and a CPU 3 is fixedly installed on the surface of the core board 2.

[0026] See attached Figure 2 The inner walls on both sides of the outer shell 1 are fixedly provided with copper rods 101 at equal distances, and the inner walls on both sides of the outer shell 1 are fixedly connected with side clamp rods 102 at both ends of a corresponding group of copper rods 101.

[0027] See attached Figure 3 The outer walls on both sides of the core board 2 are equidistantly provided with copper bottom grooves 201, and the copper rods 101 are embedded in the corresponding copper bottom grooves 201. The outer walls on both sides of the core board 2 are provided with side card grooves 202 at both ends of a corresponding group of copper bottom grooves 201, and the side card rods 102 are embedded in the corresponding side card grooves 202. The side card grooves 202 are fixedly provided with spring sheets 203 for popping the side card rods 102 out of the side card grooves 202.

[0028] According to the above structure, the structure of the chip is that the CPU3 used for calculation is installed on the surface of the core board 2, and the core board 2 is detachably assembled inside the outer shell 1, wherein the copper rod 101 is embedded in the corresponding copper bottom groove 201, mainly used as a circuit for data transmission, and the side card rod 102 is embedded in the corresponding side card groove 202 and squeezes the spring sheet 203. When the fixed state between the outer shell 1 and the core board 2 is released, the two can be quickly separated by the elastic force of the spring sheet 203. The above structure changes the traditional one-piece fixed chip structure. The core board 2 that can be separated from the outer shell 1 can carry the CPU3 to quickly detach, and can provide a mechanism to protect the CPU3 in emergencies such as accidental damage, thereby reducing the possibility of direct damage to the CPU3.

[0029] See attached Figure 2 The inner walls on the other two sides of the outer shell 1 are provided with locking holes 103, and the edge of the inner surface of the outer shell 1 is fixedly provided with height-increasing pins 104 in an equidistant array.

[0030] See attached Figure 3 , the surface of the other two sides of the core plate 2 is obliquely symmetrically provided with side grooves 204, and one end of the two side grooves 204 is fixedly connected with a through rod 205, the surface of the core plate 2 is movably assembled with a guard plate 206 for protecting the CPU 3, and the two ends of the guard plate 206 diagonally opposite are fixedly connected with end tubes 207, the through rod 205 is movably inserted into the corresponding end tube 207, and a spring 208 is connected between the top of the through rod 205 and the inner wall of the end tube 207;

[0031] See attached Figure 3 and Figure 4 The outer wall of the end tube 207 is symmetrically fixed with a spherical protrusion 209, and the outer surface of the end tube 207 is movably sleeved with a sleeve 210. Both sides of the inner wall of the sleeve 210 are provided with spiral guide grooves 212. The spherical protrusion 209 is embedded and slidably installed in the corresponding spiral guide grooves 212. The outer wall of the sleeve 210 is fixedly provided with a clamping plate 211 for temporarily inserting into the lock hole 103, and the two clamping plates 211 are arranged in an oblique symmetrical manner.

[0032] According to the above structure, when the chip is about to be impacted, the guard plate 206 will replace the CPU3 to receive the impact first, and then the guard plate 206 will approach the core plate 2, and the through rod 205 will be correspondingly retracted into the end tube 207 and squeeze the spring 208. At the same time, the end tube 207 will move relative to the sleeve 210. Since the spherical protrusion 209 on the outer wall of the end tube 207 is embedded in the spiral guide groove 212 on the inner wall of the sleeve 210, according to the direction of the spiral guide groove 212, the sleeve 210 will carry the card plate 211 to rotate together. When the card plate 211 leaves the lock hole 103, under the elastic force of each spring sheet 203, the core plate 2 will be separated from the outer shell 1 and fall into the inner side of the heightened pin 104. The above process can quickly separate the core plate 2 carrying the CPU3 from the outer shell 1 when the chip is impacted, so that the core plate 2 is in a flexible and free state, and the CPU3 is kept as far away from the impact as possible, and the integrity of important components is guaranteed as much as possible, which indirectly reduces the cost of parts replacement.

[0033] The working principle of the utility model is: when the chip is about to be impacted, the guard plate 206 will replace the CPU3 to receive the impact first, and then the guard plate 206 will approach the core board 2, and the penetration rod 205 will be correspondingly retracted into the end tube 207 and squeeze the spring 208. At the same time, the end tube 207 moves relative to the sleeve 210. Since the spherical protrusion 209 on the outer wall of the end tube 207 is embedded in the spiral guide groove 212 on the inner wall of the sleeve 210, according to the direction of the spiral guide groove 212, the sleeve 210 will carry the card plate 211 to rotate together. When the card plate 211 leaves the lock hole 103, under the elastic force of each spring sheet 203, the core board 2 will be separated from the outer shell 1 and fall into the inner side of the heightened pin 104.

[0034] The above is only a preferred embodiment of the present invention. It should be noted that, for ordinary technicians in the technical field, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be regarded as the protection scope of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention shall be implemented according to the conventional means in the field unless otherwise specified and limited.

Claims

1. An AI acceleration chip, comprising an outer shell (1) and a core board (2), characterized in that: The core plate (2) is detachably assembled inside the outer shell (1), and a CPU (3) is fixedly mounted on the surface of the core plate (2); Copper rods (101) are fixedly arranged at equal distances on the inner walls of both sides of the outer shell (1), and side clamp rods (102) are fixedly connected to the inner walls of both sides of the outer shell (1) at both ends of a corresponding group of copper rods (101), and lock holes (103) are opened on the inner walls of the other two sides of the outer shell (1); The outer walls of both sides of the core board (2) are provided with copper bottom grooves (201) at equal intervals, and the copper rods (101) are embedded in the corresponding copper bottom grooves (201); the outer walls of both sides of the core board (2) are provided with side clamping grooves (202) at both ends of a corresponding group of copper bottom grooves (201), and the side clamping rods (102) are embedded in the corresponding side clamping grooves (202); The surfaces of the other two sides of the core plate (2) are obliquely symmetrically provided with side grooves (204), and one end of each of the two side grooves (204) is fixedly connected to a through rod (205). A protective plate (206) for protecting the CPU (3) is movably assembled on the surface of the core plate (2), and both ends of the diagonally opposite sides of the protective plate (206) are fixedly connected to end tubes (207). The through rod (205) is movably inserted into the corresponding end tube (207), and a spring (208) is connected between the top of the through rod (205) and the inner wall of the end tube (207); The outer wall of the end tube (207) is symmetrically fixedly provided with a spherical protrusion (209); the outer surface of the end tube (207) is movably sleeved with a sleeve (210); both sides of the inner wall of the sleeve (210) are provided with spiral guide grooves (212); the spherical protrusion (209) is embedded and slidably installed in the corresponding spiral guide groove (212); and the outer wall of the sleeve (210) is fixedly provided with a clamping plate (211) for temporarily clamping into the lock hole (103).

2. The AI ​​acceleration chip according to claim 1, characterized in that: Heightening pins (104) are fixedly arranged in an equidistant array on the edge of the inner surface of the outer shell (1).

3. The AI ​​acceleration chip according to claim 1, characterized in that: A spring sheet (203) for ejecting the side clamping rod (102) out of the side clamping slot (202) is fixedly installed inside the side clamping slot (202).

4. The AI ​​acceleration chip according to claim 1, characterized in that: The two clamping plates (211) are arranged in an obliquely symmetrical manner.