Warpage Results Research Articles

Development of ANN-Based Warpage Prediction Model for FCCSP via Subdomain Sampling and Taguchi Hyperparameter Optimization.

This study aims to establish an accurate prediction model using artificial neural networks (ANNs) to effectively and efficiently predict the process-induced warpage of a flip-chip chip-scale package (FCCSP). To enhance model performance, a novel subdomain-based sampling strategy and Taguchi hyperparameter optimization are proposed in the ANN algorithm. To simulate the warpage behavior the FCCSP during fabrication, a process modeling approach is proposed, where the viscoelastic behavior of the epoxy molding compound is included, in which the viscoelastic properties are determined using dynamic mechanical measurement. In addition, the temperature-dependent thermal-mechanical properties of the materials in the FCCSP are assessed through thermal-mechanical analysis and dynamic mechanical analysis. The modeled warpage results are verified by the warpage measurement. Next, warpage parametric analysis is performed to identify the key factors most affecting warpage behavior for use in the construction of the warpage prediction model. Moreover, the advantages of the proposed sampling and hyperparameter tuning approaches are proved by comparing with other existing models, and the validity of the developed ANN-based deep learning warpage prediction model is demonstrated through a validation dataset.

Warpage and residual stress analyses of post-mold cure process of IC packages

To improve the production yield rate, reliability is one of the important indicators of electronic packaging products. In past research, however, the influence of the fabrication process was rarely taken into consideration. In this thesis, mold flow analysis software Moldex3D is used to develop a series analysis procedure for IC package products. The effects of many factors, including process, structure, and materials were being taken into account. Especially for epoxy molding compound, namely EMC, is studied on its properties during the molding and post-mold cure (PMC) processes. This paper adopted P–V-T-C equations, which consider both volume shrinkage due to thermal mismatch and chemical shrinkage to predict the amount of warpage and residual stresses after the mold filling process. Next, dual shift factor model for viscoelastic analysis was used to model the PMC process and predict the amount of warpage and residual stresses after PMC. And the influence of different PMC process conditions and loading conditions on the warpage results is discussed. The residual stresses after PMC simulation are set as the initial conditions for reliability analysis and then the stress distribution after two thermal cycles is analyzed. It is observed that the deformed shape of the simulation and experiment results after PMC were consistent. Both are concave downwards. In comparison with experiment results, the error of warpage simulation results was between 10% and 50%. The biggest error was found in the short direction. During two thermal cycles, it is can be found that the maximum stress of the lead frame is 505.7 MPa and the location of the possible failure is at the top left of the die. In addition, when considering or not considering the process-induced residual stress in the thermal cycle analysis, the stress states are very different.

Simulation methodology development of warpage estimation for epoxy molding compound under considerations of stress relaxation characteristics and curing conditions applied in semiconductor packaging

The thermal mismatch between different components has been a serious issue in the reliability of electronic products. The effect of epoxy molding compound (EMC) on the warpage of electronic package should be considered. The finite element method (FEM) is widely adopted to predict the warpage induced by the molding process. However, no previous studies focused on the warpage during the molding process and post process. In this study, several material characteristics of the EMC, such as cure reaction, cure shrinkage, and viscoelasticity, are measured. This study analyzes the corresponding mathematical model completely and substitutes the above properties into the FEM while the process conditions are considered through measurement. A process-oriented simulation methodology is proposed, and the estimated warpage results show that the simulation methodology with stress relaxation of EMC is in good agreement with the experimental value. The characteristics of stress relaxation behavior undergoing the molding process are discussed through the evolution of warpage. The warpage is different from the two process steps at room temperature due to the stress relaxation. This study demonstrates how the conditions of the post molding cure (PMC) process affect the warpage. The warpage can decrease with the extension of PMC time but the degree of reduction has a limit.

Modeling and Characterization of Cure-Dependent Viscoelasticity of Molded Underfill in Ultrathin Packages

Warpage control of an ultrathin package is critical in the semiconductor industry because of the large volume of molded underfill (MUF) encapsulated in such packages and because of the low stiffness of the entire package structure. The viscoelastic relaxation behavior that is observed during the MUF curing process results in a major warpage change and thus should be investigated. First, the material model of the cure-dependent viscoelasticity of MUF in the time domain and the curing kinetics model of MUF were successfully established based on the characterization results. Then, a finite-element (FE) model integrated with the material model for package warpage prediction was successfully developed. The final prediction results of package warpage obtained using a novel process modeling approach are in agreement with the experimental data measured using thermal shadow Moire. Furthermore, the effects of MUF volume change observed by changing the die size and thickness in an ultrathin package were studied using the warpage prediction model. The results indicate that package warpage can be effectively reduced by decreasing the die size and increasing the die thickness.

Investigation on forming defects and crystallization of plastic parts in combined in-mold decoration and microcellular injection molding based on a multiphase flow-solid coupled heat transfer model

Microcellular injection molding (MIM) combined with in-mold decoration (IMD) method can produce foamed parts with improved surface appearance by changing the heat transfer in the MIM mold. The temperature field has a significant effect on the forming defects and crystallization of the foamed parts. Based on finite volume method, a multiphase flow-solid coupled heat transfer model was established to calculate the temperature field in IMD/MIM process, and the implicit domain coupling algorithm (IDCA) was used to calculate the temperature field by taking the coupled heat transfer between the mold and the polymer melt into account. Compared with the experimental results, it is found that the established model can accurately predict the temperature field. The thermal response in IMD/MIM process was further analyzed. The effect of film on the forming defects of parts molded by IMD/MIM process was discovered by combining the simulation results with the experimental surface topography and warpage results, and the effect of film on the crystallization of parts molded by IMD/MIM process was also revealed.

Warpage Modeling of Ultra-Thin Packages Based on Chemical Shrinkage and Cure-Dependent Viscoelasticity of Molded Underfill

Molded underfill (MUF) is an essential component in ultra-thin flip chip packages to ensure their long-term reliability and mechanical integrity. However, the warpage evolution of package during the curing process of MUF has a critical effect on both the SMT yield and package reliability. Therefore, a thorough understanding of the MUF physical behavior under curing is essential. The present study therefore successfully establishes a novel process modeling approach based on finite element method to predict the final warpage of an ultra-thin flip chip scale package (fcCSP) in accordance with the chemical shrinkage and cure-dependent viscoelastic behavior of MUF. In describing the cure-dependent behavior of the MUF, the chemical shrinkage, curing kinetics and time-domain viscoelasticity are characterized by the pressure-volume-temperature (PVT) method, differential scanning calorimetry (DSC) and dynamic mechanical analysis (DMA), respectively. The predicted package warpage results are shown to be in good agreement with the experimental thermal Shadow Moire data. It is additionally shown that the accuracy of the warpage predictions is seriously degraded if the mechanical properties of MUF is modeled using a simple temperature-dependent elastic assumption.

Effect of warpage on the operation of a rapid cooling and heating device

Warpage of an injection molding product is caused by residual stresses from non-uniform cooling. In this paper, rapid cooling and heating devices were developed using a peltier module. Through a finite element program, it was confirmed that heat was transferred and maintained uniformly in a short time. With a traditional water cooling system and the rapid cooling and heating system developed here, warpage results were compared and analyzed. The injection molding conditions of the packing time, mold temperature, cooling time, and packing pressure were measured and compared. The cooling method of rapid cooling and heating devices undergoes less warpage than that in the conventional water cooling method, and uniform cooling was noted. The state of a polymer tested with the proposed system was confirmed by SEM photographs. The conventional water-cooling method shows a slow growth rate of a component’s morphology when the component is produced at a high mold temperature, and the rapid cooling and heating method leads to a rapid growth rate of a component’s morphology when the component is produced at a high mold temperature. This indicates that the cooling is uniform and that the internal stress decreases as the temperature of the mold slowly decreases.

A new characterizing method for warpage measurement of injection-molded thermoplastics

Injection molded plastics are widely used for mass production in auto industry. However, it is challenge to produce parts that do not warp. It is helpful to have an accurate way to measure the warpage when investigating and solving warpage issues. In this paper, a new technology is introduced to measure the part warpage quickly and accurately with the help of digital image correlation (DIC) method. The polypropylene plaque were fabricated by injection molding based on An L9 (3ˆ4) Taguchi design of experiments. The warpage of plaque are characterized for all the conditions and at different time after being ejected. Different kinds of warpage results are obtained by DIC method, such as maximum deflection, deflection distribution along desired path, warpage contour on the surface. The plaque warpage increases with the time. The maximum difference can reach as high as 64.9%. The cooling time has the largest influence on the warpage, while the packing pressure has the minor effect. Nevertheless, the warpage difference due to time varying depends on melt temperature, mold temperature, cooling time and packing pressure, in which mold temperature has the highest effects.

자동차용 콘솔 게이트 위치 선정을 위한 3차원 사출성형 시뮬레이션 활용

Injection molding simulation provided optimized design results by analyzing quality problems while the product is in assembly or in the process of manufacturing with make automobile plastics. Frequent change of design, change of injection molding, repetition of test injection which was held in the old way can now be stopped. And quality upgrade is expected instead. This report deals with the effect which the position of injection molding automobile console gate and number has on product quality including pressure at end of fill, bulk temperature at end of fill, shear stress of end of fill, residual stress at post filling end, product weld lines and warpage results. Simpoe-Mold simulates the complete manufacturing process of plastic injected parts, from filling to warpage. Simpoe-Mold users, whether they are product designers, mold makers or part manufacturers, can identify early into the design stage potential manufacturing problems, study alternative solutions and directly assess the impact of such part modification, whatever the complexity and geometry of such parts, shell part as plain solid parts.

Comparison of Warpage Measurement Capabilities and Results Obtained by Using Laser and Digital Fringe Projection Methods

Electronic packaged devices are becoming increasingly smaller in size and higher in density while requiring higher performance and superior reliability. Warpage is one of the crucial factors for the thermomechanical reliability of electronic packages and warpage control becomes a more crucial process during the printed wiring board (PWB) fabrication and package assembly processes. This requirement necessitates more accurate methods of measuring warpage. The fringe projection methods are recent trends for measuring the warpage of chip packages, PWBs, and PWB assemblies (PWBAs) because of their noncontact, full-field, and high-resolution measurement capabilities. This paper presents a comparison of two fringe projection methods: laser fringe projection (LFP) (projection moiré) and digital fringe projection (DFP). Experimental results show that digital fringe projection has higher practical resolution, and better accuracy and precision than laser fringe projection.

Effects of Copper Pattern Density and Orientation on the Modulus of BGA Substrates

Substrate material properties directly impact package mechanical performance. The component materials of a BGA substrate are the fiberglass-reinforced epoxy, copper, and solder-mask. Even when the material properties (modulus and CTE) of each component are known, the behavior of the final substrate is not easily determined because factors such as copper density and pattern have significant effects. For example, the modulus in the direction parallel to the copper traces will be different from the modulus in the direction perpendicular to the traces. Package mechanical simulations typically do not mesh the details of the copper patterns, but rather average within each layer, thus neglecting the orientation effects. Therefore FEM (finite element model) simulations may yield inaccurate stress and warpage results. In this study the effect of copper pattern density and orientation on substrate modulus was examined. Eight different test patterns were designed, each with a single simple pattern. These were fabricated into two metal layer substrates. Flexural modulus measurements from −65°C to 260°C were made using two types of instruments: DMA and Instron. Because the geometries were simple, FEM simulations were able to mesh the copper pattern details exactly. The exact FEM simulations matched the empirical data better than simulations using traditional layer averaging methods. The copper pattern effect was significant. The substrate modulus was lower in the direction perpendicular to the traces than in the direction parallel to the traces by 15% at room temperature and 35% at 260°C. The substrate modulus parallel to the copper traces increased linearly with copper pattern density as predicted by theoretical mixture rules. However, the substrate modulus in the direction perpendicular to the traces was not predicted theoretically. Therefore, more sophisticated averaging techniques are needed.

Longitudinal residual warpage of strip in continuous annealing line

After the heat treatment in continuous annealing line (CAL), the strip always exhibits phenomenon of residual warpage during punching process. This geometric defect is mainly attributed to the residual stress induced in the line and has a close relation with the temperature history of the strip. It becomes more serious for the wider strip and may significantly degrade the quality of the products. The contents of this study are mainly composed of establishing the energy model of the strip and roll to estimate the temperature distributions of the strip and the surface temperature of the roll, estimating the emissivity and equivalent heat transfer coefficient of the strip in each section of CAL, and establishing mechanical model of the strip to evaluate the distributions of contact pressure and residual stress of the strip especially along the width direction. Finally, the longitudinal residual warpage was calculated accordingly by virtual layers method (VLM). It was found that the residual warpage at both sides is larger than the central region. Moreover, the numerical results of strip warpage are well consistent with the experimental data which demonstrates the accuracy and reliability of the proposed method.

Integrative Analysis of the Injection Molding Process and Mechanical Behavior of Plastic Part

For parts made of plastics, injection molding is a common manufacturing process. Warpage and residual stress induced during the injection molding process have very important influences on the mechanical performance of injection products. In this paper, an integrative analysis of the injection molding process and mechanical performance of plastic parts is proposed. This integrative approach incorporates the effects of the manufacturing process in the mechanical simulation: (a) firstly, the finite element package MoldFlow is used to simulate the injection molding process and obtain the warpage and residual stress results. (b) Then the finite element model of plastic part including the process induced warpage and residual stress is established. Explicit dynamic finite element program LS-DYNA is used to simulate the mechanical behaviors of the molded part. Based on the integrative analysis, the influences of injection molding process parameters on mechanical behavior of a PC windshield against impact loading are studied.

Aspherical plastic lens injection molding warpage and opto-mechanical analysis

Abstract The aspherical plastic lens is commonly used in commercial optical products. The injection molding process introduces warpage and stress residue into the aspherical plastic lens. Mold flow software is generally used in injection molding simulation and successfully predicts molding warpage results. This paper attempts to link mold flow and optical design software by opto-mechanical analysis. The lens surface warpage can apply the Zernike polynomial and transform it to the optical aberration. Meanwhile, the injection molding process and the operation parameters also affect the plastic lens warpage value. Thus, according to optical software, the injection molding process parameters will affect lens optical performance. The plastic lens aberration results can evaluate the lens warpage effect and optimize the injection molding process parameters. Further, by reversing lens warpage aberration in the plastic lens model, it can preliminarily compensate for the warpage error in the lens design process.

Effects of Copper Pattern Density and Orientation on the Modulus of BGA Substrates

Substrate material properties directly impact package mechanical performance. Estimation of substrate mechanical properties based on each component layer provides maximum flexibility and better accuracy. The component materials of a BGA substrate are the fiberglass-reinforced epoxy layers, copper and soldermask. Even when the material properties (modulus and CTE) of each component are known, the behavior of the final substrate is not easily determined because factors such as copper density and pattern have significant effects. For example, the copper pattern may be oriented such that conductor traces run in one direction for a considerable length. The modulus in the direction parallel to the traces will be different from the modulus in the direction perpendicular to the traces. Package mechanical simulations typically do not mesh the details of the copper patterns, but rather average within each layer, thus neglecting the orientation effects. Therefore FEM (Finite Element Model) simulations may yield inaccurate stress and warpage results. This study utilized a test vehicle (TV) to study the copper pattern density and orientation effect on substrate modulus. Each TV sample was a two-metal-layer copper-clad laminate designed with one of eight simple geometric patterns. One set of patterns had Cu traces running parallel in one direction, with varied metal densities of 0%, 25%, 50%, 75% and 100% respectively. Measurements were made both parallel and perpendicular to the trace orientation. Also studied were a pattern of holes (continuous metal plane with unconnected holes) and a dot pattern (unconnected circular metal pads). Substrate flexural modulus measurements were made using two types of instruments: DMA and Instron. Data was collected from −65 to 260°C. FEM simulations which meshed the details of copper patterns were used to calculate the overall substrate modulus. These correlated better with the empirical data than simulations utilizing traditional layer averaging. The copper pattern had significant effects; the substrate modulus was always much lower in the direction perpendicular to the traces than the modulus in the direction parallel to the traces. The amount varied depending on temperature, but could be as high as 35%. When measured parallel to the copper traces, the substrate modulus was observed to increase with copper pattern density as predicted by theoretical mixture rules. However, the substrate modulus in the direction perpendicular to the traces was not easily predicted theoretically. Therefore, more sophisticated averaging techniques and simulations are needed.

Optimising Injection Moulding Parameter Setting in Processing Polypropylene-Clay Composites through Taguchi Method

The main purpose of this research is to optimise the injections moulding parameters in processing of thermoplastics samples made from Polypropylene (PP)-Clay with the fraction of 95% PP and 5% Clay, through Design of Experiment -Taguchi Method. The selection of defects or responses that need to be controlled for these thin layer products were warpage and shrinkage. The parameters that were taken into consideration were packing pressure; melt temperature, screw speed and filling time. In this research, the L934 orthogonal from Taguchi Method was chosen as the experiment setup. According to the results, 170°C of melt temperature, 80% of packing pressure, 70% of screw speed and 3 seconds of filling time gave minimum shrinkage of 3.45%. As for the warpage results, the best parameters combinations are 170°C of melt temperature, 70% of packing pressure, 70% of screw speed and 2 seconds of filling time. The Analysis of Variance for warpage shows that screw speed was the most significant factor with 41.83% contribution. As for shrinkage, the most influential factor was packing pressure with 98.48% contribution. Two factors have been pooled which are screw speed and filling time. Based on these findings, it is concluded that by controlling the optimum parameter setting, the best quality of desired product can be easily achieved and maintain through the in process production.

Application of Lamination Theory to Study Warpage Across PWB and PWBA During Convective Reflow Process

Out-of-plane displacement (warpage) has been a major reliability concern for board-level electronic packaging. Printed wiring board (PWB) and component warpage results principally from coefficient of thermal expansion (CTE) mismatch among the materials that make up the PWB assembly (PWBA). Warpage occurring during surface-mount assembly processes and normal operations may lead to severe solder bump failures. In this paper, the classical laminated plate theory was applied to study the warpage behavior of PWBs and PWBAs during the thermal reflow process. The rule of mixtures was used to estimate the effective material properties of PWBA composites. Closed form solutions of the differential equations for the PWB and PWBA deformation were generated from the classical lamination theory to evaluate the sample warpage. The board support conditions and thermal gradients through the assembly thickness were considered in the theoretical model. The calculated warpage results were compared with experimental results using the projection moiré measurement system in a simulated convective reflow process to evaluate the accuracy of the model. Finite element modeling was applied to estimate the sample warpage values under the same thermal loading and boundary constraints. The warpage values obtained from the theoretical models show high sensitivity to the thermal loads, through-the-thickness temperature gradients of the test samples, and CTE values of the composite materials. The relative warpage results obtained from the theoretical models and numerical models agreed well with the experimental measurement results during the thermal reflow process.

Study of Warpage and Mechanical Stress of 2.5D Package Interposers during Chip and Interposer Mount Process

As data transmission rate increases, flip chip plastic ball grid array (FCPBGA) utilizing a interposer for multiple chips is gaining popularity because of high electrical performance, ease of chip design, ease of thermal management with thermal lid, etc. The authors assessed package design configuration and key design elements for two chips application assuming 1600 signal I/Os for logic and 800 signal I/Os for memory. Then, we studied warpage behavior of the interposer, and mechanical stress of solder interconnections and low-k dielectric layer under controlled collapse chip connection (C4) pad. We set three different mount process assumptions for chip to interposer and interposer to base organic substrate. The mount process assumptions are (1) two pass reflow of chip to interposer first, then interposer to base organic substrate, (2) reversed sequence of two pass reflow which is interposer to base organic substrate first, then chip to interposer, (3) one pass reflow of chip, interposer and base organic substrate all together. We also set three different interposer material assumptions of Si, glass and organic in this study. We analyzed warpage behavior and mechanical stress using finite element method (FEM) modeling technique with a set of combinations of coefficient of thermal expansion (CTE) and elastic modulus of the interposers. The study also includes an analysis for conventional multi-chip-module (MCM) FCPBGA as a reference. We show the analysis results of interposer warpage, first principal stress at low-k dielectric layer under C4 pad and von Mises stress at solder interconnections of chip joining and interposer joining.

Effects of localized warpage and stress on chip‐on‐glass packaging: Induced light‐leakage phenomenon in mid‐sized TFT‐LCD

Abstract— Mura defects become visible in a 13.3‐in. TFT‐LCD using chip‐on‐glass (COG) packaging when the thickness of the glass substrate is decreased from 0.5 to 0.3 mm. Mura, the non‐uniform brightness in LCDs, is caused by COG packaging due to the mismatch of the coefficient of thermal expansion (CTE) and Young's modulus between the glass substrate and the IC‐driver Si chips. In this paper, a 3‐D finite‐element‐analysis (FEA) model, coupled with transient thermal analysis is first established to examine the warpage and stress behavior in the upper‐glass‐plate post‐COG‐package processing for identifying the root causes of the light‐leakage phenomenon. Prior to that, the simulated warpage results are validated by surface‐contour measurement. Data and modeling results show that a low bonding temperature together with a low modulus in novel ACF materials can effectively eliminate Mura. Besides, thinner silicon or a shorter length of Si chips as drivers offers enhanced reduction in the localized warpage, and thus can be a practical and low‐cost solution for eliminating mura defects.

Chemical Wet Etching of Silicon Wafers from a Mixture of Concentrated Acids

Warpage on the backside of silicon wafer after thinning process is examined. The thinning process includes back-grinding (BG) and wet chemical etching (WCE). Results of wafer warpage were compared to sub-surface damage from Transmission Electron Microscopy (TEM) analysis and showed that sub-surface damage on the backside of the silicon 100 would induce high wafer warpage, and reduced wafer strength. Further studies from surface roughness and topography of each surface finish is obtained by Atomic Force Microscopy (AFM) and SEM show that low surface roughness is in accordance with smooth surface condition, which comes after the wet etching process.