A method for predicting the deformation of IGBT modules during reflow soldering

Abstract

The present invention provides a method for predicting the deformation of an IGBT module in a reflow soldering process. The present invention comprises the following steps: preliminarily selecting the materials and geometric features of a substrate, solder and DBC liner according to application requirements; performing surface shape measurement on the arch surface of the substrate; detecting the material properties of components; preliminarily selecting the basic setting conditions of the reflow soldering process and performing the reflow soldering process of the IGBT module; recording the temperature change of each component in the reflow soldering, detecting the deformation and stress value of the module after the reflow soldering process, and performing flaw detection on the solder; establishing a thermal stress analysis simulation model based on the acquired data and calibrating the parameter settings; analyzing the reflow soldering process to obtain the deformation of the IGBT module substrate in the reflow soldering process and the residual stress of each component; and using the obtained simulation model to guide the actual process. The present invention is very accurate and reliable in analyzing and predicting the deformation and residual stress of the IGBT module in the reflow soldering process, which is conducive to predicting and controlling the influence of welding deformation and residual stress on the entire packaging result.

Description

Technical Field

[0001] The invention relates to the technical field of IGBT module reflow soldering, and in particular to a method for predicting the deformation amount of an IGBT module in a reflow soldering process. Background Art

[0002] IGBT (Insulated Gate Bipolar Transistor), also known as the insulated gate bipolar transistor, is a composite fully controlled, voltage-driven, power semiconductor device with a self-shutdown function. It is widely used in strategic industries such as home appliances, aircraft, ships, transportation, and power grids, and is known as the CPU in the power electronics industry. Figure 1 As shown in the figure, the IGBT module packaging structure mainly includes IGBT chip, freewheeling diode chip (FWD), bonding wire, conductive terminal, direct copper-clad ceramic board (DBC), welding layer, substrate, heat sink, silicone and packaging shell. Connecting the chip to the direct copper-clad ceramic board can effectively reduce the packaging volume and make the structure compact. The substrate and the copper layer under the DBC are connected by reflow soldering to form a chip-DBC connection for supporting the formation, and connected to the heat sink as a heat dissipation channel. Finally, high-temperature epoxy resin is used as a coating and the module is sealed with silicone gel to protect the entire packaging structure.

[0003] Reflow soldering is to place paste or sheet solder between the components to be soldered, and then cool the entire assembly to a temperature above the melting point of the solder before achieving soldering for mechanical and electrical connections between the solder ends of surface mounted components. IGBT modules are generally packaged using reflow soldering. During the packaging process, deformation and stress are inevitable due to differences in the thermodynamic properties of each component. Larger deformation and stress will affect the performance and life of the module, so it is necessary to predict the deformation of electronic devices during reflow soldering.

[0004] In the packaging structure, the IGBT heat dissipation substrate is a key component to ensure the normal operation of the IGBT module. Its function is to conduct the heat generated when the module is working so that the junction temperature of the chip is in an appropriate range. Therefore, copper or aluminum silicon carbide composite materials with good comprehensive performance are generally selected as substrate materials to ensure stable operation of the module and extend its service life. The heat dissipation substrate is also the component with the largest deformation during the reflow soldering process, and the deformation and residual stress of the substrate will also affect the deformation of other components. These deformations will have a direct impact on the coating of thermal grease and the heat dissipation performance of the module. In order to offset the impact of this deformation as much as possible, such as Figure 2 As shown in a and b, high-power IGBT modules mostly use packaging substrates with an arched lower surface or arched upper and lower surfaces to ensure that the substrate after the reflow process meets the packaging requirement of a residual arch less than 50μm, but this undoubtedly increases the difficulty of deformation prediction.

[0005] Surveys show that about 55% of electronic device failures are caused by heat dissipation and related problems. The evaluation indicators of the heat dissipation system generally include: the surface shape and residual stress of the substrate after reflow soldering, the residual stress of the solder, etc. The residual arch of the substrate after soldering meets the requirements within 50μm, while the residual stress of the substrate and solder should be reduced as much as possible.

[0006] The traditional methods for predicting the deformation of IGBT modules during the reflow process are mainly three-coordinate measurement and finite element simulation. The three-coordinate measurement method is to randomly select a number of IGBT modules from a batch of modules, scan the surface of the extracted substrate with arch before and after reflow, and then analyze the obtained data to obtain the deformation of the substrate, and use it to predict the deformation of the entire batch of workpieces. This method has a long cycle and high cost. When the solder or substrate material needs to be replaced, the sampling point location and measurement method must be changed accordingly, which is not universal. Moreover, this method can only test the deformation, and lacks the evaluation of the residual stress of the substrate after reflow.

[0007] On the other hand, the currently used finite element simulation method has many shortcomings. First, when building a CAD model, the initial arch modeling of the substrate is often ignored. For low-power IGBT modules without pre-warping, the solution can meet the basic requirements, but when solving high-power IGBT modules, it will directly affect the accuracy of the solution.

[0008] Second, the database used for simulation is also mostly established based on the data provided by the manufacturer, but these data often differ greatly from the actual values. The reasons are as follows: First, for composite materials (such as aluminum silicon carbide), composite materials composed of the same matrix and different mass (or volume) fractions of reinforcing phases will have large differences in thermodynamic properties, and manufacturers often only provide one or several sets of material performance parameters with good sales, and the reference data given are mostly range values; second, the same material will show different thermodynamic properties at different temperatures, and the parameters provided by the manufacturer often ignore the changes in material properties with temperature. In summary, using the data provided by the manufacturer for simulation will directly affect the accuracy of the solution, and it is necessary to establish an accurate and complete material property library.

[0009] Third, the melting and solidification of solder is not completed in a short time. Due to the influence of the latent heat of solder phase change, the solder will cause temperature differences inside the module during phase change, which will also make the deformation more complicated. The current simulation does not take this part as an important parameter of the material library.

[0010] Fourth, when evaluating the residual camber of the substrate, the currently used simulation method can only intuitively evaluate the residual camber at the maximum and minimum deformations, and lacks the judgment of the local residual camber in other areas. Therefore, a short-cycle, low-cost, and reliable method is needed to analyze the thermal stress of each component in the reflow process of the IGBT module, and obtain the temperature field distribution, deformation, and residual stress distribution of the IGBT module to guide production practice. Summary of the invention

[0011] According to the technical problem raised above, a method for predicting the deformation of IGBT module in reflow process based on the measured material thermodynamic property database is provided. The technical means adopted by the present invention are as follows:

[0012] A method for predicting the deformation amount of an IGBT module in a reflow soldering process comprises the following steps:

[0013] Step 1: Preliminary selection of materials and geometric features of each component of the IGBT module based on application requirements;

[0014] Step 2, measuring the surface shape or the curvature of the curved surface of the substrate selected in step 1;

[0015] Step 3: Based on the selection results of step 1, necessary thermodynamic performance tests are performed on each component;

[0016] Step 4: Preliminarily select the basic setting conditions of the reflow soldering process and implement the reflow soldering process;

[0017] Step 5: Establish a CAE initial model of the packaging structure based on the data obtained in steps 1, 2, 3, and 4;

[0018] Step 6: Based on the process implementation process of step 4, record the temperature change of the module during the reflow process, detect the deformation and stress value of the module after the reflow process, and perform flaw detection on the solder layer after the reflow process;

[0019] Step 7, evaluate the accuracy of the model constructed in step 5, and use the data obtained in step 6 to correct the CAE model;

[0020] Step 8: Use the CAE model obtained in step 7 to analyze the reflow process of the IGBT module to obtain the deformation of the substrate and the residual stress of the module;

[0021] Step 9: Use the CAE model obtained in step 7 to guide the actual production process.

[0022] Furthermore, in step 1, the preliminary selection of materials and geometric features of various components of the IGBT module according to application requirements includes the following steps:

[0023] Step 1a, determine the heat dissipation solution according to the power level of the IGBT module;

[0024] Step 1b, designing a heat dissipation structure according to the heat dissipation solution in step 1a, and determining the size of each component;

[0025] Step 1c, selecting the material of the heat dissipation structure components according to the heat dissipation structure in step 1b;

[0026] The heat dissipation solutions in step 1a include air cooling and liquid cooling;

[0027] The design of the heat dissipation structure in step 1b includes the geometric dimensions of the heat dissipation substrate, the DBC ceramic liner, and the solder and the assembly relationship between them;

[0028] The component materials described in step 1c include: heat dissipation substrate materials include copper and aluminum silicon carbide; DBC ceramic liner materials include aluminum nitride, aluminum oxide and silicon nitride; common solder materials include tin-lead alloy, copper-tin alloy and bismuth-tin alloy.

[0029] Furthermore, in step 2, the substrate is a single-sided arch substrate or a double-sided arch substrate, and the method for measuring the surface shape is a three-coordinate measurement method, and a higher model accuracy can be obtained in step 5; the tool for measuring the arch includes a vernier caliper, and only the arch is tested. When implementing step 5, an arc model needs to be used to fit the substrate surface shape.

[0030] Furthermore, the parameters to be measured in step 3 include: density, specific heat capacity, thermal conductivity, thermal expansion coefficient, Young's modulus and shear modulus of the substrate, DBC ceramic liner, bonding copper and solder; tensile strength, latent heat of phase change and Anand viscoplastic model parameters of the solder; the Anand viscoplastic model parameters include: initial deformation resistance, ratio of activation energy to universal gas constant Q / R, pre-exponential factor A, stress multiplier ξ, strain rate sensitivity of stress m, hardening or softening constant h0, deformation resistance saturation coefficient Saturation strain rate sensitivity n, strain rate sensitivity of hardening or softening a

[0031] Furthermore, the latent heat of phase change and specific heat capacity are measured by using a differential scanning calorimeter; the thermal conductivity is measured by using a laser thermal conductivity meter to measure the thermal diffusion coefficient at different temperatures, and then the thermal conductivity is calculated by formula (1) in combination with the specific heat capacity at different temperatures; the thermal expansion coefficient is measured by using a thermomechanical analyzer; as for the elastic modulus measurement part, the Young's modulus and shear modulus at different temperatures are directly measured by a dynamic thermomechanical analyzer, or the Young's modulus and Poisson's ratio at the key temperature are measured by combining the tensile test method with the binocular vision measurement method with the help of a variable temperature device; the measurement of the Anand model parameters of the solder requires a series of constant strain rate tensile tests at different temperatures and different strain rates according to the ASTM E8 / E8M-21 standard, and then the results are analyzed, and the nine parameters are obtained by fitting the Anand constitutive model;

[0032] λ c =α·ρ·C p (1)

[0033] In formula (1): c is the thermal conductivity of the material, W / (m·K); α is the thermal diffusion coefficient, mm 2 / s; ρ is density, g / cm 3 ; C p is the specific heat capacity, J / (kg·K);

[0034]

[0035] The Anand equation unifies the deformation caused by creep and plasticity, and can more accurately express the stress-strain response of the solder joint. The viscoplastic Anand equation is used to describe the performance of the solder joint, that is,

[0036]

[0037] Internal quantity evolution equation:

[0038]

[0039]

[0040]

[0041] In formulas (2), (3), (4), and (5), s0 is the initial deformation resistance, MPa; Q / R is the ratio of activation energy to the universal gas constant, K; A is the pre-exponential factor, 1 / s; ξ is the stress multiplier, dimensionless; m is the strain rate sensitivity of stress, dimensionless; h0 is the hardening or softening constant, MPa; is the saturation coefficient of deformation resistance, MPa; n is the saturation strain rate sensitivity, dimensionless; a is the strain rate sensitivity of hardening or softening, dimensionless.

[0042] Furthermore, the setting conditions of the reflow soldering process in step 4 include the selection of a heating method, a reflow oven conveyor belt speed, and a reflow temperature setting curve.

[0043] Furthermore, the setting of the reflow temperature curve includes the preheating zone heating rate, the active zone time length, the reflow zone heating rate and peak temperature, and the cooling zone cooling rate.

[0044] Furthermore, the CAE model establishment in step 5 includes the following steps:

[0045] Step 5a, establishing a CAD model of the packaging structure according to the data obtained in steps 1 and 2;

[0046] Step 5b, establishing a material property library used for analysis based on the data obtained in step 3;

[0047] Step 5c: Determine the thermal boundary conditions used for analysis based on the data obtained in step 4.

[0048] Furthermore, the method and accuracy of establishing the substrate arch surface in step 5a depends on the measurement result of step 2. The method of establishing the curved surface includes: fitting based on the surface shape measurement data using an approximate spline curve; establishing based on the arch measurement data using an arc model (Formula 6).

[0049]

[0050] In formula (6), a is the length of the substrate, mm; b is the width of the substrate, mm; c is the diagonal length of the substrate, mm; r is the radius of the imaginary circle, mm; d is the preset curvature of the substrate, mm; θ is the central angle of the imaginary arc, rad.

[0051] Furthermore, the material property library established in step 5b should include the thermal expansion coefficient, Young's modulus, shear modulus, specific heat capacity, thermal conductivity of the material at different temperatures; the phase change latent heat of the solder, and Anand model parameters.

[0052] Furthermore, the process of establishing the thermal boundary conditions in step 5c is: first determine the heat transfer method according to the type of reflow oven, and then determine the specific thermal boundary conditions according to the reflow oven settings and the sample size.

[0053] Furthermore, the temperature recording method in step 6 is a thermocouple method; the residual stress measurement method is a blind hole method or an X-ray measurement method; and the solder flaw detection method is an ultrasonic flaw detection or an X-ray flaw detection.

[0054] Furthermore, the step 7 comprises the following steps:

[0055] 7a. Use the initial model to solve and compare the obtained results with the test results in step 6 to evaluate the accuracy of the model;

[0056] 7b. Adjust the thermal boundary conditions, unit type and size so that the simulated temperature change curve of the entire welding workpiece obtained based on the CAE model is basically consistent with the corresponding actual temperature change;

[0057] 7c. Adjust the constraint application area and method, unit type and size so that the deformation and residual stress of the IGBT module obtained based on the CAE model are basically consistent with the deformation and stress data obtained in step 6; make the area where the solder residual stress is greater than its tensile strength basically consistent with the solder damage area obtained in step 6.

[0058] Furthermore, the guidance of the actual production process described in step 9 is to determine whether the solution result of step 8 meets the packaging requirements. If not, the settings corresponding to the actual process in the model are adjusted to meet the requirements, and then the actual process is modified according to the adjustment of the model.

[0059] The present invention has the following advantages: the present invention establishes a CAD model based on the preset arch of the substrate, actually measures the material properties of the material at different temperatures and establishes a material property library based on this. The module surface and solder temperature field solution results are basically consistent with the corresponding temperature changes; and the substrate deformation distribution, the surface shape of the substrate after welding and the residual stress on the substrate surface obtained based on the CAE model are basically consistent with the measured values, and the area of ​​the solder residual stress is greater than its tensile strength is basically consistent with the actual flaw detection results. The influence of the preset arch of the substrate, the thermodynamic properties of the material at different temperatures, the deformation generated in the reflow soldering process of the IGBT module, the module residual stress, and the solder residual stress (solder damage) on the welding deformation is fully considered, so that the analysis and prediction of welding deformation and residual stress are very accurate and reliable. The model constructed according to the method of the present invention is used to guide the actual production process of the IGBT module, which is very helpful to predict and control the influence of welding deformation and residual stress on the entire packaging result. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0061] Figure 1 This is a schematic diagram of the IGBT module packaging structure in the prior art.

[0062] Figure 2Schematic diagrams of two commonly used substrates with curvature in the prior art, wherein Figure 2 (a) is a single-sided arch substrate. Figure 2 (b) is a double-sided arch substrate.

[0063] Figure 3 is a schematic diagram of the arc model of the substrate of the present invention, wherein Figure 3 (a) is a schematic diagram of the substrate surface with curvature. Figure 3 (b) is a schematic diagram of the substrate arch.

[0064] Figure 4 The figure is a schematic flow chart of a method for predicting the deformation of an IGBT module in a reflow soldering process according to the present invention.

[0065] Figure 5 This is a schematic diagram of the surface shape of the lower surface of the IGBT module selected for this embodiment before welding.

[0066] Figure 6 Reflow temperature profile selected for this example.

[0067] Figure 7 Schematic diagram of CAE model mesh division in this embodiment.

[0068] Figure 8 Graph showing actual temperature variation in this embodiment.

[0069] Fig. 9 This is a schematic diagram of the surface shape of the lower surface of the IGBT module selected for this embodiment after welding.

[0070] Fig.10 FIG. 4 is a schematic diagram of solder damage in this embodiment.

[0071] Fig.11 This is a curve diagram of the solution result of the temperature change of the IGBT module in this embodiment.

[0072] Fig.12 It is a schematic diagram of the solution result of the substrate deformation in this embodiment.

[0073] Fig.13 It is a schematic diagram of the solution result of the residual stress of the substrate in this embodiment.

[0074] Fig.14 Schematic diagram of the solder residual stress solution results in this embodiment. DETAILED DESCRIPTION

[0075] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0076] The present invention preliminarily selects the materials and geometric features of the substrate, solder and DBC liner according to the application requirements, and measures the surface shape of the arch surface of the substrate; detects the material properties of the components; preliminarily selects the basic setting conditions of the reflow process and performs the reflow process of the IGBT module, and records the temperature changes of each component in the reflow process; detects the deformation and stress value of the module after the reflow process, and performs flaw detection on the solder; establishes a thermal stress analysis simulation model based on the acquired data and calibrates the parameter settings, analyzes the reflow process, obtains the deformation of the IGBT module substrate in the reflow process and the residual stress of each component; and uses the obtained simulation model to guide the actual process. The present invention not only establishes an actual measurement material property library that changes with temperature to increase the reliability of the model, but also adds a step of modifying the CAE simulation model based on the temperature change, deformation and residual stress. The analysis and prediction of the deformation and residual stress of the IGBT module in the reflow process is very accurate and reliable, which is conducive to predicting and controlling the influence of welding deformation and residual stress on the entire packaging result.

[0077] Specifically, Figure 4 The figure shows a flow chart of a method for predicting the deformation of an IGBT module in a reflow soldering process according to an embodiment of the present invention. This embodiment will be described in conjunction with an IGBT module for flexible DC transmission, that is, predicting the deformation and stress value of the module in reflow soldering. It should be understood that the deformation prediction method of the invention can also be applied to the reflow soldering process of other electronic devices. Figures 4 to 14 A detailed example illustrates the process of predicting the deformation of an IGBT module during reflow soldering.

[0078] First, in step 1, the materials and geometric features of each component of the IGBT module are preliminarily selected according to application requirements.

[0079] In step 1, the preliminary selection of materials and geometric features of various components of the IGBT module according to application requirements includes the following steps:

[0080] Step 1a, determine the heat dissipation solution according to the power level of the IGBT module;

[0081] Step 1b, designing a heat dissipation structure according to the heat dissipation solution in step 1a, and determining the size of each component;

[0082] Step 1c, selecting the material of the heat dissipation structure components according to the heat dissipation structure in step 1b;

[0083] The heat dissipation solutions in step 1a include air cooling and liquid cooling;

[0084] The design of the heat dissipation structure in step 1b includes the geometric dimensions of the heat dissipation substrate, the DBC ceramic liner, and the solder and the assembly relationship between them;

[0085] The component materials described in step 1c include: heat dissipation substrate materials include copper and aluminum silicon carbide; DBC ceramic liner materials include aluminum nitride, aluminum oxide and silicon nitride; common solder materials include tin-lead alloy, copper-tin alloy and bismuth-tin alloy.

[0086] In this embodiment, the module power is relatively large, and air cooling is selected as the heat dissipation solution; the substrate selects a single-sided arch substrate with an arched lower surface and a size of 187×136×5, the DBC ceramic liner size is 58×49×2, the solder size is 57×48×0.1, 6 groups of DBC liner plates are welded on the substrate, and the gap between two adjacent liner plates is 2mm; the heat dissipation substrate material is aluminum silicon carbide, the DBC ceramic liner material is aluminum nitride, and the solder material is tin-lead alloy.

[0087] Step 2: Measure the surface shape of the arched surface of the substrate selected in step 1.

[0088] Furthermore, in step 2, the substrate is a single-sided arch substrate or a double-sided arch substrate, and the method for measuring the surface shape is a three-coordinate measurement method, and a higher model accuracy can be obtained in step 5; the tool for measuring the arch includes a vernier caliper, and only the arch is tested. When implementing step 5, an arc model needs to be used to fit the substrate surface shape.

[0089] In this embodiment, the substrate is a single-sided arch substrate; a three-coordinate measuring instrument is used to measure the surface shape of the lower surface of the substrate to obtain higher accuracy. Figure 5 For the test results.

[0090] Step 3: Based on the selection results of step 1, necessary thermodynamic performance tests are performed on each component.

[0091] The parameters to be measured in step 3 include: density, specific heat capacity, thermal conductivity, thermal expansion coefficient, Young's modulus and shear modulus of substrate, DBC ceramic liner, bonding copper and solder; tensile strength, phase change latent heat and Anand viscoplastic model parameters of solder; the Anand viscoplastic model parameters include: initial deformation resistance, ratio of activation energy to universal gas constant Q / R, pre-exponential factor A, stress multiplier ξ, strain rate sensitivity of stress m, hardening or softening constant h0, deformation resistance saturation coefficient Saturation strain rate sensitivity n, strain rate sensitivity of hardening or softening a.

[0092] The latent heat of phase change and specific heat capacity are measured by differential scanning calorimeter. The thermal conductivity is measured by laser thermal conductivity meter to measure the thermal diffusion coefficient at different temperatures, and then the thermal conductivity is calculated by formula (1) in combination with the specific heat capacity at different temperatures. The thermal expansion coefficient is measured by thermomechanical analyzer. As for the elastic modulus measurement, the Young's modulus and shear modulus at different temperatures are directly measured by dynamic thermomechanical analyzer, or the Young's modulus and Poisson's ratio at key temperatures are measured by combining the tensile test method with binocular vision measurement method with the help of variable temperature equipment. The measurement of the Anand model parameters of solder requires a series of constant strain rate tensile tests at different temperatures and different strain rates according to ASTM E8 / E8M-21 standard, and then the results are analyzed to fit the nine parameters by the Anand constitutive model.

[0093] λ c =α·ρ·C p (1)

[0094] In formula (1): c is the thermal conductivity of the material, W / (m·K); α is the thermal diffusion coefficient, mm 2 / s; ρ is density, g / cm 3 ; C p is the specific heat capacity, J / (kg·K);

[0095]

[0096] The Anand equation unifies the deformation caused by creep and plasticity, and can more accurately express the stress-strain response of the solder joint. The viscoplastic Anand equation is used to describe the performance of the solder joint, that is,

[0097]

[0098] Internal quantity evolution equation:

[0099]

[0100]

[0101]

[0102] In formulas (2), (3), (4), and (5), s0 is the initial deformation resistance, MPa; Q / R is the ratio of activation energy to the universal gas constant, K; A is the pre-exponential factor, 1 / s; ξ is the stress multiplier, dimensionless; m is the strain rate sensitivity of stress, dimensionless; h0 is the hardening or softening constant, MPa; is the saturation coefficient of deformation resistance, MPa; n is the saturation strain rate sensitivity, dimensionless; a is the strain rate sensitivity of hardening or softening, dimensionless.

[0103] Table 1 is the test results of this embodiment.

[0104] Table 1 Test results of material properties of each component

[0105]

[0106] Step 4: Preliminarily select the basic setting conditions for the reflow soldering process and implement the reflow soldering process.

[0107] Among them, the setting conditions of the reflow soldering process include the selection of heating method, the speed of the reflow oven conveyor belt, and the reflow temperature setting curve. The setting of the reflow temperature curve includes the heating rate of the preheating zone, the time length of the active zone, the heating rate and peak temperature of the reflow zone, and the cooling rate of the cooling zone.

[0108] In this embodiment, the selected reflow oven is a hot air reflow oven, and the main heat transfer modes are heat convection and heat conduction; the conveyor belt speed is 80 cm / min; Figure 6 It is the reflow temperature curve used in the reflow soldering process. The reflow cycle is 1055s, the heating rate in the preheating zone is 2°C / s, the time length in the active zone is 400s, the heating rate in the reflow zone is 1.98°C / s, the peak temperature is 230°C, and the cooling rate in the cooling zone is 1.9°C / s.

[0109] Step 5: Establish a CAE initial model of the packaging structure based on the data obtained in steps 1, 2, 3, and 4;

[0110] Specifically comprising: step 5a, establishing a CAD model of the packaging structure according to the data obtained in steps 1 and 2;

[0111] Step 5b, establishing a material property library used for analysis based on the data obtained in step 3;

[0112] Step 5c: Determine the thermal boundary conditions used for analysis based on the data obtained in step 4.

[0113] The method and accuracy of establishing the arched surface of the substrate in step 5a depends on the measurement result of step 2. The method of establishing the curved surface includes: fitting based on the surface measurement data using an approximate spline curve; or Figure 3 As shown, the arc model (Formula 6) is used to establish the arc model based on the crown measurement data.

[0114]

[0115] In formula (6), a is the length of the substrate, mm; b is the width of the substrate, mm; c is the diagonal length of the substrate, mm; r is the radius of the imaginary circle, mm; d is the preset curvature of the substrate, mm; θ is the central angle of the imaginary arc, rad.

[0116] In this embodiment, step 2 uses the three-coordinate method for measurement, so the CAD model is established based on the measurement data by fitting using an approximate spline curve.

[0117] The material property library established in step 5b should include the thermal expansion coefficient, Young's modulus, shear modulus, specific heat capacity, thermal conductivity of the material at different temperatures; the phase change latent heat of the solder, and Anand model parameters.

[0118] The process of establishing the thermal boundary conditions in step 5c is as follows: first determine the heat transfer method according to the reflow oven type, and then determine the specific thermal boundary conditions according to the reflow oven settings and sample size.

[0119] In this embodiment, the reflow oven is a hot air reflow oven, and the heat transfer mode is heat conduction and heat convection. The thermal boundary conditions are determined according to the reflow oven settings and the sample size, and ANSYS-Workbench is selected for CAE modeling. Specifically, the thermal-stress analysis module can be used, and the mesh type can be a hexahedron, such as Figure 7 shown.

[0120] The established material property library includes the thermal expansion coefficient, Young's modulus, shear modulus, specific heat capacity, thermal conductivity of the material at different temperatures; the phase change latent heat of the solder, and Anand model parameters.

[0121] The process of establishing the thermal boundary conditions is as follows: first, the heat transfer mode is determined to be heat conduction and heat convection according to the hot air reflow oven, and then the heat conduction temperature curve and the heat convection heat transfer coefficient are determined according to the reflow oven settings and sample size.

[0122] Step 6: Based on the process implementation process of step 4, record the temperature change of the module during the reflow process, detect the deformation and stress value of the module after the reflow process, and perform flaw detection on the solder layer after the reflow process.

[0123] Among them, the temperature recording method is the thermocouple method; the residual stress measurement method is the blind hole method or the X-ray measurement method; the solder flaw detection method is the ultrasonic flaw detection or the X-ray flaw detection.

[0124] In this embodiment, the temperature variation curve of the heat dissipation substrate surface of the IGBT module is recorded by the thermocouple method. Figure 8 ; The surface shape of the lower surface of the heat dissipation substrate after the process is completed is obtained using a three-coordinate measuring instrument, such as Fig. 9Combined with the measurement results in step 2, it can be seen that the maximum deformation of the IGBT module is 0.2643mm; the residual stress on the substrate surface was detected using the blind hole method, and the average value obtained was 45.876MPa; the solder was tested using the ultrasonic flaw detection method, and the results are as follows Fig.10 , the stress damage area of ​​solder accounts for about 5%.

[0125] Step 7: Evaluate the accuracy of the model constructed in step 5 and use the data obtained in step 6 to correct the CAE model.

[0126] Specifically including: 7a, using the initial model to solve, and comparing the obtained results with the detection results of step 6 to evaluate the accuracy of the model;

[0127] 7b. Adjust the thermal boundary conditions, unit type and size so that the simulated temperature change curve of the entire welding workpiece obtained based on the CAE model is basically consistent with the corresponding actual temperature change;

[0128] 7c. Adjust the constraint application area and method, unit type and size so that the deformation and residual stress of the IGBT module obtained based on the CAE model are basically consistent with the deformation and stress data obtained in step 6; make the area where the solder residual stress is greater than its tensile strength basically consistent with the solder damage area obtained in step 6.

[0129] In this embodiment, the initial model is used for solving, and the obtained results are compared with the detection results of step 6. It is found that the temperature field and the static structure solution results are poorly consistent.

[0130] The thermal boundary conditions and unit size were adjusted so that the simulated temperature change curve of the entire welding workpiece obtained based on the CAE model was basically consistent with its corresponding actual temperature change;

[0131] The constraint application area and method were adjusted so that the deformation and residual stress of the IGBT module obtained based on the CAE model were basically consistent with the deformation and stress data obtained in step 6; the area where the solder residual stress was greater than its tensile strength was basically consistent with the solder damage area obtained in step 6.

[0132] Step 8: Use the CAE model obtained in step 7 to analyze the reflow process of the IGBT module to obtain the deformation of the substrate and the residual stress of the module;

[0133] In this embodiment, the reflow soldering process of the module is analyzed, and the average temperature change is obtained as follows: Fig.11 ; The deformation distribution is as follows Fig.12 , the maximum deformation is 0.26191mm; the residual stress distribution of the substrate is as follows Fig.13 ; The residual stress distribution of solder is as follows Fig.14 .

[0134] Step 9: Use the CAE model obtained in step 7 to guide the actual production process.

[0135] In this embodiment, the solution obtained in step 8 is: the residual camber of the substrate is 38.8 μm, and the residual stress is also much smaller than the yield strength of each material. Therefore, it is considered that the parameters selected for the process are reasonable and there is no need to adjust the process.

[0136] At this point, the method for predicting the deformation of the IGBT module in the reflow process shown in the figure is basically completed. The above embodiments study the deformation prediction and process improvement of the substrate of the IGBT module for flexible DC transmission, and preliminarily select the materials and geometric features of each component of the IGBT module according to the application requirements; measure the surface shape of the selected substrate with arch surface; perform necessary thermodynamic performance tests on each component; preliminarily select the basic setting conditions of the reflow process and implement the reflow process; establish the CAE initial model of the packaging structure; record the temperature change of the module in the reflow process, detect the deformation and stress value of the module after the reflow process, and perform flaw detection on the solder layer after the reflow process; evaluate the accuracy of the constructed model and correct the CAE model; use the obtained CAE model to analyze the reflow process of the IGBT module to obtain the deformation of the substrate and the residual stress of the module. Further, this CAE prediction model can be used to guide the actual production process.

[0137] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting the deformation of an IGBT module during a reflow soldering process, characterized in that: The steps include: Step 1: Preliminary selection of materials and geometric features of each component of the IGBT module based on application requirements; Step 2, measuring the surface shape or the curvature of the curved surface of the substrate selected in step 1; Step 3: Based on the selection results of step 1, necessary thermodynamic performance tests are performed on each component; Step 4: Preliminarily select the basic setting conditions of the reflow soldering process and implement the reflow soldering process; Step 5: Establish a CAE initial model of the packaging structure based on the data obtained in steps 1, 2, 3, and 4; Step 6: Based on the process implementation process of step 4, record the temperature change of the module during the reflow process, detect the deformation and stress value of the module after the reflow process, and perform flaw detection on the solder layer after the reflow process; Step 7, evaluate the accuracy of the model constructed in step 5, and use the data obtained in step 6 to correct the CAE model; Step 8: Use the CAE model obtained in step 7 to analyze the reflow process of the IGBT module to obtain the deformation of the substrate and the residual stress of the module; Step 9: Use the CAE model obtained in step 7 to guide the actual production process.

2. The method for predicting the deformation of an IGBT module in a reflow soldering process according to claim 1, characterized in that: In step 1, the preliminary selection of materials and geometric features of various components of the IGBT module according to application requirements includes the following steps: Step 1a, determine the heat dissipation solution according to the power level of the IGBT module; Step 1b, designing a heat dissipation structure according to the heat dissipation solution in step 1a, and determining the size of each component; Step 1c, selecting the material of the heat dissipation structure components according to the heat dissipation structure in step 1b; The heat dissipation solutions in step 1a include air cooling and liquid cooling; The design of the heat dissipation structure in step 1b includes the geometric dimensions of the heat dissipation substrate, the DBC ceramic liner, and the solder and the assembly relationship between them; The component materials described in step 1c include: heat dissipation substrate materials include copper and aluminum silicon carbide; DBC ceramic liner materials include aluminum nitride, aluminum oxide and silicon nitride; common solder materials include tin-lead alloy, copper-tin alloy and bismuth-tin alloy.

3. The method for predicting the deformation of an IGBT module in a reflow soldering process according to claim 2, characterized in that: The parameters to be measured in step 3 include: density, specific heat capacity, thermal conductivity, thermal expansion coefficient, Young's modulus and shear modulus of all or part of the materials at different temperatures, tensile strength, latent heat of phase change and Anand viscoplastic model parameters of the solder; the Anand viscoplastic model parameters include: initial deformation resistance, ratio of activation energy to universal gas constant Q / R, pre-exponential factor A, stress multiplier ξ, strain rate sensitivity of stress m, hardening or softening constant h0, deformation resistance saturation coefficient Saturation strain rate sensitivity n, strain rate sensitivity of hardening or softening a; The latent heat of phase change and specific heat capacity are measured by differential scanning calorimeter; the thermal conductivity is measured by laser thermal conductivity meter to measure the thermal diffusion coefficient at different temperatures, and then the thermal conductivity is calculated by combining the specific heat capacity at different temperatures; the thermal expansion coefficient is measured by thermomechanical analyzer; regarding the elastic modulus measurement part, the Young's modulus and shear modulus at different temperatures are directly measured by dynamic thermomechanical analyzer, or the Young's modulus and Poisson's ratio at key temperatures are measured by combining the tensile test method with the binocular vision measurement method with the help of variable temperature equipment; the enthalpy change of solder is measured by synchronous thermal analyzer to measure the energy change during the phase change process; the measurement of the Anand model parameters of solder requires a series of constant strain rate tensile tests at different temperatures and different strain rates according to ASTME8 / E8M-21 standards, and then the results are analyzed, and the above 9 parameters are obtained by fitting the Anand constitutive model.

4. The method for predicting the deformation of an IGBT module in a reflow soldering process according to claim 2, characterized in that: The setting conditions of the reflow soldering process in step 4 include the selection of the heating method, the reflow furnace conveyor belt speed, and the reflow temperature setting curve; the setting of the reflow temperature curve includes the preheating zone heating rate, the active zone time length, the reflow zone heating rate and peak temperature, and the cooling zone cooling rate.

5. The method for predicting the deformation of an IGBT module in a reflow soldering process according to claim 2, characterized in that: The method and accuracy of establishing the arched surface of the substrate in step 5a depends on the measurement result of step 2; the method of establishing the curved surface includes: fitting based on the surface shape measurement data using an approximate spline curve; establishing based on the arch measurement data using an arc model (Formula 6); In formula (6), a is the length of the substrate, mm; b is the width of the substrate, mm; c is the diagonal length of the substrate, mm; r is the radius of the imaginary circle, mm; d is the preset camber of the substrate, mm; θ is the central angle of the imaginary arc, rad; The material property library established in step 5b should include the thermal expansion coefficient, Young's modulus, shear modulus, specific heat capacity, thermal conductivity of the material at different temperatures; the phase change latent heat of the solder, and Anand model parameters; The process of establishing the thermal boundary conditions in step 5c is as follows: first determine the heat transfer method according to the reflow oven type, and then determine the specific thermal boundary conditions according to the reflow oven settings and sample size.

6. The method for predicting the deformation of an IGBT module in a reflow soldering process according to claim 2, characterized in that: The temperature recording method in step 6 is the thermocouple method; the residual stress measurement method is the blind hole method or the X-ray measurement method; the solder flaw detection method is the ultrasonic flaw detection or the X-ray flaw detection.

7. The method for predicting the deformation of an IGBT module in a reflow soldering process according to claim 2, characterized in that: The step 7 comprises the following steps: 7a. Use the initial model to solve and compare the obtained results with the test results in step 6 to evaluate the accuracy of the model; 7b. Adjust the thermal boundary conditions, unit type and size so that the simulated temperature change curve of the entire welding workpiece obtained based on the CAE model is basically consistent with the corresponding actual temperature change; 7c. Adjust the constraint application area and method, unit type and size so that the deformation and residual stress of the IGBT module obtained based on the CAE model are basically consistent with the deformation and stress data obtained in step 6; make the area where the solder residual stress is greater than its tensile strength basically consistent with the solder damage area obtained in step 6.

8. The method for predicting the deformation of an IGBT module in a reflow soldering process according to claim 2, characterized in that: The guidance of the actual production process described in step 9 is to determine whether the solution of step 8 meets the packaging requirements. If not, adjust the settings corresponding to the actual process in the model to meet the requirements, and then modify the actual process based on the adjustment of the model.

Images