A smart monitoring method and system for high heat dissipation chip packaging
By using pulsed power excitation and thermal stress theory in the chip packaging process, combined with multi-exponential coupling analysis, the problem of difficulty in assessing the effects of thermal shock and thermal stress in existing technologies is solved, enabling efficient and accurate quality assessment and risk identification of chip packages.
Patent Information
- Application Number
- CN202610795956.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-04
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2046-06-04
AI Technical Summary
In the current chip packaging process, monitoring methods are unable to capture the spatial non-uniformity of heat conduction inside the package, making it impossible to accurately assess the impact of thermal shock and thermal stress on chip reliability, resulting in large quality judgment errors.
By employing pulsed power excitation and temperature field sequence acquisition, combined with thermal stress theory and multi-index coupling analysis, a dynamically corrected risk index is generated. By simulating heat accumulation and transfer, equivalent thermal resistance and junction temperature are obtained, temperature difference cycling and thermal shock frequencies are identified, thermal stress risks are analyzed, and a quality failure assessment index is generated.
It enables precise assessment of internal thermal stress and heat dissipation efficiency of chip packages, accurately identifies high-risk areas, quantifies thermal fatigue damage under dynamic operating conditions, and improves the accuracy and reliability of quality assessment.
Smart Images

Figure CN122652256B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of semiconductor testing technology, specifically to an intelligent monitoring method and system for high heat dissipation chip packaging. Background Technology
[0002] High-heat-dissipation chip packaging is widely used in new energy, high-end electronic equipment and other fields. Its heat dissipation performance and structural reliability directly determine the chip's operational stability and lifespan. During the chip packaging manufacturing process, the solder layer, the chip-substrate interface and other parts are prone to becoming weak points. Even minor problems at the time of manufacture can lead to increased thermal resistance, thermal stress accumulation and crack initiation under long-term thermal cycling loads.
[0003] In the current chip packaging process, monitoring methods mostly use steady-state temperature measurement or single thermal resistance monitoring, which makes it difficult to capture the spatial non-uniformity of heat conduction inside the package. They also have low sensitivity to aging of weak parts such as chip direct connection interfaces and do not deeply analyze the impact of thermal shock and thermal stress cycling on package reliability, resulting in large errors and inaccurate judgments of chip quality level.
[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention
[0005] The purpose of this invention is to provide an intelligent monitoring method and system for high heat dissipation chip packaging, so as to solve the problems mentioned in the background art.
[0006] To achieve the above objectives, the present invention provides the following technical solution:
[0007] A smart monitoring method for high heat dissipation chip packages includes the following steps:
[0008] Step 1: Divide the package to be tested into N sub-regions, apply pulsed power excitation to the package, and simultaneously collect the time-series data of the surface temperature of each sub-region to obtain the temperature field sequence.
[0009] Step 2: Obtain the equivalent thermal resistance and equivalent junction temperature of the sub-region by simulating heat accumulation and transfer, and generate the heat dissipation efficiency index of the sub-region based on the equivalent thermal resistance.
[0010] Step 3: Use peak-valley detection algorithm to identify the periodic peak and valley values of the equivalent junction temperature, generate temperature difference cycle, obtain the time interval of the temperature difference cycle, analyze based on the time interval, and generate thermal shock frequency;
[0011] Step 4: Perform correlation analysis on the equivalent junction temperature data and the mechanical parameters of the package sub-region material. Obtain the thermal stress of the corresponding sub-region through thermal stress theory analysis. Compare and analyze the thermal stress with the critical stress threshold of the sub-region material to obtain the thermal stress risk index. Based on the thermal stress analysis process, generate the stress cycle strength.
[0012] Step 5: Analyze the stress deviation and thermal shock frequency deviation based on the stress cycle intensity and thermal shock frequency, and couple them with the thermal stress risk index to generate a dynamic correction risk index. The dynamic correction risk index is used to reflect the influence of stress amplitude and cycle frequency on the thermal stress risk index and to characterize the degree of dynamic damage acceleration.
[0013] Step 6: Analyze the heat dissipation efficiency index of each sub-region horizontally, combine it with the dynamic correction risk index for correlation analysis, obtain the quality failure assessment index of the sub-region, and generate the package quality failure assessment score based on the weighted method.
[0014] Step 7: Compare the package quality failure assessment score with the risk threshold and output the package quality level.
[0015] Furthermore, a pulse power lasting 5 seconds is applied through the test interface to collect the temperature field sequence of each sub-region in real time, with a sampling rate greater than 100Hz, covering the pulse period and the cooling stage.
[0016] Furthermore, for the k-th sub-region, it is simplified to a thermal RC circuit, and a first-order thermal dynamic equation is established. Based on ambient temperature, temperature field sequence, and pulse power data, heat accumulation and transfer in the sub-region are simulated, and the transient temperature response is quantitatively analyzed. The formula used is as follows:
[0017]
[0018] in, , , , These represent the equivalent heat capacity, equivalent junction temperature, temperature field sequence, and equivalent thermal resistance of the solder layer for the k-th sub-region, respectively. For pulse power, , Here, t represents the ambient thermal resistance and ambient temperature, respectively, and t represents the time index. The equivalent thermal resistance is used to reflect the degree of obstruction of the heat conduction path in the sub-region, and the equivalent junction temperature is used to reflect the estimated temperature distribution of the nodes inside the sub-region. The equivalent parameters of the package are obtained by recursive least squares method, and the equivalent parameters of the package include the equivalent thermal resistance and the equivalent junction temperature.
[0019] Furthermore, a correlation analysis is performed based on the equivalent thermal resistance to generate a heat dissipation efficiency index. The heat dissipation efficiency index is the reciprocal of the equivalent thermal resistance, and the formula is as follows: The heat dissipation efficiency index is used to reflect the degree of heat dissipation in a sub-region.
[0020] Furthermore, the equivalent junction temperature data is preprocessed by denoising and normalization, and the peak value of the equivalent junction temperature in the time series of the k-th sub-region is identified by the peak-valley detection algorithm. Valley value Obtain temperature difference cycle The formula used is: The temperature difference cycle is used to reflect the temperature change value between adjacent peak and valley temperature differences of the i-th equivalent junction temperature in the k-th sub-region. The time interval is used to reflect the time history of the corresponding temperature difference cycle. The average time interval is obtained after averaging. The thermal shock frequency of the k-th sub-region is generated by reciprocal calculation. The formula used is: The thermal shock frequency is used to reflect the frequency of thermal cycling in the sub-region.
[0021] Furthermore, the material mechanical parameters include Young's modulus, coefficient of thermal expansion, and ambient temperature. Correlation analysis is performed based on these material mechanical parameters and the equivalent junction temperature. The thermal stress in the k-th sub-region at time t is calculated using thermal stress theory. The formula used is: , , These are the Young's modulus and coefficient of thermal expansion of the k-th sub-region, respectively. For ambient temperature, The equivalent junction temperature of the k-th sub-region at time t is given. The critical stress threshold is used to reflect the maximum stress limit that the material in the sub-region can withstand under thermomechanical loads. It is obtained based on the material's inherent properties. The thermal stress is used to reflect the internal stress of the material within the package sub-region caused by temperature changes. The thermal stress risk index is also mentioned. This value is used to reflect the risk level of thermal stress in a sub-region. Where k represents the sub-region and t is the time series. The critical stress threshold for the k-th sub-region is determined by analyzing the magnitude of thermal stress changes caused by temperature cycling, based on the thermal stress analysis process. The formula used is: The stress cycle intensity is used to reflect the stress change experienced by the k-th sub-region during the i-th temperature difference cycle. It represents the equivalent junction temperature difference between the i-th peak and valley in the time series of the k-th sub-region.
[0022] Furthermore, the dynamically adjusted risk index The formula used to characterize the degree of dynamic damage acceleration is:
[0023]
[0024] in, , These are the dynamic correction risk index and thermal stress risk index of the k-th sub-region at time t, respectively. , These are the weighting factors for thermal shock and stress, respectively. , For standard frequency, This represents the average stress cyclic intensity of the k-th sub-region.
[0025] Furthermore, with the maximum heat dissipation efficiency index As a normalization benchmark, the heat dissipation efficiency index of each sub-region is normalized, and correlation analysis is performed in conjunction with the dynamically corrected risk index to obtain the quality failure assessment index of the k-th sub-region at time t. The formula used is:
[0026]
[0027] in As a risk adjustment factor, it is used to prevent global maxima from being masked by local factors. A weighted method is used to generate a package quality failure assessment score. The formula used is:
[0028]
[0029] in, The importance weight factor for the k-th sub-region of the encapsulation body. As an integral variable, it is used to reflect past moments on the time axis. It is a time decay function used to reflect the gradual weakening of the impact of historical risks on current risks, and is used for package quality failure assessment scores. The risk value of the package is used to reflect the overall assessment, and the test time for each package is one hour.
[0030] The present invention also provides an intelligent monitoring system for high heat dissipation chip packages, for executing an intelligent monitoring method for high heat dissipation chip packages, comprising:
[0031] The detection module divides the package to be tested into N sub-regions, applies pulsed power excitation to the package, and simultaneously collects the time-series data of the surface temperature of each sub-region to obtain the temperature field sequence.
[0032] The data analysis module obtains the equivalent thermal resistance and equivalent junction temperature of the sub-region by simulating heat accumulation and transfer, and generates the heat dissipation efficiency index of the sub-region based on the equivalent thermal resistance.
[0033] The peak-valley identification module uses a peak-valley detection algorithm to identify the periodic peak and valley values of the equivalent junction temperature, generate temperature difference cycles, obtain the time interval of the temperature difference cycles, and analyze the time interval to generate the thermal shock frequency.
[0034] The risk index analysis module performs correlation analysis on the equivalent junction temperature data and the mechanical parameters of the package sub-region material. It obtains the thermal stress of the corresponding sub-region through thermal stress theory analysis, compares it with the critical stress threshold of the sub-region material, obtains the thermal stress risk index, and generates the stress cycle strength based on the thermal stress analysis process.
[0035] The dynamic correction module analyzes the stress deviation and thermal shock frequency deviation based on stress cycle intensity and thermal shock frequency, and couples them with the thermal stress risk index to generate a dynamic correction risk index.
[0036] The output module performs a horizontal analysis of the heat dissipation efficiency index of each sub-region, combines the dynamic correction risk index for correlation analysis, obtains the quality failure assessment index of the sub-region, generates the package quality failure assessment score based on the weighted method, compares the package quality failure assessment score with the risk threshold, and outputs the package quality level.
[0037] Compared with the prior art, the beneficial effects of the present invention are:
[0038] This invention establishes thermal dynamic equations, conducts thermal stress theoretical analysis, and uses multi-index coupling to perform synergistic analysis of multiple parameters such as heat dissipation efficiency, thermal stress, and thermal shock frequency. It also performs correlation analysis on temperature difference cycles, thermal shock frequency, and stress cycle intensity to obtain accurate stress cycles, making the analysis results more accurate. After comprehensive analysis, it obtains a dynamic correction risk index and a quality failure assessment index to quantify the accelerating effect of thermal fatigue damage under dynamic operating conditions. Through regional comprehensive risk assessment and risk accumulation analysis, it can accurately locate local high-risk areas in chip packaging, obtain a quality failure assessment index reflecting the comprehensive risk of the sub-region, and a package quality failure assessment score reflecting the overall risk of the package, thereby enabling accurate testing and quality level assessment of the package. Attached Figure Description
[0039] Figure 1 This is a schematic diagram of the overall method flow of the present invention;
[0040] Figure 2 This is a schematic diagram of the system modules of the present invention. Detailed Implementation
[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.
[0042] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.
[0043] Example:
[0044] Please see Figure 1 The present invention provides a technical solution:
[0045] A smart monitoring method for high heat dissipation chip packages includes the following steps:
[0046] S1. Divide the package to be tested into N sub-regions, apply continuous pulse power excitation to the package, and synchronously collect the time series data of the surface temperature of each sub-region to obtain the temperature field sequence; divide the sub-regions based on the package structure and thermal conductivity characteristics.
[0047] A 5-second pulse power is applied via the test interface to collect the temperature field sequence of each sub-region in real time, with a sampling rate greater than 100Hz, covering the pulse period and the cooling phase. The package is divided into multiple small regions. Heat conduction within the package is typically spatially non-uniform, and different regions have different materials, structures, and heat dissipation. Dividing the package into sub-regions makes the thermal field measurement more accurate. The pulse power excitation generates a transient thermal pulse inside the package material, and heat diffuses from the heat source to the surroundings, forming a temporal temperature change. Compared with steady-state temperature measurement, pulse excitation increases the amount of thermal response information and improves the ability to identify the thermal properties of the material. The acquisition process uses non-contact infrared temperature acquisition.
[0048] S2. Obtain the equivalent thermal resistance and equivalent junction temperature of the sub-region by simulating heat accumulation and transfer, and generate the heat dissipation efficiency index of the sub-region based on the equivalent thermal resistance.
[0049] For the k-th sub-region, it is simplified to a thermal RC circuit. A first-order thermal dynamic equation is established. Based on ambient temperature, temperature field sequence, and pulse power data, heat accumulation and transfer in the sub-region are simulated, and the transient temperature response is quantitatively analyzed. The formula used is as follows:
[0050]
[0051] in, , , , These represent the equivalent heat capacity, equivalent junction temperature, temperature field sequence, and equivalent thermal resistance of the solder layer for the k-th sub-region, respectively. For pulse power, , The equivalent thermal resistance and ambient temperature are respectively represented by t, where t is a time index. The equivalent thermal resistance reflects the degree of obstruction in the heat conduction path of the sub-region. Ambient thermal resistance refers to the thermal impedance between the surface of the package and its surrounding environment, reflecting the degree of obstruction encountered by heat dissipation from the surface of the package to the environment. It is determined by a combination of steady-state experimental and computational methods. A known steady-state power is applied to the package to bring its temperature to a steady state, thereby obtaining the relevant temperature and calculating the equivalent ambient thermal resistance, and taking the average value. The equivalent junction temperature reflects the estimated temperature distribution of the nodes inside the sub-region. The equivalent parameters of the package are obtained by recursive least squares method, including the equivalent thermal resistance and the equivalent junction temperature. The heat conduction process inside the package is analogized to a dynamic system coupled with resistance and capacitance. The pulse power serves as the heat input, driving the temperature change. Thermal resistance limits the heat flow path, and thermal capacitance determines the time lag of the temperature response. The rate of heat accumulation corresponds to the rate of change of the junction temperature in region k, reflecting the response of the region's heat capacity to temperature changes. It directly serves as a heat energy input, driving up the temperature and is the main driving force behind temperature changes. This represents the heat flow rate through the thermal resistance of the solder layer from the junction temperature to the surface temperature. Its magnitude reflects the rate at which heat energy is transferred from the junction to the surface. This indicates the rate at which heat is dissipated from the surface into the environment across the thermal resistance of the environment. The higher the power, the more heat is transferred, the faster the junction temperature rises, the greater the thermal resistance, the more heat conduction is hindered, the faster the junction temperature accumulates, the greater the temperature response lag, the difference between the surface temperature and the junction temperature promotes heat dissipation, reduces the junction temperature rise rate, and the smaller the ambient thermal resistance, the faster the heat dissipation. Online parameter estimation is performed using the recursive least squares method to dynamically estimate the equivalent parameters of the package.
[0052] Based on the analysis of equivalent thermal resistance, a heat dissipation efficiency index is generated. The heat dissipation efficiency index is the reciprocal of the equivalent thermal resistance, and the formula is as follows: The heat dissipation efficiency index is used to reflect the degree of heat dissipation in a sub-region. The higher the heat dissipation efficiency index, the lower the thermal resistance, the higher the heat conduction efficiency, and the better the heat dissipation performance. It is used to evaluate the heat dissipation quality of a sub-region and to detect heat dissipation reduction or abnormalities in a timely manner. The heat dissipation efficiency index is dynamically adjusted with changes in thermal resistance. When the thermal resistance decreases, the heat dissipation efficiency index increases; when the thermal resistance increases, the heat dissipation efficiency index decreases.
[0053] S3. The peak and valley detection algorithm is used to identify the periodic peak and valley values of the equivalent junction temperature, generate temperature difference cycle, obtain the time interval of the temperature difference cycle, and analyze based on the time interval to generate the thermal shock frequency.
[0054] The equivalent junction temperature data undergoes denoising and normalization preprocessing to filter out noise and abnormal fluctuations in the original equivalent junction temperature data, eliminate differences in data magnitude, and ensure the accuracy of peak and valley detection. The peak and valley detection algorithm then identifies the ith equivalent junction temperature peak value in the time series of the k-th sub-region. Valley value Obtain temperature difference cycle The formula used to quantify the peak-valley temperature variation amplitude of the i-th temperature cycle in the k-th sub-region is as follows: The temperature difference cycle is used to reflect the temperature change value between adjacent peak and valley temperature differences of the i-th equivalent junction temperature in the k-th sub-region. The time interval is used to reflect the time history of the corresponding temperature difference cycle. The average time interval is obtained after averaging. The thermal shock frequency of the k-th sub-region is generated by reciprocal calculation. The formula used is: The thermal shock frequency is the frequency of thermal cycling in the sub-region.
[0055] S4. Correlation analysis is performed on the equivalent junction temperature data and the mechanical parameters of the material in the package sub-region. The thermal stress of the corresponding sub-region is obtained through thermal stress theory analysis. The thermal stress is compared with the critical stress threshold of the material in the sub-region to obtain the thermal stress risk index. Based on the thermal stress analysis process, the stress cycle strength is generated. The equivalent junction temperature is combined with the mechanical properties of the material in the sub-region. Temperature changes cause thermal expansion or contraction of the material, and material stiffness affects the stress distribution. The combination of the two can accurately assess the thermal stress level.
[0056] The material mechanical parameters include Young's modulus, coefficient of thermal expansion, and ambient temperature. Correlation analysis is performed based on these material mechanical parameters and the equivalent junction temperature. The thermal stress in the k-th sub-region at time t is calculated using thermal stress theory. To solve for the stress field caused by thermal expansion, the stress generated by thermal expansion driven by temperature difference is based on the following formula: Thermal stress is proportional to the material stiffness, coefficient of thermal expansion, and temperature difference. , These are the Young's modulus and coefficient of thermal expansion of the k-th sub-region, respectively. For ambient temperature, The equivalent junction temperature of the k-th sub-region at time t is given. The critical stress threshold reflects the maximum stress limit that the material in the sub-region can withstand under thermomechanical loads. It is obtained based on the material's inherent properties, measured through experiments such as tensile, compression, three-point bending, and fatigue tests to determine the material's yield strength, ultimate strength, and fatigue limit. The package undergoes extensive thermal cycling loads, and the fatigue limit directly affects its lifespan and reliability. As the critical stress threshold, thermal stress reflects the internal stress generated by temperature changes within the material of the package sub-region. The thermal stress risk index is also relevant. This value is used to reflect the risk level of thermal stress in a sub-region. The actual stress is compared with the maximum allowable stress of the material to quantify the risk level caused by thermal stress. A thermal stress risk index greater than 1 indicates that the thermal stress exceeds the critical threshold and the risk is greater; conversely, a value closer to 1 or lower indicates a lower risk. Here, k is used to index the sub-region, and t is the time series. The critical stress threshold for the k-th sub-region is determined by analyzing the magnitude of thermal stress changes caused by temperature cycling, based on the thermal stress analysis process. The formula used is: Based on the temperature change caused by thermal cycling, the corresponding stress change amplitude is calculated to reflect the fatigue loading strength of the material under thermal cycling. The stress cycling strength is used to reflect the stress change experienced by the k-th sub-region in the i-th thermal cycle. The equivalent junction temperature difference of the i-th peak and valley in the time series of the k-th sub-region is given. The stress cycle intensity is used for thermal fatigue damage analysis and life prediction. The stress cycle intensity is proportional to the material stiffness, coefficient of thermal expansion and temperature difference cycle amplitude.
[0057] S5. Based on the stress cycle intensity and thermal shock frequency, analyze the stress deviation and thermal shock frequency deviation, and couple them with the thermal stress risk index to generate a dynamic correction risk index. The dynamic correction risk index is used to reflect the influence of stress amplitude and cycle frequency on the thermal stress risk index and to characterize the degree of dynamic damage acceleration.
[0058] The dynamically adjusted risk index The formula used to characterize the degree of dynamic damage acceleration is:
[0059]
[0060] in, , These are the dynamic correction risk index and thermal stress risk index of the k-th sub-region at time t, respectively. , These are the weighting factors for thermal shock and stress, respectively. The weighting of thermal shock frequency reflects the influence of thermal cycling frequency on thermal stress risk, while the weighting of stress cycling intensity reflects the influence of stress amplitude on thermal stress risk. This is based on fitting material fatigue experimental data, specifically using accelerated fatigue tests to obtain fatigue life data of materials under different frequencies and stress amplitudes. The weights are determined by normalizing the contribution; a larger contribution corresponds to a higher weight. The standard frequency, defined experimentally, is a reference thermal shock frequency used to demonstrate the thermal cycling characteristics of materials or devices under standard operating conditions. It helps quantify the impact of frequency deviations on risk. The standard frequency is selected based on specific thermal cycling frequency ranges and loading rates specified in electronic device thermal cycling test standards, combined with the application field of the package. Alternatively, it can be determined based on accelerated life testing. Let be the average stress cycle intensity of the k-th sub-region. Thermal fatigue damage is not only related to the magnitude of stress, but also closely related to the frequency of thermal cycling; higher cycle frequencies and larger stress amplitudes will accelerate the damage.
[0061] The thermal stress risk index is weighted and superimposed with frequency and stress deviation to generate a dynamic corrected risk index, which reflects the accelerated damage risk under dynamic working conditions. The dynamic corrected risk index can more accurately reflect the true risk level under different thermal cycling frequencies and stress amplitudes. The higher the value of the dynamic corrected risk index, the more severe the risk. It dynamically reflects the accelerated trend of thermomechanical fatigue damage. When the actual frequency is lower than the standard frequency, the long-term thermal load and lower thermal shock frequency usually give the material a longer stress relaxation time, thereby accelerating the accumulation of thermal fatigue and crack propagation. The promoting effect of low-frequency cycling on damage is stronger than that of high-frequency cycling, and the dynamic risk increases. The greater the stress cycle intensity, the greater the dynamic risk. As the thermal stress risk index increases, the dynamic risk also increases. The weighting factor adjusts the contribution of the two influencing factors in the risk index.
[0062] S6. Analyze the heat dissipation efficiency index of each sub-region horizontally, combine the dynamic correction risk index to conduct correlation analysis, obtain the quality failure assessment index of the sub-region, and generate the package quality failure assessment score based on the weighted method.
[0063] With maximum heat dissipation efficiency index As a normalization benchmark, the heat dissipation efficiency index of each sub-region is normalized, and correlation analysis is performed in conjunction with the dynamically corrected risk index to obtain the quality failure assessment index of the k-th sub-region at time t. Individual heat dissipation performance or thermal stress risk may easily overlook important hidden dangers. Comprehensive consideration can more fully reflect the overall quality of a sub-region, making it easier to identify areas with potential faults or failure zones. The formula used is:
[0064]
[0065] Among them, the first item The main fusion item, the second item This is a risk adjustment item to prevent danger zones from being masked by averages. This is a risk adjustment factor used to prevent global maxima from being masked by local factors. It is set based on historical engineering experience and test data, and can be initially set between 0.1 and 0.5. If there is extreme sensitivity to local risks, the value can be appropriately increased. The quality failure assessment index quantitatively reflects the thermal management capability of this sub-region; a higher value indicates higher risk and worse quality, facilitating accurate location of high-risk areas in the package. When heat dissipation efficiency is poor, the package quality failure assessment score increases. The larger the value, the higher the overall weight of high-risk areas, enhancing the ability to identify extreme risks. A weighted method is used to generate a package quality failure assessment score. The formula used is:
[0066]
[0067] Based on the quality failure assessment index of each sub-region, and according to the importance weight of each sub-region combined with historical time-series information, the package quality failure assessment score is obtained by integrating through a time decay function. The importance weight factor for the k-th sub-region of the encapsulation body. These are integral variables and historical time variables, used to reflect past moments on the timeline. It is a time decay function. , This is the decay rate constant, which determines the rate at which the impact of historical risks decays. A larger value indicates that the impact of historical risks weakens rapidly, while a smaller value indicates that the impact of historical risks persists for a long time. It is the independent variable of the integral, used to iterate through all time points from the past time 0 to the current time t, by... Integration is performed to accumulate risk data from all historical moments according to weights, reflecting the impact of historical risks on current risks. A time decay function is used to show that the impact of historical risks on current risks gradually weakens. This results in a package quality failure assessment score. The overall quality of the package is used to reflect the comprehensive evaluation, and to reflect the importance, comprehensive risk and historical risk accumulation of each sub-region. The greater the importance weight, the greater the contribution to the overall risk. The greater the comprehensive risk of the sub-region, the higher the overall risk. The higher the package quality failure assessment score, the worse the package quality. Historical risks have a greater impact in the near future and a smaller impact in the long term. The test time for each package is one hour.
[0068] S7. Compare the one-hour test package quality failure assessment score with the risk threshold. , The comparison is performed, and the package quality level and risk threshold are output. , Determined jointly by thermal cycling accelerated life testing and material mechanical limits, when the test package quality failure assessment score is less than the risk threshold... The output is "quality qualified" when the test package quality failure assessment score is within the risk threshold. Risk threshold Between these thresholds, the output is considered medium risk, and the package quality requires further manual inspection. If the package quality failure assessment score exceeds the risk threshold, the risk is considered medium. A high-risk output indicates that the package is of substandard quality and needs to be reworked or discarded. A low-risk output indicates that the package is in good condition and of acceptable quality. A medium-risk output indicates that the package may have abnormalities, such as signs of temperature rise or thermal stress accumulation, requiring continuous monitoring and further manual inspection. A high-risk output indicates that the package is of substandard quality and needs to be reworked or discarded.
[0069] Reference Figure 2 The present invention also provides an intelligent monitoring system for high heat dissipation chip packaging, for executing an intelligent monitoring method for high heat dissipation chip packaging, comprising:
[0070] The detection module divides the package to be tested into N sub-regions, applies pulsed power excitation to the package, and simultaneously collects the time-series data of the surface temperature of each sub-region to obtain the temperature field sequence.
[0071] The data analysis module obtains the equivalent thermal resistance and equivalent junction temperature of the sub-region by simulating heat accumulation and transfer, and generates the heat dissipation efficiency index of the sub-region based on the equivalent thermal resistance.
[0072] The peak-valley identification module uses a peak-valley detection algorithm to identify the periodic peak and valley values of the equivalent junction temperature, generate temperature difference cycles, obtain the time interval of the temperature difference cycles, and analyze the time interval to generate the thermal shock frequency.
[0073] The risk index analysis module performs correlation analysis on the equivalent junction temperature data and the mechanical parameters of the package sub-region material. It obtains the thermal stress of the corresponding sub-region through thermal stress theory analysis, compares it with the critical stress threshold of the sub-region material, obtains the thermal stress risk index, and generates the stress cycle strength based on the thermal stress analysis process.
[0074] The dynamic correction module analyzes the stress deviation and thermal shock frequency deviation based on stress cycle intensity and thermal shock frequency, and couples them with the thermal stress risk index to generate a dynamic correction risk index.
[0075] The output module performs a horizontal analysis of the heat dissipation efficiency index of each sub-region, combines the dynamic correction risk index for correlation analysis, obtains the quality failure assessment index of the sub-region, generates the package quality failure assessment score based on the weighted method, compares the package quality failure assessment score with the risk threshold, and outputs the package quality level.
[0076] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.
[0077] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.
[0078] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.
[0079] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.
Claims
1. A smart monitoring method for high heat dissipation chip packaging, characterized in that, The specific steps include: S1. Divide the package to be tested into N sub-regions, apply pulse power excitation to the package, and synchronously collect the surface temperature time series data of each sub-region to obtain the temperature field sequence. S2. Obtain the equivalent thermal resistance and equivalent junction temperature of the sub-region by simulating the heat accumulation and transfer process, and generate the heat dissipation efficiency index of the sub-region based on the equivalent thermal resistance. S3. The peak and valley detection algorithm is used to identify the periodic peak and valley values of the equivalent junction temperature, generate temperature difference cycle, obtain the time interval of the temperature difference cycle, and analyze based on the time interval to generate the thermal shock frequency. S4. Perform correlation analysis on the equivalent junction temperature data and the mechanical parameters of the material in the sub-region of the package. Obtain the thermal stress of the corresponding sub-region through thermal stress theory analysis. Compare and analyze the thermal stress with the critical stress threshold of the sub-region material to obtain the thermal stress risk index. Based on the thermal stress analysis process, generate the stress cycle strength. S5. Based on the stress cycle intensity and thermal shock frequency, analyze the stress deviation and thermal shock frequency deviation, and couple them with the thermal stress risk index to generate a dynamic correction risk index. The dynamic correction risk index is used to reflect the influence of stress amplitude and cycle frequency on the thermal stress risk index and to characterize the degree of dynamic damage acceleration. S6. Analyze the heat dissipation efficiency index of each sub-region, combine it with the dynamic correction risk index to perform correlation analysis, obtain the quality failure assessment index of the sub-region, and generate the package quality failure assessment score based on the weighted method. S7. Compare the package quality failure assessment score with the risk threshold and output the package quality level.
2. The intelligent monitoring method for high heat dissipation chip packaging according to claim 1, characterized in that: A pulse power lasting 5 seconds is applied through the test interface to collect the temperature field sequence of each sub-region in real time.
3. The intelligent monitoring method for high heat dissipation chip packaging according to claim 2, characterized in that: For the k-th sub-region, it is simplified to a thermal RC circuit. A first-order thermal dynamic equation is established. Based on ambient temperature, temperature field sequence, and pulse power data, heat accumulation and transfer in the sub-region are simulated, and the transient temperature response is quantitatively analyzed. The formula used is as follows: in, , , , These represent the equivalent heat capacity, equivalent junction temperature, temperature field sequence, and equivalent thermal resistance of the solder layer for the k-th sub-region, respectively. For pulse power, , Here, t represents the ambient thermal resistance and ambient temperature, respectively, and t represents the time index. The equivalent thermal resistance is used to reflect the degree of obstruction of the heat conduction path in the sub-region, and the equivalent junction temperature is used to reflect the estimated temperature distribution of the nodes inside the sub-region. The equivalent parameters of the package are obtained by recursive least squares method, and the equivalent parameters of the package include the equivalent thermal resistance and the equivalent junction temperature.
4. The intelligent monitoring method for high heat dissipation chip packaging according to claim 3, characterized in that: Correlation analysis was performed based on equivalent thermal resistance to generate a heat dissipation efficiency index. The heat dissipation efficiency index is the reciprocal of the equivalent thermal resistance, and the formula used is: The heat dissipation efficiency index is used to reflect the degree of heat dissipation in a sub-region.
5. The intelligent monitoring method for high heat dissipation chip packaging according to claim 4, characterized in that: The equivalent junction temperature data is preprocessed by denoising and normalization. The peak value of the equivalent junction temperature in the time series of the k-th sub-region is identified by the peak-valley detection algorithm. Valley value Obtain temperature difference cycle The formula used is: The temperature difference cycle is used to reflect the temperature change value between adjacent peak and valley temperature differences of the i-th equivalent junction temperature in the k-th sub-region. The time interval is used to reflect the time history of the corresponding temperature difference cycle. The average time interval is obtained after averaging. The thermal shock frequency of the k-th sub-region is generated by reciprocal calculation. The formula used is: The thermal shock frequency is used to reflect the frequency of thermal cycling in the sub-region.
6. The intelligent monitoring method for high heat dissipation chip packaging according to claim 5, characterized in that: The material mechanical parameters include Young's modulus, coefficient of thermal expansion, and ambient temperature. Correlation analysis is performed based on these material mechanical parameters and the equivalent junction temperature. The thermal stress in the k-th sub-region at time t is calculated using thermal stress theory. The formula used is: , , These are the Young's modulus and coefficient of thermal expansion of the k-th sub-region, respectively. For ambient temperature, The equivalent junction temperature of the k-th sub-region at time t is given. The critical stress threshold is used to reflect the maximum stress limit that the material in the sub-region can withstand under thermomechanical loads. It is obtained based on the material's inherent properties. The thermal stress is used to reflect the internal stress of the material within the package sub-region caused by temperature changes. The thermal stress risk index is also mentioned. This value is used to reflect the risk level of thermal stress in a sub-region. Where k represents the sub-region and t is the time series. The critical stress threshold for the k-th sub-region is determined by analyzing the magnitude of thermal stress changes caused by temperature cycling, based on the thermal stress analysis process. The formula used is: The stress cycle intensity is used to reflect the stress change experienced by the k-th sub-region during the i-th temperature difference cycle. It represents the equivalent junction temperature difference between the i-th peak and valley in the time series of the k-th sub-region.
7. The intelligent monitoring method for high heat dissipation chip packaging according to claim 6, characterized in that: The dynamically adjusted risk index The formula used to characterize the degree of dynamic damage acceleration is: in, , These are the dynamic correction risk index and thermal stress risk index of the k-th sub-region at time t, respectively. , These are the weighting factors for thermal shock and stress, respectively. , For standard frequency, This represents the average stress cyclic intensity of the k-th sub-region.
8. The intelligent monitoring method for high heat dissipation chip packaging according to claim 7, characterized in that: With maximum heat dissipation efficiency index As a normalization benchmark, the heat dissipation efficiency index of each sub-region is normalized, and correlation analysis is performed in conjunction with the dynamically corrected risk index to obtain the quality failure assessment index of the k-th sub-region at time t. The formula used is: in As a risk adjustment factor, it is used to prevent global maxima from being masked by local factors. A weighted method is used to generate a package quality failure assessment score. The formula used is: in, This is the importance weighting factor for the k-th sub-region of the encapsulation. As an integral variable, it is used to reflect past moments on the time axis. It is a time decay function used to reflect the gradual weakening of the impact of historical risks on current risks, and is used for package quality failure assessment scores. The risk value of the package is used to reflect the overall assessment, and the test time for each package is one hour.
9. An intelligent monitoring system for high heat dissipation chip packaging, used to execute the intelligent monitoring method for high heat dissipation chip packaging as described in claim 1, characterized in that, include: The detection module divides the package to be tested into N sub-regions, applies pulsed power excitation to the package, and simultaneously collects the time-series data of the surface temperature of each sub-region to obtain the temperature field sequence. The data analysis module obtains the equivalent thermal resistance and equivalent junction temperature of the sub-region by simulating heat accumulation and transfer, and generates the heat dissipation efficiency index of the sub-region based on the equivalent thermal resistance. The peak-valley identification module uses a peak-valley detection algorithm to identify the periodic peak and valley values of the equivalent junction temperature, generate temperature difference cycles, obtain the time interval of the temperature difference cycles, and analyze the time interval to generate the thermal shock frequency. The risk index analysis module performs correlation analysis on the equivalent junction temperature data and the mechanical parameters of the package sub-region material. It obtains the thermal stress of the corresponding sub-region through thermal stress theory analysis, compares it with the critical stress threshold of the sub-region material, obtains the thermal stress risk index, and generates the stress cycle strength based on the thermal stress analysis process. The dynamic correction module analyzes the stress deviation and thermal shock frequency deviation based on stress cycle intensity and thermal shock frequency, and couples them with the thermal stress risk index to generate a dynamic correction risk index. The output module performs a horizontal analysis of the heat dissipation efficiency index of each sub-region, combines the dynamic correction risk index for correlation analysis, obtains the quality failure assessment index of the sub-region, generates the package quality failure assessment score based on the weighted method, compares the package quality failure assessment score with the risk threshold, and outputs the package quality level.
Citation Information
Patent Citations
Integrated circuit packaging quality evaluation method and system based on nondestructive testing technology
CN119199485A
Dynamic method and system for thermal fatigue evaluation of integrated circuit package
CN119358359A
