A server mainboard memory stick plug-in test system

Abstract

The application discloses a kind of server mainboard memory stick plug-in test system, it is related to memory stick test technical field, including test main cabinet, the inside installation of test main cabinet is used for test test clamping track group, manipulator and control module, one end of manipulator is equipped with plug-in part, control module is connected with mechanical plug-in module, data acquisition module, algorithm processing module, result output module and early warning module, mechanical plug-in module is used to execute the automatic plug-in operation of manipulator, data acquisition module is used to collect plug-in test data, algorithm processing module is used to determine contact state score based on plug-in test data, calculate and determine fault type and generate parameter adjustment instruction, the quality and stability of the present application are evaluated by two dimensions, multi-source parameter fusion and quantitative adjustment, accurate identification is realized, false and true nature fault is accurately distinguished and dynamic optimization closed loop is formed, and contact state and system reliability are continuously improved.

Description

Technical Field

[0001] This invention relates to the field of memory module testing technology, specifically a server motherboard memory module insertion and removal testing system. Background Technology

[0002] The memory module insertion and removal test on the server motherboard is a core part of ensuring the reliability of the memory subsystem. As the capacity and speed of server memory increase, the contact performance between the memory module and the slot directly affects the stability of data transmission.

[0003] The shortcomings of existing technologies are mainly reflected in three aspects: incomplete contact condition assessment, inaccurate fault diagnosis, and lack of quantitative basis for parameter adjustment, as detailed below:

[0004] First, existing systems judge contact status solely by the absolute difference between real-time contact resistance and a reference value. However, in actual testing, instantaneous interference can cause short-term fluctuations, making it easy to misjudge based on absolute values ​​alone. Second, existing systems identify faults using only single indicators such as contact resistance or bit error rate. However, in practice, insufficient insertion / removal force or excessive speed can also lead to false faults, which existing methods cannot distinguish. This often results in false faults being misjudged as true faults or true faults being missed, leading to "single-dimensional, low-confidence" fault identification. Finally, when poor contact is detected, existing systems require manual adjustment of insertion / removal force / speed based on experience. However, the adjustment amount lacks mathematical basis, which may result in insufficient or excessive adjustment, leading to low efficiency and poor performance. Summary of the Invention

[0005] The purpose of this invention is to provide a server motherboard memory module insertion and removal testing system, which solves the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution, including a test main cabinet, wherein the test main cabinet is internally equipped with a test clamping track assembly, a robotic arm and a control module for testing, one end of the robotic arm is equipped with a plug-in part, and a four-wire resistance measurement sensor, a force sensor, a displacement sensor and a memory controller are integrated and installed near the plug-in part of the robotic arm.

[0007] The control module is connected to a mechanical plug-in module, a data acquisition module, an algorithm processing module, a result output module, and an early warning module;

[0008] The specific implementation is as follows:

[0009] The mechanical insertion / removal module is used to perform automatic insertion / removal operations of the robotic arm, including controlling and adjusting the insertion / removal force and speed.

[0010] The data acquisition module is used to acquire insertion and removal test data detected and transmitted by the four-wire resistance measurement sensor, the force sensor, the displacement sensor, and the memory controller.

[0011] The algorithm processing module includes a built-in contact state evaluation unit, a fault detection unit, and an adaptive adjustment unit.

[0012] Used to calculate and determine the contact status score, identify the fault type, and generate parameter adjustment instructions based on the insertion and removal test data;

[0013] The result output module is used to display the contact status score, the fault identification result, and the new parameter information after the parameter adjustment instruction is given, and supports test report generation and storage.

[0014] The early warning module is used to provide real-time early warnings for the contact status score and the fault type.

[0015] Optionally, the insertion and removal test data includes standard contact resistance value, real-time contact resistance value, real-time standard deviation of resistance fluctuation, historical standard deviation of resistance fluctuation, real-time force, real-time speed, standard insertion and removal force, standard insertion and removal speed, bit error rate, true and false fault threshold, adjustment coefficient and fault safety value.

[0016] The four-wire resistance measurement sensor acquires real-time contact resistance values ​​in real time using the four-wire measurement method and records historical resistance data to calculate the real-time standard deviation of resistance fluctuation.

[0017] The force sensor acquires real-time force and speed during the insertion and removal process by connecting a pressure sensor and a displacement encoder in series at the actuator end of the robotic arm.

[0018] The memory controller has a built-in bit error rate acquisition unit;

[0019] The bit error rate acquisition unit obtains the bit error rate by writing a preset data template to the memory module and reading and verifying it, and by calculating the ratio of the number of erroneous data bits to the total number of data bits during the read and write process.

[0020] Optionally, the specific logic for calculating and determining the contact status score by the built-in contact status assessment unit is as follows:

[0021] Based on the standard contact resistance value and the real-time contact resistance value, calculate the absolute resistance deviation value after multiplying the contact resistance by the weight;

[0022] Based on the real-time standard deviation of the resistance fluctuation and the historical standard deviation of the resistance fluctuation, calculate the stability deviation value of the resistance fluctuation multiplied by the weight;

[0023] The contact state score is determined by adding the absolute resistance deviation value and the stability deviation value.

[0024] Optionally, the fault discrimination unit is used to calculate and determine the fault discrimination index, and the specific logic is as follows:

[0025] The contact status score is set as the core parameter;

[0026] Combined with force deviation: Calculate the force deviation rate by multiplying the real-time force and the standard insertion / extraction force by a weight;

[0027] Combined with speed deviation: Calculate the speed deviation rate by multiplying the real-time speed and the standard insertion / removal speed by a weight;

[0028] Combined with the bit error rate: Calculate the correct bit rate after removing the bit error rate and multiplying it by the weight;

[0029] The contact status score is multiplied by a weight and then added sequentially to the force deviation rate, the speed deviation rate, and the correct code rate to calculate and determine the fault discrimination index.

[0030] Optionally, based on the result of the fault discrimination index, the fault type is determined as follows:

[0031] If the fault discrimination index is lower than the true / false fault threshold, the fault type is determined to be a true fault, and the early warning module issues an early warning without generating the parameter adjustment instruction.

[0032] If the fault discrimination index is higher than the true and false fault threshold, then the fault type is determined to be faultless and there is no need to generate the parameter adjustment instruction.

[0033] If the fault discrimination index is between the true / false fault threshold and the fault safety value, then the fault type is determined to be a false fault and the parameter adjustment instruction is generated.

[0034] Optionally, the adaptive adjustment unit is used to calculate and determine the parameter adjustment amount, and the specific logic is as follows:

[0035] Calculate the severity of the fault: Calculate the fault risk coefficient after removing the fault discrimination index;

[0036] Calculate the baseline fault risk coefficient after removing the true and false fault thresholds;

[0037] Calculate the proportion of the fault risk coefficient to the baseline fault risk coefficient to determine the fault severity coefficient;

[0038] Calculate the baseline adjustment amount: Calculate the adjustment coefficient multiplied by the sum of the standard insertion and extraction force and the standard insertion and extraction speed to determine the baseline parameter adjustment amount;

[0039] The parameter adjustment amount is determined based on the product of the fault severity coefficient and the parameter adjustment base amount.

[0040] Optionally, based on the result of the parameter adjustment amount, the specific generation of the parameter adjustment instruction is as follows:

[0041] L new =L0 + T × 0.6;

[0042] V new =V0 - T × 0.4;

[0043] in:

[0044] L new To adjust the intensity, V new The adjusted speed is represented by L0, which is the standard insertion / removal force, and V0, which is the standard insertion / removal speed. 0.6 and 0.4 are both adjustment coefficients.

[0045] The adjusted force L after generation new and the adjusted speed V new The results are received and transmitted by the output module and the control module, and then executed by the robotic arm.

[0046] Optionally, the standard contact resistance value and the standard deviation of the historical resistance fluctuation are obtained in the following ways:

[0047] 100 sets of motherboards and memory modules of the same model were selected, and their contact resistance and standard deviation were measured under the condition of no mechanical deviation. The average value was taken to determine the standard contact resistance value and the historical standard deviation of the resistance.

[0048] The standard insertion / removal force and the standard insertion / removal speed are obtained in the following ways:

[0049] The robotic arm was used to perform a force-contact resistance test on a motherboard of the same model: the speed was fixed and the force was adjusted.

[0050] Record the real-time force for each set of forces, and select the force with the smallest and most stable real-time force as the standard insertion and extraction force.

[0051] Similar to the force-contact resistance experiment, fix the force and adjust the speed;

[0052] Record the real-time speed for each set of speeds, and select the speed with the smallest real-time speed and the smallest fluctuation as the standard insertion and extraction speed;

[0053] Based on the 3σ principle, the critical value for true and false faults is determined to be 0.3;

[0054] The failsafe value is determined to be 0.5 based on the 5σ principle.

[0055] Optionally, the test clamping track group holds the server motherboard, and both sides of the test clamping track group are connected to the material conveying track group;

[0056] The test main cabinet is equipped with a display screen, and the result output module is connected to the display screen. An automatic board loading machine is installed on one side of the test main cabinet, and an automatic board unloading machine is installed on the other side of the test main cabinet away from the automatic board loading machine.

[0057] The automatic board loading machine is connected to the material conveying track group via a board loading track group for loading the server motherboard, and the automatic board unloading machine is connected to the material conveying track group via a board unloading track group for unloading the server motherboard.

[0058] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0059] I. The built-in contact state evaluation unit of the present invention uses the combination of absolute resistance deviation value and stability deviation value to calculate and determine the contact state score, thereby achieving the purpose of simultaneously evaluating the absolute quality stability of contact resistance, which avoids misjudgment due to instantaneous interference and can identify potential contact problems.

[0060] Second, the fault discrimination unit of the present invention constructs a fault discrimination index by fusing multi-source data such as contact state score, force deviation rate, speed deviation rate and correct code rate, and further distinguishes between "false" and "true" faults based on the fault type.

[0061] Third, the adaptive adjustment unit of the present invention calculates the severity of the fault and the baseline adjustment amount, outputs the parameter adjustment base amount, and adjusts the parameter adjustment base amount accordingly, and distributes the force and speed proportionally, thereby increasing the contact area and reducing collision damage in the subsequent insertion and removal test, and thus realizing intelligent closed-loop optimization. Attached Figure Description

[0062] Figure 1 This is a front view of the main structural device of the present invention;

[0063] Figure 2 This is an enlarged schematic diagram of the test clamping track assembly inside the test cabinet in this invention;

[0064] Figure 3 This is a schematic diagram showing the orientational connection between the robotic arm and the insertion / removal part in this invention;

[0065] Figure 4This is a flowchart illustrating the process of implementing insertion and removal tests for each module in the server motherboard memory module insertion and removal test system.

[0066] Figure 5 This is a schematic diagram of the overall module architecture within the system of this invention.

[0067] In the diagram: 1-Test main cabinet machine, 2-Display screen, 3-Automatic board loading machine, 4-Board loading track assembly, 5-Automatic board unloading machine, 6-Board unloading track assembly, 7-Material conveying track assembly, 8-Test clamping track assembly, 9-Server motherboard, 10-Robotic arm, 11-Plug-in / plug-out unit. Detailed Implementation

[0068] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0069] This server motherboard memory module insertion and removal test system differs from existing systems. Existing systems suffer from limitations such as simplistic contact status assessment, ambiguous fault diagnosis, lack of quantitative basis for adjustment, low efficiency, and poor performance. In contrast, this algorithm unit achieves comprehensive contact status assessment, accurate fault diagnosis, and intelligent parameter adjustment.

[0070] Example 1, please refer to Figures 1 to 5 This embodiment provides a server motherboard memory module insertion and removal testing system, including a test cabinet 1. The test cabinet 1 is equipped with a test clamping rail group 8, a robotic arm 10 and a control module. The test clamping rail group 8 holds a server motherboard 9. One end of the robotic arm 10 is equipped with an insertion and removal part 11. A four-wire resistance measurement sensor, a force sensor, a displacement sensor and a memory controller are integrated and installed near the insertion and removal part 11 on the robotic arm 10.

[0071] The control module is connected to a mechanical plug-in module, a data acquisition module, an algorithm processing module, a result output module, and an early warning module;

[0072] The specific implementation is as follows:

[0073] Mechanical insertion / removal module: used to perform automatic insertion / removal operations of robotic arm 10, including controlling and adjusting insertion / removal force and speed;

[0074] Data acquisition module: used to acquire and transmit insertion and removal test data detected by the four-wire resistance measurement sensor, force sensor, displacement sensor and memory controller;

[0075] Algorithm processing module: Includes built-in contact state evaluation unit, fault discrimination unit, and adaptive adjustment unit;

[0076] Used to calculate and determine contact status scores, identify fault types, and generate parameter adjustment instructions based on insertion and removal test data;

[0077] Results output module: Used to display contact status score, fault identification results and new parameter information after parameter adjustment instructions, and supports test report generation and storage;

[0078] Early warning module: Used to provide early warnings in real time by scoring the contact status and identifying the type of fault.

[0079] The insertion and removal test data includes standard contact resistance value, real-time contact resistance value, standard deviation of real-time resistance fluctuation, standard deviation of historical resistance fluctuation, real-time force, real-time speed, standard insertion and removal force, standard insertion and removal speed, bit error rate, critical value for true and false faults, adjustment coefficient, and fault safety value.

[0080] The four-wire resistance measurement sensor acquires real-time contact resistance values ​​using a four-wire measurement method and records historical resistance data to calculate the real-time standard deviation of resistance fluctuations.

[0081] The force sensor acquires the real-time force and speed during the insertion and removal process by connecting a pressure sensor and a displacement encoder in series at the actuator end of the robotic arm 10.

[0082] The memory controller has a built-in bit error rate acquisition unit;

[0083] The bit error rate acquisition unit obtains the bit error rate by writing a preset data template to the memory module and reading and verifying it, and by calculating the ratio of the number of erroneous data bits to the total number of data bits during the read and write process.

[0084] In the scenario of the test main cabinet 1 performing plug-in / plug-out tests in this embodiment, the control module issues operation instructions, the mechanical plug-in / plug-out module controls the robotic arm 10, and the robotic arm 10, together with the plug-in / plug-out part 11, performs automatic plug-in / plug-out operations on the server motherboard 9 and memory modules on the test clamping rail group 8.

[0085] During the insertion and removal process, the four-wire resistance measurement sensor, force sensor and displacement sensor are used for real-time sensing and detection. After signal conversion, historical standard data and real data are transmitted to the data acquisition module. At the same time, the memory controller sends read and write instructions to the memory module during the test and counts the number of errors that occur during the read and write process in real time, thereby obtaining the bit error rate, and also transmits the bit error rate to the data acquisition module.

[0086] The data acquisition module transmits the insertion and removal test data to the algorithm processing module. The algorithm processing module, through calculations by the contact state evaluation unit, fault discrimination unit, and adaptive adjustment unit, determines the contact state score, identifies the fault type, and generates parameter adjustment instructions and new parameter information.

[0087] The new parameter information is transmitted to the result output module and the control module at the same time. Based on the new parameter information, the control module controls the robotic arm 10 to perform a new round of automatic insertion and removal operations.

[0088] Furthermore, the standard contact resistance value and the standard deviation of historical resistance fluctuations are obtained as follows:

[0089] 100 sets of motherboards and memory modules of the same model were selected, and the contact resistance and standard deviation were measured under the condition of no mechanical deviation. The average value was taken to determine the standard contact resistance value and the standard deviation of the historical resistance fluctuation.

[0090] The methods for obtaining standard insertion / removal force and standard insertion / removal speed are as follows:

[0091] A force-contact resistance experiment was conducted on a motherboard of the same model using a robotic arm 10: the speed was fixed and the force was adjusted.

[0092] Record the real-time force for each set of forces, and select the force with the smallest and most stable real-time force as the standard insertion and extraction force.

[0093] Similar to the force-contact resistance experiment, fix the force and adjust the speed;

[0094] Record the real-time speed for each group of speeds, and select the speed with the smallest real-time speed and the smallest fluctuation as the standard insertion and extraction speed;

[0095] Based on the 3σ principle, the critical value for true and false faults is determined to be 0.3;

[0096] The fail-safe value is determined to be 0.5 based on the 5σ principle.

[0097] Please see Figures 4 to 5 The specific logic for calculating and determining the contact status score using the built-in contact status assessment unit is as follows:

[0098] Based on the standard contact resistance value and the real-time contact resistance value, calculate the absolute deviation of the resistance after multiplying the contact resistance by the weight;

[0099] Based on the real-time standard deviation of resistance fluctuation and the historical standard deviation of resistance fluctuation, calculate the stability deviation value of resistance fluctuation multiplied by weight;

[0100] The contact condition score is determined by adding the absolute resistance deviation value and the stability deviation value.

[0101] The 3σ principle is often used to identify "low-probability anomalies", while the 5σ principle, with a fault safety value of 0.5, corresponds to a higher confidence level requirement.

[0102] In this embodiment: First, the formula for calculating the contact state score in this algorithm unit is as follows:

[0103]

[0104] in:

[0105] SJ represents the contact condition score, D0 represents the standard contact resistance value, D represents the real-time contact resistance value, w1 represents the weighting coefficient, and BC represents the contact resistance score. HST BC represents the historical standard deviation of the resistance, while BC represents the real-time standard deviation of the resistance.

[0106] And 0.6≤w1≤0.8;

[0107] The result is the absolute resistance deviation value. The result is the stability deviation value;

[0108] If the real-time contact resistance value D > the standard contact resistance value D0, it indicates that the contact area is insufficient or there is oxidation, which leads to an increase in resistance.

[0109] If the real-time contact resistance value D < the standard contact resistance value D0, the contact may deform due to excessive force.

[0110] If the real-time standard deviation of resistance fluctuation BC > the historical standard deviation of resistance fluctuation BC HST This indicates that the contact resistance is affected by factors such as mechanical vibration and uneven insertion and removal speed, resulting in increased fluctuations.

[0111] The absolute resistance deviation value, as a major component of the contact condition score (SJ), quantifies the "absolute quality" of the contact resistance and is used to evaluate the basic contact performance between the memory module and the slot. The stability deviation value, as a supplementary part of the contact condition score (SJ), quantifies the "stability" of the contact resistance and avoids misjudgment of poor contact due to transient interference.

[0112] Please see Figures 4 to 5 The fault discrimination unit is used to calculate and determine the fault discrimination index, and the specific logic is as follows:

[0113] Set the contact status rating as the core parameter;

[0114] Combined with force deviation: Calculate the force deviation rate by multiplying the real-time force and the standard insertion / extraction force by a weight;

[0115] Combined speed deviation: Calculate the speed deviation rate by multiplying the real-time speed and the standard insertion / removal speed by a weight;

[0116] Combined with bit error rate: Calculate the correct bit rate after removing the bit error rate and multiplying it by the weight;

[0117] The contact status score is multiplied by a weight and then added sequentially to the force deviation rate, speed deviation rate, and correct code rate to calculate and determine the fault discrimination index.

[0118] In this embodiment, the formula for calculating the fault discrimination index is as follows:

[0119]

[0120] in:

[0121] GP is the fault discrimination index, L0 is the standard insertion / removal force, L is the real-time force, V0 is the standard insertion / removal speed, V is the real-time speed, and WW is the bit error rate.

[0122] w2, w3, w4, and w5 are all weighting coefficients, and w2 + w3 + w4 + w5 = 1;

[0123] The result is the force deviation rate. The result is the speed deviation rate, and the result of w5×(1-WW) is the fault discrimination index.

[0124] Furthermore, based on the results of the fault discrimination index, the fault type is determined as follows:

[0125] If the fault discrimination index is lower than the threshold for true and false faults, the fault type is determined to be a true fault, the early warning module issues an early warning and does not need to generate parameter adjustment instructions;

[0126] If the fault discrimination index is higher than the threshold for true and false faults, the fault type is determined to be faultless and no parameter adjustment command needs to be generated.

[0127] If the fault discrimination index is between the critical value for true and false faults and the fault safety value, then the fault type is determined to be a false fault and a parameter adjustment instruction is generated.

[0128] This algorithm unit comprehensively evaluates the degree of deviation of contact state, mechanical parameters, and electrical parameters by combining contact state scoring with multi-source data fusion of force deviation, speed deviation, and bit error rate. It outputs a fault discrimination index GP to distinguish between false faults caused by poor contact and true faults caused by hardware damage, and determines whether to trigger the calculation of the adaptive adjustment unit.

[0129] Furthermore, if it is determined to be a genuine fault, it must be repaired directly and cannot be improved by adjustment;

[0130] If the fault is determined to be false, it can be improved by triggering the calculation of the adaptive adjustment unit and adjusting 10 parameters of the robotic arm.

[0131] If it is determined to be fault-free, it reflects that the risk of failure is low and no adjustment is required;

[0132] The adaptive adjustment unit is triggered only in the event of a false fault. It is used to optimize the insertion and removal parameters of the robotic arm 10 to improve the fault discrimination index GP to a fault-safe value. When the fault discrimination index GP ≥ 0.5, it means that the current insertion and removal parameters have met the system's preset safety standards. The contact status of the memory module, the force / speed deviation, and the bit error rate are all within acceptable ranges, so no additional adjustment is required.

[0133] Please see Figures 4 to 5 The adaptive adjustment unit is used to calculate and determine the parameter adjustment amount, and the specific logic is as follows:

[0134] Calculate the severity of the fault: Calculate the fault risk coefficient after removing the fault discrimination index;

[0135] Calculate the baseline fault risk coefficient after removing the critical values ​​for true and false faults;

[0136] Calculate the proportion of the failure risk coefficient to the baseline failure risk coefficient to determine the failure severity coefficient;

[0137] Calculate the baseline adjustment amount: Calculate the adjustment coefficient by multiplying the sum of the standard insertion and extraction force and the standard insertion and extraction speed to determine the baseline adjustment amount of the parameters;

[0138] The parameter adjustment amount is determined by multiplying the fault severity coefficient and the parameter adjustment base amount.

[0139] In this embodiment, the algorithm unit first calculates the parameter adjustment amount using the following formula:

[0140]

[0141] in:

[0142] T is the parameter adjustment amount, GP CRI w6 is the threshold value for true and false faults, and w6 is the adjustment coefficient.

[0143] And 0.1≤w6≤0.2;

[0144] The result of 1-GP is the failure risk coefficient, 1-GP CRI The result is the baseline failure risk coefficient. The result is the fault severity coefficient, and the result of w6×(L0+V0) is the parameter adjustment base quantity.

[0145] Furthermore, based on the result of the parameter adjustment, the specific generation of the parameter adjustment instruction is as follows:

[0146] L new =L0 + T × 0.6;

[0147] V new =V0 - T × 0.4;

[0148] in:

[0149] L new To adjust the intensity, V new The adjusted speed is represented by L0, which is the standard insertion / removal force, and V0, which is the standard insertion / removal speed. 0.6 and 0.4 are both adjustment coefficients.

[0150] Adjusted strength L after generation new and the adjusted speed V new The results are received and transmitted by the output module and the control module, and then executed by the robotic arm 10.

[0151] In this algorithm unit, the parameter adjustment amount T is proportionally allocated to the force (+60%) and the velocity (-40%), which conforms to Hertzian contact theory and collision energy formula. Specifically, the increased force increases the contact area, while the decreased velocity reduces collision damage.

[0152] Example 2, please refer to Figures 1 to 5 Both sides of the test clamping track group 8 are connected to the material conveying track group 7;

[0153] The test main cabinet 1 is equipped with a display screen 2, and the result output module is connected to the display screen 2. An automatic board loading machine 3 is installed on one side of the test main cabinet 1, and an automatic board unloading machine 5 is installed on the other side of the test main cabinet 1 away from the automatic board loading machine 3.

[0154] The automatic loading machine 3 is connected to the material conveying track group 7 by a loading track group 4 for loading the server motherboard 9, and the automatic unloading machine 5 is connected to the material conveying track group 7 by a unloading track group 6 for unloading the server motherboard 9.

[0155] In this embodiment, the automatic board loading machine 3 starts and the server motherboard 9 containing memory modules to be tested is transferred to the material conveying track group 7 via the board loading track group 4. The material conveying track group 7 further transports the server motherboard 9 to the test clamping track group 8. The fixing clamps of the test clamping track group 8 accurately position and clamp the motherboard and memory modules to ensure the mechanical consistency of the insertion and removal test.

[0156] The control module sends instructions to the robotic arm 10 through the mechanical plugging and unplugging module, and the robotic arm 10 drives the plugging and unplugging part 11 to perform automatic plugging and unplugging operations.

[0157] The output module will display the contact state score SJ, fault diagnosis result, and adjusted force L. new and the adjusted speed V new The results are displayed on the screen 2 of the main test unit 1, and a test report is generated.

[0158] If the fault is false, the early warning module issues a warning. After receiving the parameter adjustment command, the control module adjusts the insertion and removal operation of the robotic arm 10 through the mechanical insertion and removal module. Specifically, the force L is adjusted. new and the adjusted speed V new This leads to the execution of the next round of testing to optimize the contact state;

[0159] After the test is completed, the test clamping track group 8 releases the server motherboard 9, and the material conveying track group 7 transfers the server motherboard 9 to the unloading track group 6, where the automatic unloading machine 5 completes the unloading, thus ending the closed loop of the entire test process.

[0160] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A server motherboard memory module insertion and removal testing system, characterized in that, The test cabinet includes a test main unit (1), which is equipped with a test clamping track group (8), a robotic arm (10) and a control module. One end of the robotic arm (10) is equipped with a plug-in part (11), and a four-wire resistance measurement sensor, a force sensor, a displacement sensor and a memory controller are integrated and installed near the plug-in part (11) of the robotic arm (10). The control module is connected to a mechanical plug-in module, a data acquisition module, an algorithm processing module, a result output module, and an early warning module; The specific implementation is as follows: The mechanical insertion and removal module is used to perform automatic insertion and removal operations of the robotic arm (10), including controlling and adjusting the insertion and removal force and speed; The data acquisition module is used to acquire insertion and removal test data detected and transmitted by the four-wire resistance measurement sensor, the force sensor, the displacement sensor, and the memory controller. The algorithm processing module includes a built-in contact state evaluation unit, a fault detection unit, and an adaptive adjustment unit. Based on the insertion and removal test data, the system calculates and determines a contact state score and a fault identification index. Based on the fault identification index, it determines the fault type and generates parameter adjustment instructions. The formula for calculating the contact state score is as follows: ； In the formula, SJ is the contact condition score, D0 is the standard contact resistance value, D is the real-time contact resistance value, w1 is the weighting coefficient, and BC... HST BC represents the historical standard deviation of the resistance, while BC represents the real-time standard deviation of the resistance. The formula for calculating the fault discrimination index is as follows: ； In the formula, GP is the fault discrimination index, L0 is the standard insertion and removal force, L is the real-time force, V0 is the standard insertion and removal speed, V is the real-time speed, WW is the bit error rate, and w2, w3, w4 and w5 are all weighting coefficients. The specific generation of parameter adjustment instructions is as follows: ； L new =L0+T×0.6; V new =V0-T×0.4; In the formula, T is the parameter adjustment amount, and GP CRI The threshold for true and false faults is w6, which is the adjustment coefficient, and L is the value of the threshold for false faults. new To adjust the intensity, V new The adjusted speed is represented by L0, which is the standard insertion / removal force, and V0, which is the standard insertion / removal speed. 0.6 and 0.4 are both adjustment coefficients. The result output module is used to display the contact status score, the fault identification result, and the new parameter information after the parameter adjustment instruction is given, and supports test report generation and storage. The early warning module is used to provide real-time early warnings for the contact status score and the fault type.

2. The server motherboard memory module insertion and removal testing system according to claim 1, characterized in that, The insertion and removal test data includes standard contact resistance value, real-time contact resistance value, standard deviation of real-time resistance fluctuation, standard deviation of historical resistance fluctuation, real-time force, real-time speed, standard insertion and removal force, standard insertion and removal speed, bit error rate, critical value for true and false faults, adjustment coefficient, and fault safety value. The four-wire resistance measurement sensor acquires real-time contact resistance values ​​in real time using the four-wire measurement method and records historical resistance data to calculate the real-time standard deviation of resistance fluctuation. The force sensor acquires the real-time force and speed during the insertion and removal process by connecting a pressure sensor and a displacement encoder in series at the execution end of the robotic arm (10). The memory controller has a built-in bit error rate acquisition unit; The bit error rate acquisition unit obtains the bit error rate by writing a preset data template to the memory module and reading and verifying it, and by calculating the ratio of the number of erroneous data bits to the total number of data bits during the read and write process.

3. The server motherboard memory module insertion and removal testing system according to claim 2, characterized in that: The specific logic for the contact status scoring is as follows: Based on the standard contact resistance value and the real-time contact resistance value, calculate the absolute resistance deviation value after multiplying the contact resistance by the weight; Based on the real-time standard deviation of the resistance fluctuation and the historical standard deviation of the resistance fluctuation, calculate the stability deviation value of the resistance fluctuation multiplied by the weight; The contact state score is determined by adding the absolute resistance deviation value and the stability deviation value.

4. The server motherboard memory module insertion and removal testing system according to claim 3, characterized in that: The fault discrimination unit is used to calculate and determine the fault discrimination index, and the specific logic is as follows: The contact status score is set as the core parameter; Combined with force deviation: Calculate the force deviation rate by multiplying the real-time force and the standard insertion / extraction force by a weight; Combined with speed deviation: Calculate the speed deviation rate by multiplying the real-time speed and the standard insertion / removal speed by a weight; Combined with the bit error rate: Calculate the correct bit rate after removing the bit error rate and multiplying it by the weight; The contact status score is multiplied by a weight and then added sequentially to the force deviation rate, the speed deviation rate, and the correct code rate to calculate and determine the fault discrimination index.

5. The server motherboard memory module insertion and removal testing system according to claim 4, characterized in that: Based on the results of the fault discrimination index, the fault type is determined as follows: If the fault discrimination index is lower than the true / false fault threshold, the fault type is determined to be a true fault, and the early warning module issues an early warning without generating the parameter adjustment instruction. If the fault discrimination index is higher than the true and false fault threshold, then the fault type is determined to be faultless and there is no need to generate the parameter adjustment instruction. If the fault discrimination index is between the true / false fault threshold and the fault safety value, then the fault type is determined to be a false fault and the parameter adjustment instruction is generated.

6. The server motherboard memory module insertion and removal test system according to claim 5, characterized in that: The adaptive adjustment unit is used to calculate and determine the parameter adjustment amount, and the specific logic is as follows: Calculate the severity of the fault: Calculate the fault risk coefficient after removing the fault discrimination index; Calculate the baseline fault risk coefficient after removing the true and false fault thresholds; Calculate the proportion of the fault risk coefficient to the baseline fault risk coefficient to determine the fault severity coefficient; Calculate the baseline adjustment amount: Calculate the adjustment coefficient multiplied by the sum of the standard insertion and extraction force and the standard insertion and extraction speed to determine the baseline parameter adjustment amount; The parameter adjustment amount is determined based on the product of the fault severity coefficient and the parameter adjustment base amount.

7. The server motherboard memory module insertion and removal testing system according to claim 6, characterized in that: The adjusted force L new and the adjusted speed V new The results are received and transmitted by the output module and the control module and then executed by the robotic arm (10).

8. The server motherboard memory module insertion and removal test system according to claim 2, characterized in that: The standard contact resistance value and the standard deviation of the historical resistance fluctuation are obtained as follows: 100 sets of motherboards and memory modules of the same model were selected, and their contact resistance and standard deviation were measured under the condition of no mechanical deviation. The average value was taken to determine the standard contact resistance value and the historical standard deviation of the resistance. The standard insertion / removal force and the standard insertion / removal speed are obtained in the following ways: The mechanical arm (10) is used to perform a force-contact resistance test on the same type of motherboard: the speed is fixed and the force is adjusted. Record the real-time force for each set of forces, and select the force with the smallest and most stable real-time force as the standard insertion and extraction force. Similar to the force-contact resistance experiment, fix the force and adjust the speed; Record the real-time speed for each set of speeds, and select the speed with the smallest real-time speed and the smallest fluctuation as the standard insertion and extraction speed; Based on the 3σ principle, the critical value for true and false faults is determined to be 0.3; The failsafe value is determined to be 0.5 based on the 5σ principle.

9. A server motherboard memory module insertion and removal testing system according to claim 7, characterized in that: The test clamping track group (8) holds the server motherboard (9), and both sides of the test clamping track group (8) are connected to the material conveying track group (7). The test main cabinet (1) is equipped with a display screen (2), the result output module is connected to the display screen (2), an automatic board loading machine (3) is installed on one side of the test main cabinet (1), and an automatic board unloading machine (5) is installed on the other side of the test main cabinet (1) away from the automatic board loading machine (3). The automatic loading machine (3) is connected to the material conveying track group (7) by a loading track group (4) for loading the server motherboard (9), and the automatic unloading machine (5) is connected to the material conveying track group (7) by a unloading track group (6) for unloading the server motherboard (9).

Images