Huei-Huang Lee

Application of Taguchi's method on the optimal process design of an injection molded PC/PBT automobile bumper

Abstract The manufacturing process of an injection molded polycarbonate/ poly(butylene terephthalate) automobile bumper was optimized, using Taguchis method, to eliminate silver streaks on the surface of the product. First, a set of process factors was selected and a preliminary experiment was conducted with an L 12 orthogonal array to screen the significant factors, check the validity of the data fitting model, and predict an improved process condition to overcome the filling difficulties. Second, using the selected control factors, a principal experiment was conducted with an L 18 orthogonal array to predict an optimal process condition that solve the quality problem. Finally, the predicted process condition was tested on an 2500-ton injection molding machine to confirm the prediction. The results reveal that the main factors that directly relate to the generation of silver streaks are the mold temperature, the fill time, the fill/ postfill switch over control, and the injection rate.

Prediction of paddle shift via 3-D TSOP modeling

This study develops a methodology to simulate the molding process of thin small outline packages (TSOPs) and to predict the paddle shift caused by pressure difference and viscous stress during molding through the help of a true three-dimensional (3-D) Computational Fluid Dynamics (CFD) software.

Simulation of warpage considering both thermal and cure induced shrinkage during molding in IC packaging

O C C U ~ owing to differences in shrinkage among constituent materials. Thermal shrinkage is usually considered as the main cause for warpage in an IC. However, cure induced warpage is also important during molding. Analysis without considering c u e induced shrinkage can not he accurate. This paper used the P-V-T-C relation and CTE of an encapsulation material to predict the amount of warpage and experimentally verifying the results. Comparing the simulation and experiment results of an EMC-copper composite shows that warpage predictions with P-V-T-C relation and CTE can better predict the amount of warpage of an IC package after molding. For an EMC-copper composite, the amount of warpage is proportional to the cure pressure and inversely proportional to the mold temperature. Besides the P-V-T-C relation and CTE, elastic modulus is found to play an important roll on the amount of warpage in simulation. With this study, the simulation shows that this approach can well predict the amount of warpage for an EMC composite after molding without too much computation effort.

The Development of Paddle Shift Analysis for IC Packaging Processing

During IC packaging processes, defects such as warpage, paddle shift and wire sweep etc, may happen. There was little literature about paddle shift and wire sweep analyses while most research in the past were focusing on warpage phenomenon in package outline. Excessive paddle shift reduces the encapsulation protection for the components, and may result in failure due to excessive wire sweep and stress, then cause the signal breakdown or a short circuit. Computer-aided analysis is one of the feasible tools for simulating and predicting the occurrence of such defects in molding process, even prior to the commencement of mass production of a package. Because of several analyses, such as mold filling analysis, wire sweep analysis, paddle shift analysis, warpage analysis, and stress strain analysis need to be assembled in one simulation series, more than one mesh models are needed for the analysis during the overall IC packaging analysis. The number of elements of those mesh model are often more than 600,000. For such large mesh model, a high quality methodology of mesh generation would be used. Unbalance pressure between top and bottom cavities was used as the loading for the analysis. In this study, a thin small outline package TSOP-I 48L produced by ChipMOS Technologies Ltd., was used for paddle shift simulation. In the simulation results, the lead shift phenomenon agreed very well with the experiments. This example demonstrated the feasibility and usefulness of this approach.

Prediction of Flow Induced Lead Deflection of Leadframe Assemblies in Plastic Encapsulation during Molding

Although leadframe assemblies have been well developed and plastic encapsulation has been popular and common for semiconductor packaging, lots of related problems still exist in the packages. During molding process, viscous epoxy molding compound (EMC) flows over wires and drags them for a short distance, then induces the well-known wire sweep problem. Similar phenomena can also happen to the leads of a leadframe and make them shift. The lead deflection phenomena were rarely mentioned or discussed before. However, extreme deflection for one lead could be sufficiently severe to cause leads contact and damage the device. A numerical simulation was developed to predict the lead deflection of a leadframe assembly which was encapsulated with EMC after molding process. X-ray detection of the encapsulated leads and optical microscope inspections of the encapsulated leads, were introduced to investigate and verify the lead shifting phenomenon. The experimental results showed satisfactory consistency with the simulation. The deflection of the leads was calculated by applying the distributed pressure from fluid onto the exterior surface of the leads. The distributed pressure of the flowing compound during encapsulation was obtained from a 3-D flow simulation. Experiments were conducted to obtain proper properties of materials for the flow and deflection simulation. Some key parameters such as package thickness, gate shape and down-set distance were studied to evaluate their effects on the deflection of the leads. It was found that the flow effect of the EMC during molding did play an important roll for the lead deflection and should be carefully considered when designing the leadframe. The results also showed that the approach proposed in tins research did provide a useful tool for the leadframe design engineers to avoid the lead deflection problems during encapsulation.

Mold Adhesion Force Measurement

During the encapsulation process of IC packages, when epoxy molding compound (EMC) is filling the mold cavity and cured in the mold, adhesion effects occur in the interface between EMC and mold surface. Adhesion effects can cause many problems. For example, too large an adhesion force may damage an IC during ejection and cause the package to fail and thus lower the yield rate and reliability. To get rid of the mold adhesion problems, improving the mold design and applying suitable surface treatments such as mold surface coating are the common approaches. Applying suitable surface coating is a more popular and practical approach. How to evaluate the mold adhesion force and use the data to improve products yield rate are the main issues for packaging. This paper uses a semi-automatic EMC adhesion force test instrument that had been developed and fabricated to measure normal and shear adhesion forces between the mold surface and EMC. By measuring the adhesion force, one can judge how much does a specific type of surface treatment help in reducing the amount of mold adhesion force. Nine kinds of various mold surface coating were tested with this instrument to measure the magnitude of adhesion force between EMC to determine the effectiveness of mold coating. The measured data showed that with different coatings on mold surface, the mold adhesion force can be very different. This paper also discussed the issue of successive normal force test. Engineers would use the most effective mold coating material to execute the continuous normal adhesion experiment. The variation of normal force during successive molding test could be used to predict the time for mold cleaning. By counting the total number of shots when the normal force begins to rise, engineers can accurately predict the number of shots for a specific kind of mold surface coating to be cleaned. This paper also described the effects of defrosting period on mold adhesion force of EMC. Defrosting is a process to increase the frozen EMC temperature to room temperature and stay at room temperature for some time before molding. It was found by molding engineers that increased defrosting time period would increase the frequency of mold cleaning. But there had been no quantitative description on how much mold adhesion force increase during the defrosting process. A semi-automatic EMC adhesion force test instrument was used to evaluate the effects of defrosting period on mold adhesion force of EMC. It was found that increase the defrosting time will increase the amount of adhesion force between EMC and mold surface. The higher is the relative humidity during defrosting on the adhesion force, the higher is the increase of adhesion force.Copyright