Electric-thermal-magnetic-force coupling field finite element simulation life prediction method for board-level electronic packaging device

Abstract

The invention provides an electro-thermal-magnetic-force coupling field finite element simulation life prediction method for a board-level electronic packaging device, relates to the field of electronic packaging reliability analysis and finite element simulation, and aims to solve the problems that life prediction depends on single / partial field coupling, neglects the influence of electromagnetic force and magnetic field, lacks a multi-field transmission mechanism, and cannot predict the life of the board-level electronic packaging device. Therefore, the problems of large prediction deviation and incapability of realizing full-field coupling are solved. According to the method, a real welding spot model is firstly built and repaired, a board-level packaging three-dimensional structure subdivision grid is constructed, electromagnetic force and additional heat are obtained through electro-magnetic coupling, heat-force coupling loading is input to calculate heat stress, at least two fatigue models are substituted, and the service life of a key welding spot is calculated. Electro-thermal-magnetic-force full-coupling simulation is achieved, life prediction precision is improved through multi-model fusion, multi-package and multi-load adaptation is achieved, a reliable basis is provided for electronic package design, the test period is shortened, the experiment cost is reduced, and development of package reliability research towards multi-field life prediction is promoted.

Description

Technical Field

[0001] This invention relates to the field of electronic packaging reliability analysis and finite element simulation technology, and in particular to a finite element simulation method for predicting the lifetime of board-level electronic packaging devices using electro-thermal-magnetic-force coupling fields. Background Technology

[0002] As electronic packaging advances towards higher integration and power density, devices simultaneously endure complex coupling effects from electric, magnetic, thermal, and force fields during service. Under high-frequency and high-current conditions, the Lorentz force and eddy current effect induced by electromagnetic fields are significantly enhanced, easily leading to localized stress concentration and fatigue damage at solder joints. Furthermore, mismatches in the thermal expansion coefficients of different materials can cause thermal stress and structural warping, while vibration and mechanical shock further accelerate interface degradation and crack propagation.

[0003] Existing technologies have achieved multi-field coupled simulation of electromagnetism, thermodynamics, and vibration in board-level packaging, which can obtain stress distribution results. However, they only use a single three-interval model for single-point lifetime prediction, which has the following drawbacks: ① It cannot adapt to thermo-magnetic-vibration composite loads; ② It does not consider the prediction differences of different fatigue models, resulting in insufficient accuracy; ③ It does not associate with a material parameter library, making it difficult to extend to different packaging types.

[0004] Therefore, there is an urgent need for a life prediction method that integrates multiple models and adapts to multiple loads to fill the gap in the engineering evaluation system. Summary of the Invention

[0005] This invention proposes a finite element simulation method for lifetime prediction of board-level electronic packaging devices using electro-thermal-magnetic-force coupling fields. By setting up a magnetic-electric coupling field analysis of the component before performing the overall thermal-mechanical analysis, the force and thermal effects of the component under the action of electro-magnetic coupling are obtained. Subsequently, the calculated temperature distribution and stress distribution results are applied as preset fields in the subsequent thermal-mechanical coupling structural analysis to achieve full electro-magnetic-thermal-force coupling analysis. This solves the problems in the prior art where lifetime prediction is based only on a single physical field (thermal field or force field) or partial field (thermal-electric-force) coupling, only considers the current thermal effect while ignoring the electromagnetic force effect and magnetic field influence, and lacks a multi-field transfer mechanism from the electro-magnetic field analysis results to the thermal-mechanical analysis, resulting in a large deviation between the predicted results and the actual service life, and the inability to achieve full-field coupling analysis.

[0006] A finite element simulation method for lifetime prediction of electro-thermal-magnetic-mechanical coupling fields of board-level electronic packaging devices includes the following steps: S1. Based on the actual structure of the board-level electronic packaging component, the solder joint process parameters are adjusted using a morphology simulation algorithm. Gravitational potential energy, surface energy, and interfacial tension are iteratively calculated according to the principle of minimum free energy to obtain a real solder joint morphology model. The closed defects on the surface of the solder joint in the real solder joint morphology model are repaired and then solidified. S2. Construct a three-dimensional structure model of the board-level package based on the actual device of the board-level electronic packaging component, and perform high-quality mesh generation on the three-dimensional structure model; S3. Apply current boundary conditions to the three-dimensional structural model to form a conductive loop, and obtain the electromagnetic force distribution and the additional heat field caused by electromagnetic effects by solving the electromagnetic field equations. S4. Input the additional heat field and electromagnetic force distribution obtained in S3 into the three-dimensional structural model as the initial field, apply thermal cycling, thermal storage or mechanical vibration loads to the three-dimensional structural model, and calculate the thermal stress and deformation response. S5. Substitute the thermal stress and strain amplitude values ​​obtained in S4 into at least two fatigue life models selected from the Basquin model, Coffin-Manson model, and Engelmaier model, and calculate the life distribution of key weld points using a weighted fusion strategy.

[0007] Furthermore, in S1, the surface tension, solid-liquid interface energy, and gravitational potential energy parameters of the liquid solder are adjusted according to the principle of minimum liquid phase interface energy to constrain the dynamic morphological changes of the liquid phase during reflow soldering. The process is iterated until the solder joint surface reaches a stable state, thereby obtaining a true solder joint morphology model. The closed defect repair adopts interface reconstruction and topology repair algorithms.

[0008] Furthermore, in S2, the element quality coefficient of the high-quality mesh is ≥0.11, and the mesh is a tetrahedral mesh.

[0009] Furthermore, in S3, when solving the electro-magnetic field equations, based on Ampere's circuital law and the Biot-Savart law, the parameters of the current boundary conditions are: peak current 10A, frequency 100MHz. In the three-dimensional structural model, the conductor material is Cu, the solder joint material is Sn63Pb37 eutectic solder, and the structural area force of the electromagnetic force distribution is 10. 2 -10 3 N / m 2 And it is concentrated in the current source and the bending area of ​​the wire.

[0010] Furthermore, in S4, the thermal cycling load is set according to the IPC-TM650-2672 standard, with the following parameters: upper temperature limit 125℃, lower temperature limit -55℃, heating rate 3℃ / s, and isothermal holding time 5min. In the thermo-mechanical coupled field analysis, the current is kept on to calculate the Joule heating effect, thereby realizing full-field coupling of electricity, heat, magnetism, and force.

[0011] Furthermore, in S5, the calculation formula for the Basquin model is as follows:

[0012] Where, σ a For stress amplitude, The fatigue strength coefficient, denoted as fatigue life, and b as the fatigue strength index.

[0013] Furthermore, in S5, the Coffin-Manson lifetime prediction formula is:

[0014] Where Δγ is the equivalent non-shear strain range; Δε is the equivalent plastic strain amplitude, Δγ=√3Δε. denoted as the fatigue elongation coefficient, c as the fatigue ductility index, and the equivalent plastic strain amplitude is related to the temperature cycle amplitude ΔT, the weld point geometric parameters, and the difference in the material's thermal expansion coefficient.

[0015] Furthermore, in S5, the calculation formula for the Engelmaier model is as follows:

[0016] Where Δγ is the shear strain range.

[0017] A storage medium storing a computer program, which, when executed by a processor, implements the above-described method for predicting the lifetime of board-level electronic packaging devices using finite element simulation of electro-thermal-magnetic-force coupling fields.

[0018] A computer device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-described method for predicting the lifetime of board-level electronic packaging devices using finite element simulation of electro-thermal-magnetic-force coupling fields.

[0019] Compared with the prior art, the present invention achieves significant beneficial effects through the above technical solution: (1) Theoretical level: This invention realizes for the first time the full coupling simulation of multiple physical fields such as electricity, heat, magnetism and force, and establishes a lifetime prediction module on this basis, forming a full-link system from morphology modeling to field response calculation to lifetime assessment, filling the gap in the field; (2) Engineering level: By adapting to various packaging structures (BGA, CGA, QFP, SOP, DIP, etc.) and different load environments (thermal cycling, electromagnetic vibration, electrical fatigue, etc.), a wider range of lifetime prediction applications have been achieved; (3) Improved accuracy and efficiency: The multi-field result mapping mechanism significantly reduces the prediction error, and the error between lifetime calculation and experimental results is controlled within 10%; (4) High engineering promotion value: It can provide electronic packaging designers with a basis for reliability assessment, shorten the testing cycle and reduce experimental costs; (5) Technological evolution trend: This invention promotes the transition from "multi-field stress analysis" to "multi-field lifetime prediction", representing the new technology development direction of electronic packaging reliability research. Attached Figure Description

[0020] Figure 1 The image shows the simulation results of the solder joint morphology. Figure 1 (a) is a simulation result of the BGA solder joint morphology; Figure 1 (b) is a simulation result of the DIP solder joint morphology; Figure 1 (c) shows the simulation results of the pin solder joint morphology for QFP and SOP; Figure 2 A 3D model of a printed circuit board assembly; Figure 3 To create a current loop model diagram in electromagnetic simulation software; Figure 4 The diagram shows the stress distribution and plastic deformation results from the thermal cycling simulation. Figure 4 (a) represents the maximum stress value of the solder joint of the CGA device; Figure 4 (b) represents the maximum stress value of the solder joint of the QFP device; Figure 4 (c) represents the maximum stress value of the solder joint of the SOP device; Figure 4 (d) represents the maximum stress value of the solder joint of the DIP device; Figure 5 A comparison chart of multi-model fusion lifetime prediction results and experimental results; Figure 6 The flowchart illustrates a finite element simulation method for predicting the lifetime of a board-level electronic packaging device using electro-thermal-magnetic-force coupling fields. Detailed Implementation

[0021] 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.

[0022] Reference Figure 6 As shown, this invention discloses a finite element simulation method for lifetime prediction of board-level electronic packaging devices using electro-thermal-magnetic-mechanical coupling fields, comprising the following steps: S1. Based on the actual structure of the board-level electronic packaging component, the solder joint process parameters are adjusted using a morphology simulation algorithm. Gravitational potential energy, surface energy, and interfacial tension are iteratively calculated according to the principle of minimum free energy to obtain a real solder joint morphology model. The closed defects on the surface of the solder joint in the real solder joint morphology model are repaired and then solidified. S2. Construct a three-dimensional structure model of the board-level package based on the actual device of the board-level electronic packaging component, and perform high-quality mesh generation on the three-dimensional structure model; S3. Apply current boundary conditions to the three-dimensional structural model to form a conductive loop, and obtain the electromagnetic force distribution and the additional heat field caused by electromagnetic effects by solving the electromagnetic field equations. S4. Input the additional heat field and electromagnetic force distribution obtained in S3 into the three-dimensional structural model as the initial field, apply thermal cycling, thermal storage or mechanical vibration loads to the three-dimensional structural model, and calculate the thermal stress and deformation response. S5. Substitute the thermal stress and strain amplitude values ​​obtained in S4 into at least two fatigue life models selected from the Basquin model, Coffin-Manson model, and Engelmaier model, and calculate the life distribution of key weld points using a weighted fusion strategy.

[0023] Specifically, this invention constructs and repairs a realistic solder joint morphology model based on the principle of minimum free energy, and combines it with high-quality mesh-generated 3D structural modeling, laying a precise geometric foundation for multi-physics coupling simulation and significantly improving the accuracy of local stress distribution description. By first performing electro-magnetic coupling field analysis to obtain the electromagnetic force distribution and additional heat field, and then using these as initial field inputs for thermo-mechanical coupling analysis and superimposing corresponding loads, it achieves full-field coupling simulation of electro-thermal-magnetic-force, completely restoring the real scenario of multi-field synergy during device service, filling the gap in incomplete multi-physics coupling in existing technologies. By substituting thermal stress and strain amplitudes into at least two fatigue life models and combining them with a weighted fusion strategy to calculate the life distribution of key solder joints, the adaptability advantages of different models are fully utilized, effectively reducing the deviation between the predicted results and the actual service life. At the same time, it is adaptable to various packaging structures and load environments, forming a complete link system from topography modeling to field response calculation to life assessment. This not only provides electronic packaging designers with reliable reliability assessment basis, shortens the testing cycle and reduces experimental costs, but also promotes the transition of electronic packaging reliability research from "multi-field stress analysis" to "multi-field life prediction", which has broad engineering application value and technological evolution significance.

[0024] Furthermore, in S1, the surface tension, solid-liquid interface energy, and gravitational potential energy parameters of the liquid solder are adjusted according to the principle of minimum liquid phase interface energy to constrain the dynamic morphological changes of the liquid phase during reflow soldering. The process is iterated until the solder joint surface reaches a stable state, thereby obtaining a true solder joint morphology model. The closed defect repair adopts interface reconstruction and topology repair algorithms.

[0025] Specifically, when constructing a realistic solder joint morphology model, this invention adjusts the surface tension, solid-liquid interface energy, and gravitational potential energy parameters of the liquid solder based on the principle of minimum liquid phase interface energy. It also constrains the dynamic morphological changes of the liquid phase during reflow soldering until the solder joint surface reaches a stable state. This process perfectly matches the natural formation mechanism of solder joints in actual reflow soldering processes, accurately reproducing the true geometric morphology of solder joints in different packaging forms such as BGA, QFP, SOP, and DIP. This reduces subsequent multiphysics coupling simulation errors caused by deviations between solder joint morphology modeling and reality. Simultaneously, addressing the issue of localized unclosed solid-liquid interfaces that easily occur in morphology simulation of complex solder joint structures, an interface reconstruction and topology repair algorithm is used for targeted repair. This ensures the integrity and surface closure of the solder joint model, enabling the repaired model to be accurately identified and successfully solidified by finite element preprocessing software. This provides reliable geometric model support for the smooth implementation of subsequent 3D structural modeling, mesh generation, and electro-thermal-magnetic-mechanical full-field coupling analysis.

[0026] Furthermore, in S2, the element quality coefficient of the high-quality mesh is ≥0.11, and the mesh is a tetrahedral mesh.

[0027] Specifically, after constructing the 3D structural model of the board-level package, this invention uses tetrahedral meshes to perform high-quality subdivision of the model and ensures that the element quality coefficient is ≥0.11. The tetrahedral mesh can accurately adapt to the geometric morphology of complex structures such as solder joints, pads, and wire interconnections in board-level electronic packaging components, preserving key structural details to the greatest extent during model discretization and avoiding simulation data deviations caused by mismatch between mesh morphology and actual structure. The element quality coefficient ≥0.11 standard can effectively avoid the finite element calculation non-convergence problem easily caused by low-quality meshes, ensuring accurate solutions for electromagnetic force distribution and additional heat field in subsequent electro-magnetic coupled field analysis, as well as reliable calculations of thermal stress and deformation response in thermo-mechanical coupled field analysis. This provides a solid foundation for the subsequent extraction of accurate thermal stress and strain amplitude data, thereby ensuring the accuracy of key solder joint lifetime prediction work based on these data, and further enhancing the stability and reliability of the entire electro-thermal-magnetic-mechanical coupled field finite element simulation lifetime prediction process.

[0028] Furthermore, in S3, when solving the electro-magnetic field equations, based on Ampere's circuital law and the Biot-Savart law, the parameters of the current boundary conditions are: peak current 10A, frequency 100MHz. In the three-dimensional structural model, the conductor material is Cu, the solder joint material is Sn63Pb37 eutectic solder, and the structural area force of the electromagnetic force distribution is 10. 2 -10 3 N / m 2 And it is concentrated in the current source and the bending area of ​​the wire.

[0029] Specifically, in the process of calculating the electro-magnetic coupling field, this invention solves the electro-magnetic field equations based on Ampere's circuital law and the Biot-Savart law. These two fundamental theorems of electromagnetics provide rigorous theoretical support for the quantitative calculation of electromagnetic effects, enabling accurate derivation of the self-excited magnetic field formed by the current path when the device is energized and the force exerted by the magnetic field on the wires and solder joints. This avoids deviations in the calculation of electromagnetic force distribution and additional heat field caused by a lack of theoretical basis in the solution method. Simultaneously, the current boundary condition parameters are set to a peak current of 10A and a frequency of 100MHz, parameters suitable for board-level electronic packages. The simulation of the device's electromagnetic environment under high frequency and high current conditions allows for a more accurate representation of the actual electromagnetic effects, providing a reliable initial field input for subsequent thermo-mechanical coupling analysis. Furthermore, the selection of Cu as the conductor material and Sn63Pb37 eutectic solder as the solder joint material, both commonly used in electronic packaging, ensures that their electromagnetic conductivity, resistivity, and other physical properties perfectly match those of the actual device, effectively avoiding simulation distortion caused by discrepancies between material properties and real-world conditions. The structural area force with a clearly defined electromagnetic force distribution is 10. 2 -10 3 N / m 2 Furthermore, by focusing on the current source and the bending area of ​​the conductor, the key parts that affect the mechanical response of the structure by electromagnetic force can be accurately located. In subsequent thermo-mechanical coupling analysis, these high-stress risk areas can be targeted, avoiding the deviation in local stress calculation caused by uniformly applying electromagnetic force to the whole structure. This further ensures the accuracy of thermal stress and deformation response calculations and lays the foundation for extracting reliable thermal stress and strain amplitude data.

[0030] Furthermore, in S4, the thermal cycling load is set according to the IPC-TM650-2672 standard, with the following parameters: upper temperature limit 125℃, lower temperature limit -55℃, heating rate 3℃ / s, and isothermal holding time 5min. In the thermo-mechanical coupled field analysis, the current is kept on to calculate the Joule heating effect, thereby realizing full-field coupling of electricity, heat, magnetism, and force.

[0031] Specifically, in the thermo-mechanical coupling field analysis, this invention sets the thermal cycling load according to the IPC-TM650-2672 standard. This standard is an authoritative industry standard in the field of electronic packaging reliability testing, ensuring that the setting of the thermal cycling load fully matches the temperature environment requirements of the actual service process of board-level electronic packaging devices, avoiding the disconnect between the simulation scenario and the real service conditions caused by inconsistent load standards or deviations from industry specifications. At the same time, the thermal cycling load parameters are specifically set as follows: upper temperature limit of 125℃, lower temperature limit of -55℃, heating rate of 3℃ / s, and holding time of 5min. These parameters accurately reproduce the extreme temperature change process and temperature stabilization stage that the device may experience in actual application, and can realistically simulate temperature alternation. The influence of changes on the thermal stress generated by the device structure provides a realistic load input for the accurate calculation of subsequent thermal stress and deformation response. In addition, maintaining the current state in the thermo-mechanical coupled field analysis to calculate the Joule heating effect, this operation incorporates the Joule heat generated by the electric field with the additional heat field and electromagnetic force distribution obtained from the electro-magnetic coupled field analysis into the thermo-mechanical analysis system. This completely realizes the full coupling simulation of the four fields of electro-thermal-magnetic-force, fully restoring the real scenario of the multi-physics field synergy when the device is powered on. This avoids the thermal field calculation deviation caused by ignoring the Joule heating effect in traditional analysis, thereby ensuring the accuracy of the thermal stress and deformation response calculation results and providing key support for the subsequent extraction of reliable thermal stress and strain amplitude data.

[0032] Furthermore, in S5, the calculation formula for the Basquin model is as follows:

[0033] Where, σ a For stress amplitude, The fatigue strength coefficient, denoted as fatigue life, and b as the fatigue strength index.

[0034] Specifically, this invention introduces the Basquin model into lifetime prediction calculations, providing a standardized theoretical basis for lifetime calculations and effectively avoiding prediction biases caused by fuzzy model parameters or unclear calculation logic. The model itself is suitable for stress-dominated fatigue failure scenarios, and its input σ... aThe data originates from the thermal stress results obtained from the electro-thermal-magnetic-force full-field coupling analysis mentioned earlier, ensuring a high degree of consistency between the model's application scenario and the data source. This allows for the accurate capture of the impact of stress amplitude changes on the fatigue life of solder joints, ensuring that the calculation process closely aligns with the actual failure mechanism of the device. Furthermore, the Basquin model, as one of the various fatigue life models selected in this invention, provides a clear computational logic that lays the foundation for the subsequent weighted fusion strategy. This ensures that the prediction results of this model and other fatigue life models have a unified computational dimension and comparable standards, thereby facilitating the orderly implementation of parallel predictions using multiple models. This supports the accuracy of the final calculation of the critical solder joint lifetime distribution, pushing the error between the lifetime prediction results and experimental verification results to within 10%, thus guaranteeing the reliability of the entire lifetime prediction system.

[0035] Furthermore, in S5, the Coffin-Manson lifetime prediction formula is:

[0036] Where Δγ is the equivalent non-shear strain range; Δε is the equivalent plastic strain amplitude, Δγ=√3Δε. denoted as the fatigue elongation coefficient, c as the fatigue ductility index, and the equivalent plastic strain amplitude is related to the temperature cycle amplitude ΔT, the weld point geometric parameters, and the difference in the material's thermal expansion coefficient.

[0037] Specifically, this invention employs the Coffin-Manson lifetime prediction formula in lifetime prediction calculations, providing a rigorous theoretical framework for fatigue lifetime calculations dominated by plastic strain, effectively avoiding prediction deviations caused by parameter ambiguity or unclear formula logic. Crucially, this invention correlates the equivalent plastic strain amplitude with the temperature cycle amplitude ΔT, solder joint geometric parameters, and differences in the material's thermal expansion coefficient. This correlation perfectly aligns with the actual failure mechanisms of board-level electronic packaging devices—thermal stress caused by the mismatch in thermal expansion coefficients of different materials in electronic packaging, temperature fluctuations due to temperature cycling, and the influence of solder joint geometry on local strain are all core factors leading to the accumulation of plastic strain at solder joints and subsequent fatigue failure. This correlation ensures that the model calculations are no longer limited to... Instead of being limited to the theoretical level, this model deeply integrates the actual structure and service environment of the device, making the strain data source closer to real-world scenarios. Simultaneously, the equivalent plastic strain amplitude data required by the model originates from the deformation response results obtained from the aforementioned electro-thermal-magnetic-mechanical full-field coupling analysis, ensuring that the model input data has the reliability consistent with actual service conditions. Furthermore, its adaptability to plastic strain-dominated failure scenarios complements other models such as Basquin, providing predictive data support covering different failure types for subsequent weighted fusion strategies. This further ensures the comprehensiveness and accuracy of multi-model parallel prediction, helping to control the error between the final critical solder joint lifetime prediction results and experimental verification results within 10%, thus improving the engineering practicality and credibility of the entire lifetime prediction system.

[0038] Furthermore, in S5, the calculation formula for the Engelmaier model is as follows:

[0039] Where Δγ is the shear strain range.

[0040] Specifically, this invention introduces the Engelmaier model into the life prediction calculation, providing a rigorous theoretical basis for calculating the fatigue life of solder joints dominated by shear strain, effectively avoiding prediction deviations caused by fuzzy model parameters or unclear calculation logic. The model itself is suitable for fatigue failure scenarios caused by shear deformation of solder joints, and the required shear strain range Δγ data originates from the deformation response results obtained from the aforementioned electro-thermal-magnetic-force full-field coupling analysis. This ensures a high degree of consistency between the model's application scenario and the data source, accurately capturing the influence of shear strain changes on the fatigue life of solder joints, aligning with common shear failure mechanisms of solder joints in board-level electronic packaging devices. Simultaneously, this invention combines the prediction results of the Engelmaier model with those of other fatigue life models through a weighted fusion strategy. The weights are automatically determined by the degree of strain dominance (plastic strain / shear strain ratio) and load type (thermal cycling / thermal-magnetic coupling / thermal-magnetic-vibration composite), adapting to different failure-dominant factors and service environments without manual intervention. This not only solves the limitation of a single model only being suitable for specific failure scenarios but also ensures the synergy and rationality of the prediction results from multiple models. More importantly, the error between the lifetime prediction results and the experimental verification results under this weighted fusion strategy is ≤ ±8%, which is significantly better than the error level of the traditional thermal-mechanical coupling prediction. This further improves the accuracy and reliability of the lifetime distribution calculation of key solder joints, providing electronic packaging designers with a more reliable basis for reliability assessment, and helping to shorten the testing cycle and reduce experimental costs.

[0041] A storage medium storing a computer program, which, when executed by a processor, implements the above-described method for predicting the lifetime of board-level electronic packaging devices using finite element simulation of electro-thermal-magnetic-force coupling fields.

[0042] Specifically, the storage medium provided by this invention stores a computer program for a finite element simulation lifetime prediction method for the electro-thermal-magnetic-mechanical coupling field of a corresponding board-level electronic packaging device. This allows the lifetime prediction method to be stably stored and conveniently called in program form, eliminating the dependence on specific hardware development environments. Whether it is different workstations of the electronic packaging design team or simulation analysis needs at different stages of R&D, the entire process from solder joint morphology modeling, 3D mesh generation, electro-magnetic coupling field calculation, thermal-mechanical coupling field analysis to multi-model fusion lifetime prediction can be completely reproduced simply by reading the computer program in the storage medium. There is no need to repeatedly carry out the underlying technology development of the method, which significantly improves the reusability and engineering dissemination efficiency of the lifetime prediction technology. Meanwhile, when executed by a processor, this computer program can accurately perform the aforementioned full-field coupled simulation of electro-thermal-magnetic-force (such as solving the electromagnetic force distribution based on Ampere's circuital law and setting thermal cycling loads in conjunction with the IPC-TM650-2672 standard), parallel calculation of multiple fatigue models (Basquin, Coffin-Manson, Engelmaier models), and weighted fusion strategies. It can adapt to various packaging structures such as BGA, QFP, SOP, and DIP, as well as various load environments such as thermal cycling, thermal-magnetic coupling, and thermal-magnetic-vibration composites. Moreover, the error between the lifetime prediction results and experimental verification results is controlled within a reliable range. Therefore, with the help of this storage medium, electronic packaging designers can conveniently obtain reliability assessment basis that meets actual needs, effectively shorten the testing cycle, reduce experimental costs, and further promote the transition of electronic packaging reliability research towards "multi-field lifetime prediction," allowing this lifetime prediction system to be efficiently implemented in a wider range of engineering scenarios.

[0043] A computer device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-described method for predicting the lifetime of board-level electronic packaging devices using finite element simulation of electro-thermal-magnetic-force coupling fields.

[0044] Specifically, the computer device provided by this invention stores the computer program for the finite element simulation lifetime prediction method of the corresponding board-level electronic packaging device's electro-thermal-magnetic-force coupling field in its memory, and simultaneously relies on the processor to achieve efficient execution of the program, providing a stable hardware carrier and computational support for the entire lifetime prediction process. The memory can properly store the material parameters required for simulation (such as the thermal expansion coefficient of Cu, fatigue performance parameters of Sn63Pb37 solder), load condition data (such as the temperature range of thermal cycling, current boundary parameters), and intermediate simulation results, avoiding the tediousness of data loss or repeated calls; while the processor can efficiently complete the simulation from the solder joint... The entire process, from topographic modeling (iterative calculation based on the principle of minimum free energy), 3D mesh generation (ensuring element quality coefficient ≥ 0.11), electromagnetic coupling field solution (based on Ampere's circuital law and Biot-Savart law), thermo-mechanical coupling field analysis (applying thermal cycling loads according to IPC-TM650-2672 standard) to multi-model lifetime prediction (parallel calculation and weighted fusion of Basquin, Coffin-Manson, and Engelmaier models), ensures smooth data transfer between each step and timely calculation results, avoiding simulation lag or result delays due to insufficient computing power. Furthermore, since this computer device can fully reproduce the lifetime prediction function that is compatible with various packaging structures (BGA, QFP, SOP, DIP, etc.) and various load environments (thermal cycling, thermo-magnetic coupling, thermo-magnetic-vibration composite, etc.) when executing programs, and can control the error between the prediction results and experimental verification results within a reliable range, electronic packaging designers do not need to rely on complex multi-device combinations. They can conveniently conduct device reliability assessments and quickly obtain critical solder joint lifetime distribution data through this computer device alone. This effectively shortens the testing cycle, reduces experimental costs, and further promotes the transition of electronic packaging reliability research from "multi-field stress analysis" to "multi-field lifetime prediction," making this lifetime prediction system more operable and implementable in actual engineering design.

[0045] Example 1: This embodiment takes board-level electronic packaging components including typical packaging forms such as BGA, DIP, QFP, and SOP as the research object, and uses the electro-thermal-magnetic-force multi-physics field coupled finite element lifetime prediction method described in this invention to systematically verify the solder joint morphology evolution, electromagnetic effect transmission and lifetime calculation process.

[0046] I. Solder joint morphology simulation and geometric modeling For solder joints with different packaging forms, reflow soldering morphology evolution simulation was performed based on the principle of minimum liquid phase interface energy. By adjusting the surface tension of the liquid solder, solid-liquid phase interface energy, and gravitational potential energy parameters, the dynamic morphological changes of the liquid phase during reflow soldering were constrained, and simulation results of the stable state of the solder joint surface were obtained.

[0047] The simulation input parameters are as follows for different device types: 1) BGA solder joint: Solder joint height 0.28 mm, upper pad diameter 0.28 mm, lower pad diameter 0.36 mm, solder volume 0.146 mm². 3 ; 2) QFP and SOP pin configurations: Both adopt gull-wing pin structure, with a pin height of 0.02 mm and a solder volume of 0.03 mm². 3 ; 3) DIP solder joint: outer pad diameter 2 mm, through-hole diameter 1 mm, solder volume 0.3 mm. 3 .

[0048] The morphology of the solder joints after reflow soldering obtained from the simulation is as follows: Figure 1 As shown.

[0049] During the morphology simulation, some complex weld joints exhibited a problem of partially unclosed solid-liquid interfaces. After repair using interface reconstruction and topology repair algorithms, the weld joint surface was closed, ensuring that it could be recognized and solidified by the finite element preprocessing software.

[0050] Subsequently, a three-dimensional geometric model was established based on the actual PCB assembly structure (see...). Figure 2 The model is configured with wire routing, pads, and interconnect structures to ensure consistency with the actual device structure parameters.

[0051] II. Mesh Generation and Electro-Magnetic Coupling Analysis The repaired 3D solid model is discretized to generate a high-quality tetrahedral mesh with an element quality coefficient ≥0.11, ensuring computational accuracy and convergence.

[0052] In electro-magnetic field analysis, according to Ampere's circuital law, the current path of a device under energized conditions generates a self-excited magnetic field in the surrounding space. As the signal frequency increases, this magnetic field generates a Lorentz force in the form of an additional load in the conductors and solder joints, thereby altering the local stress distribution.

[0053] To quantitatively calculate this effect, a current loop model was established in electromagnetic simulation software (see...). Figure 3 The current source parameters are set as follows: peak current 10 A; frequency 100 MHz; wire material is Cu, and solder joint material is Sn63Pb37 eutectic solder.

[0054] Calculation results show that the surface force of the structure caused by the magnetic field excited by the current is within 10 2 -10 3 N / m 2The magnitude of the effects is mainly concentrated in the current source and the bending region of the conductor. This result is used as input to the subsequent thermo-mechanical analysis module of the prestressed field to reflect the influence of electromagnetic coupling on the structural mechanical response.

[0055] III. Thermo-mechanical Coupling Analysis and Load Setting According to the IPC-TM650-2672 standard, the upper limit of the thermal cycling load temperature is set to 125 ℃, the lower limit is -55 ℃, the heating rate is 3 ℃ / s, and the isothermal holding time is 5 min.

[0056] In the thermo-mechanical analysis, the electromagnetic thermal effect and Lorentz force distribution results obtained from the preceding electro-magnetic field analysis are used as initial boundary conditions and superimposed on the external thermal cycle load curve. The model maintains the current on-state to calculate the Joule heating effect, achieving full-field coupling of electro-thermal-magnetic-mechanical forces.

[0057] The temperature and stress field distributions of the component under four thermal cycles were calculated. Stress contour plots and plastic strain-time relationships at key solder joints are shown below. Figure 4 As shown.

[0058] The results show that the electromagnetic coupling effect significantly increases the temperature gradient and stress amplitude in the local solder joint area, especially at the edge solder joints of QFP and BGA devices, where stress concentration is most obvious.

[0059] IV. Life Prediction and Results Analysis To verify the broad applicability of the proposed electro-thermal-magnetic-mechanical multiphysics coupling simulation lifetime prediction framework, four typical package structures—BGA, QFP, SOP, and DIP—were selected for multi-model lifetime prediction calculations under different load and temperature cycling conditions. The stress distribution results (maximum equivalent stress) obtained from the simulation were compared and predicted using the Basquin model, Coffin-Manson model, and Engelmaier model to evaluate the prediction differences and applicability of each model under different service conditions.

[0060] The three types of lifetime models are selected as follows: Basquin model calculation formula:

[0061] In the formula σ a Stress amplitude (half the difference between the maximum and minimum stress); Fatigue strength coefficient; b: Fatigue life (number of cycles); c: Fatigue strength index; Coffin-Manson life prediction formula:

[0062] In the formula: Δγ is the equivalent non-shear strain range; Δε is the equivalent plastic strain amplitude, Δγ=√3Δε. denoted as the fatigue elongation coefficient and c as the fatigue ductility index, and the equivalent plastic strain amplitude is related to the temperature cycle amplitude ΔT, the weld point geometric parameters, and the difference in the material's thermal expansion coefficient, forming an engineering-grade Coffin-Manson formula.

[0063] Engelmaier lifespan prediction formula:

[0064] Where: N f : Fatigue life (number of cycles); Δγ: Shear strain range; : Solder fatigue ductility coefficient; c: Fatigue ductility index.

[0065] Lifetime predictions were performed based on the simulation stress results for various packaging types. The specific results are shown in Table 1.

[0066]

[0067] Table 1 Comparison of multi-model lifetime prediction results for different packaging structures The results show that under fully coupled electro-thermal-magnetic-mechanical conditions, the lifetimes of different packaging forms vary significantly. BGA and DIP, due to their larger solder joints and the significant impact of thermal-magnetic coupling, have relatively shorter lifetimes; while SOP and QFP devices, with their smaller structural dimensions, more uniform heat dissipation paths, and lower stress concentration, have longer lifetimes. The predicted trends across the models are consistent, validating the stability of the method.

[0068] In further research, three typical load conditions were applied to the QFP package: Thermal cycling load (–55~125 ℃); Electrothermal-magnetic combined load (10 A, 100 MHz); Combined load of energization and vibration (resonant frequency 200 Hz, amplitude 0.2 mm).

[0069]

[0070] Table 2. Prediction results of QFP solder joint life under different load conditions As shown in Table 2, with the superposition of electromagnetic load and vibration load, the stress level of the weld joint gradually increases, and the lifespan shows a significant decreasing trend. This pattern is consistent with the theoretical expectation of the synergistic effect of electromigration-accelerated failure and thermal fatigue, verifying the reliability and interpretability of the model in this invention under multi-load composite environments.

[0071] To further improve the universality of lifetime prediction, this invention introduces a multi-model fusion strategy in the lifetime calculation module, which takes a weighted average of the predictions from the Basquin, Coffin-Manson, and Engelmaier models, with the weights automatically determined by the degree of strain dominance and the type of load.

[0072] Reference Figure 5 As shown, the weighted lifetime prediction results are compared with the experimental verification results (accelerated aging test lifetime), and the error is controlled within ±8%, which is significantly better than the traditional thermo-mechanical coupling prediction (error of about ±20%).

[0073] The results show that the lifetime prediction framework proposed in this invention maintains high consistency and stability under different models, structures, and load conditions, and can realize a unified, multi-field, and scalable lifetime prediction system for packaged devices.

[0074] Existing technologies only address the differences in stress-life prediction at a single point in the packaging structure. This invention achieves the following breakthroughs in lifetime prediction: 1) Covers multiple packaging types and geometries, enabling universal lifetime assessment of board-level packaging systems; 2) Supports various physical loads (thermal cycling, electromagnetic, vibration) and their arbitrary combinations; 3) Integrate multiple fatigue models and optimize life prediction accuracy through fusion algorithms; 4) Achieve full-process modeling from stress response to lifetime distribution, forming a lifetime prediction platform that can be used for integrated design, simulation, and evaluation.

[0075] Therefore, this invention significantly broadens the application boundaries of multiphysics finite element technology in the field of electronic packaging lifetime prediction, extending it from single-field response calculation to a multi-model lifetime assessment system for complex service environments, fully demonstrating the innovation and wide applicability compared to the paper.

[0076] The embodiments of the present invention disclosed above are merely illustrative of the invention. These embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention.

Claims

1. A finite element simulation method for lifetime prediction of electro-thermal-magnetic-mechanical coupling fields of board-level electronic packaging devices, characterized in that, Includes the following steps: S1. Based on the actual structure of the board-level electronic packaging component, the solder joint process parameters are adjusted using a morphology simulation algorithm. Gravitational potential energy, surface energy, and interfacial tension are iteratively calculated according to the principle of minimum free energy to obtain a real solder joint morphology model. The closed defects on the surface of the solder joint in the real solder joint morphology model are repaired and then solidified. S2. Construct a three-dimensional structure model of the board-level package based on the actual device of the board-level electronic packaging component, and perform high-quality mesh generation on the three-dimensional structure model; S3. Apply current boundary conditions to the three-dimensional structural model to form a conductive loop, and obtain the electromagnetic force distribution and the additional heat field caused by electromagnetic effects by solving the electromagnetic field equations. S4. Input the additional heat field and electromagnetic force distribution obtained in S3 into the three-dimensional structural model as the initial field, apply thermal cycling, thermal storage or mechanical vibration loads to the three-dimensional structural model, and calculate the thermal stress and deformation response. S5. Substitute the thermal stress and strain amplitude values ​​obtained in S4 into at least two fatigue life models selected from the Basquin model, Coffin-Manson model, and Engelmaier model, and calculate the life distribution of key weld points using a weighted fusion strategy.

2. The finite element simulation lifetime prediction method for the electro-thermal-magnetic-mechanical coupling field of board-level electronic packaging devices according to claim 1, characterized in that, In S1, the surface tension, solid-liquid interface energy, and gravitational potential energy parameters of the liquid solder are adjusted according to the principle of minimum liquid phase interface energy to constrain the dynamic morphological changes of the liquid phase during reflow soldering. The process is iterated until the solder joint surface reaches a stable state, and a real solder joint morphology model is obtained. The closed defect repair adopts interface reconstruction and topology repair algorithms.

3. The finite element simulation lifetime prediction method for the electro-thermal-magnetic-mechanical coupling field of board-level electronic packaging devices according to claim 2, characterized in that, In S2, the element quality coefficient of the high-quality mesh is ≥0.11, and the mesh is a tetrahedral mesh.

4. The finite element simulation lifetime prediction method for the electro-thermal-magnetic-mechanical coupling field of board-level electronic packaging devices according to claim 3, characterized in that, In S3, the electro-magnetic field equations are solved according to Ampere's circuital law and the Biot-Savart law. The parameters of the current boundary conditions are: peak current 10A, frequency 100MHz. In the three-dimensional structural model, the wire material is Cu, the solder joint material is Sn63Pb37 eutectic solder, and the structural area force of the electromagnetic force distribution is 10. 2 —10 3 N / m 2 And it is concentrated in the current source and the bending area of ​​the wire.

5. The finite element simulation lifetime prediction method for the electro-thermal-magnetic-mechanical coupling field of board-level electronic packaging devices according to claim 4, characterized in that, In S4, the thermal cycling load is set according to the IPC-TM650-2672 standard, with the following parameters: upper temperature limit 125℃, lower temperature limit -55℃, heating rate 3℃ / s, and isothermal holding time 5min. In the thermo-mechanical coupled field analysis, the current is kept on to calculate the Joule heating effect, thereby realizing the full-field coupling of electricity, heat, magnetism and force.

6. The finite element simulation lifetime prediction method for the electro-thermal-magnetic-mechanical coupling field of board-level electronic packaging devices according to claim 5, characterized in that, In S5, the calculation formula for the Basquin model is: Where, σ a For stress amplitude, The fatigue strength coefficient, denoted as fatigue life, and b as the fatigue strength index.

7. The finite element simulation lifetime prediction method for the electro-thermal-magnetic-mechanical coupling field of board-level electronic packaging devices according to claim 6, characterized in that, In S5, the Coffin-Manson lifetime prediction formula is: Where Δγ is the equivalent non-shear strain range; Δε is the equivalent plastic strain amplitude, Δγ=√3Δε. denoted as the fatigue elongation coefficient, c as the fatigue ductility index, and the equivalent plastic strain amplitude is related to the temperature cycle amplitude ΔT, the weld point geometric parameters, and the difference in the material's thermal expansion coefficient.

8. The finite element simulation lifetime prediction method for the electro-thermal-magnetic-mechanical coupling field of board-level electronic packaging devices according to claim 7, characterized in that, In S5, the calculation formula for the Engelmaier model is: Where Δγ is the shear strain range.

9. A storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the finite element simulation lifetime prediction method for the electro-thermal-magnetic-force coupling field of the board-level electronic packaging device as described in claims 1-8.

10. A computer device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the finite element simulation lifetime prediction method for the electro-thermal-magnetic-force coupling field of a board-level electronic package device as described in any one of claims 1-8.

Images