科技日新月異，對於產品需求日益進步，微小化以及高效能普及成為今日發展重點。FC-PBGA (Flip Chip Plastic Ball Grid Array)的發展成為此重點一個重要課題，利用FC-PBGA以及3D system in package，進而推廣出高密度封裝，以及微小化目的。
使用有限元素分析軟體ANSYS建立一精確之三維有限元素熱傳分析模型，於自然對流與強制對流環境中，針對一散熱強化型覆晶球柵陣列組合體的熱傳導、熱對流與熱輻射之行為進行模擬分析。這個散熱強化型覆晶球柵陣列組合體是基本型覆晶球柵陣列組合體包封著封膠，接著在封膠上面黏附著一片鋁質散熱板，最後以錫球構裝在印刷電路板上。本文報告了以上組合體之熱傳行為、散熱效能與溫度分佈。最後進行參數化分析，討論散熱板、封膠、黏著劑、凸塊、錫球、PCB、散熱孔的設計參數對這個組合體散熱性能的影響。
此研究材料元件包含基板 (Stubstrate)、錫球 (Solder Ball)、晶片 (Chip)、封膠 (Molding Compound)、散熱孔 (Thermal Via)、印刷機路基板PCB (Print Circuit Bpard )、凸塊 (solderbump)。

On the Thermal Performance of a Thermally Enhanced FC-PBGA Assembly
Jie Yu You
Jun-Huang Wu
Department of Mechanical Engineering of National Cheng Kung University
SUMMARY
In this thesis, the three-dimensional finite element analysis using the commercial ANSYS software is performed to study the thermal performance of a thermally enhanced FC-PBGA (Flip-Chip Plastic Ball Grid Array) assembly in both natural and forced convection environments. The thermally enhanced FC-PBGA assembly is a basic FC-PBGA package that is overmolded with molding compound, after which an aluminum heat spreader is adhered to the top of the molding compound and subsequently mounted on a PCB (Printed Circuit Board). The thermal behavior of the assembly is presented, thermal performance is analyzed by considering various design parameters of the aluminum heat spreader, molding compound , adhesive, solder ball , solder bump , PCB , and thermal via.
Keyword : Solder ball; FC-PBGA; Thermal performance
INTRODUCTION
In this thesis, the ANSYS 15.0 is used to study the heat transfer of FC-PBGA by the finite element method, and its heat transfer efficiency is improved by analyzing the Material size. Change the material size including heat spreader, molding compound , adhesive, solder ball , solderbump , PCB and thermal via to perform the parametrize. This thesis study.
MATERIAL AND METHOD
In this thesis, ANSYS 15.0 finite element analysis software is used to analyze the FC-PBGA structure by the finite element theory. The FC-PBGA assembly is composed of many original materials. The structure contains the molding compound, solder ball, Substrate, adhesive, thermal via, PCB, Substrate Signal, Chip, Copper layer, and heat spreader plate. The assumptions used in the analysis are as follows: 1. Due to the model symmetry, take a quarter of the model to reduce the analysis time.2. There is no deterioration of the material after temperature loading. 3. There is no stress in the initial state of the model, and the residual stress of the assembly process is not taken into account. 4. The symmetrical surface without any displacement. 5. Material is homogeneous and isotropic. 6. Does not consider the effect of gravity caused by the combination. 7. There is no impurities and voids are inside the assembly. 8. The structure itself does not absorb water vapor or the temperature change caused by the Popcorn Effect.
RESULTS AND DISCUSSTION
To discuss the distance, height ,and radius of solder balls to do temperature analysis and discussion. As shown in Figure 1. it’s effect of solder ball spacing variety on the highest assembly temperature and thermal resistance. It can be found that the smaller the solder ball spacing, the higher the temperature. In contrast, the greater the spacing, the lower the temperature. In general, from the perspective of heat transfer, the smaller the distance between solder balls, the harder it will be for the gap in the middle to allow the air to dissipate the heat to the outside world. Therefore, it is the best when the distance between solder balls is 1.3 mm from the simulation. From the curves in Figure 1., it is found that the impact of solder ball spacing on the natural convection effect, the temperature change is quite large, from 77 °C to 69 °C, mainly forced convection too much influence, resulting in forced convection of tin The ability of the ball to dissipate heat is poor, and it highlights the fact that in natural convection, changes in solder ball pitch have a significant effect on temperature. As shown in Figure 2., for the ball height changes and the maximum temperature and thermal resistance of the assembly. It can be seen from the figure that the change of solder ball height does not change much for the maximum temperature of the assembly, so the solder ball height changes have poor thermal performance. As shown in Fig. 3., it is the change of the center radius of the solder ball and the maximum temperature and thermal resistance of the assembly. The figure shows that the larger the radius, the higher the temperature of the assembly. When doing this simulation, because the solder ball spacing is fixed to 1.3 mm. The higher the center radius, the smaller the gap between the solder ball and the solder ball, which is the main reason for the slight increase in temperature.
CONCLUSION
Type 3 has the best heat dissipation and wind speed is the dominant cooling parameter. The smaller the solder ball pitch, the higher the thermal resistance. Understanding the changes in solder ball spacing has a high impact on natural convection and highlights the strong influence of forced convection. When the number of PCB layers is 4, 6 and solder ball spacing is 1.3 mm, the heat dissipation effect is the best. The bumps are closest to the wafer. The larger the bump radius, the smaller the thermal resistance; the thicker the bump height, the greater the thermal resistance. The effect of aluminum heat sink width is more significant than thickness, because increasing the width can effectively increase the cooling area of the heat sink. The results of the study found that reducing the gap of the molding compound or increasing the heat transfer coefficient of the molding compound contributes to the heat dissipation of the assembly because both reduce the thermal resistance of the heat transfer path. The width of the molding compound has little effect on the heat dissipation performance of the assembly. The effect of the glue gap is far greater than the effect of the width. The adhesive has little effect on the heat dissipation performance of the assembly, but the larger the thermal conductivity, the smaller the thickness contributes to the improvement of the heat dissipation performance of the assembly. There are thermal vias on both the PCB and the substrate. The heat dissipation capability is best, and the larger the heat sink radius, the smaller the thermal resistance.

[1] Wu, G. H., Ju, S. H. and Tsein, T. C. (2005) Three-dimensional Finite Element Analysis of Thermomechanical Behavior in Flip-chip Packages under Temperature Cycling Conditions, Journal of Reinforced Plastics and Composites, 24, 1895-1907.
[2] Joiner, B. and Montes de Oca, T. (2002) Thermal Performance of Flip Chip Ball Grid Array Packages, Eighteenth Annual IEEE Semiconductor Thermal Measurement and Management Symposium, 50-56.
[3] Lu, Xiangning , Shi, Tielin and Liao, Guanglan (2012) Thermal conduction analysis and characterization of solder bumps in flip chip package, Sciencedirect Applied Thermal Engineering, 36, 181-187.
[4] Lee, Tien-Yu (2000) An Investigation of Thermal Enhancement on Flip Chip Plastic BGA Packages Using CFD Tool, IEEE Transactions on Components and Packaging Technologies, 23, 481-489.
[5] Shim, K. W. and Lo, W. Y. (2006) Solder Fatigue Modeling of Flip-Chip Bumps in Molded Packages, Thirty-First IEEE/CPMT International Electronic Maunfacturing Technology Symposium, 109-114.
[6] Mertol, A. (1996) Thermal Performance Comparison of High Pin Count Cavity-Up Enhanced Plastic Ball Grid Array (EPBGA) Packages, IEEE Transactions on Components, Packaging and Manufacturing Technology: Part B, 19, 427-443.
[7] Ellison, G. N. (1989) Thermal Computations For Electronic Equipment, E. Robert , Krieger Publishing Company, Malabar, FL. USA..
[8] Chen,Wen-Hwa, Cheng, Hsien-Chie and Shen, Hsin-An (2003) An Effective Methodol-ogy for Thermal Characterization of Electronic Packaging, IEEE Transactions on Compon-ents and Packaging Technologies, 26, 222-232.
[9] Tarter T S, Goetz M P and Pagageorge M: 'BGA performance characteristics: a users design guide'. Proc of Tech Programme, SMI, 1995, 245-254.
[10] Retuta, D., Ma, Y.Y., Kanth, R., Tan, H.B., Sun, A., and Tanary, S., Thermal Performance Enhancement for CSP Packages, Proc 57th Electronic Components and Technology Conf, Reno, NV, USA, May-June. 2007,p p. 1690-1695.
[11] Miaowen Chen, Leo Huang, George Pan, Nicholas Kao, Don Son Jiang, “Thermal Analyses of a Package-on-Package (POP) Structure for Tablet Application” , 2014 IEEE 16th Electronics Packaging Technology Conference (EPTC)
[12] Bor Zen Hong and Tsomg-Dih Yuan ,"Integrated Flow-Thermal mechanical and Reliability Analysis ofa Low Air Cooled Flip Chip-PBGA Package
[13]Xiang Qiu, Jun Wang, "Study on Heat Dissipation in Package On Package (POP) ", 11th International Conference on Electronic Packaging Technology & High Density Packaging, 2010.
[14]黃舜治, Fatigue Analysis for Free-Lead Flip-Chip Ball Grid Array Assembly,國立成功大學機械工程學系, 2014.
[15]Ramakrishna,K.; T.Y. Lee, "Prediction of thermal performance of flip chip-plastic ball grid array (FC-PBGA) packages: effect of substrate physical design", ITHERM 2002. Eighth Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (Cat. No.02CH37258), pp528-537, 2002.
[16]C. F. Lin1, G. H. Wu1*, S. H. Ju2, "Investigation of Thermal Characterization of a Thermally Enhanced FC-PBGA Assembly", 1Department of Mechanical Engineering, National Cheng-Kung University, 2Department of Civil Engineering, National Cheng-Kung University, 2013
[17] K.W. Shim, and W.Y. Lo, “Solder Fatigue Modeling of Flip-Chip Bumps in Molded Packages,” IEEE International Electronic Manufacturing Technology, pp. 109-114, 2006.
[18] K. Biswas, S. Liu, X. Zhang, T.C. Chai, “Effects of detailed substrate modeling and solder layout design on the and level solder joint reliability for the large die FCBGA,” IEEE International Conference on Thermal, Mechanical and Multiphysics Simulation and Experiments in Micro-Electronics and Micro-Systems, pp. 1-7, 2008.
[19] H.U. Akay, Y. Liu, M. Tassaian, “Simplification of Finite Element Models for Thermal Fatigue Life Prediction of PBGA Packages,” ASME Journal of Electronic Packaging, Vol.125, pp.347-353, 2003.
[20] J.H. Lau, S.-W. Ricky Lee , "Modeling and Analysis of96.5Sn3.5Ag Lead-Free Solder Joints of Wafer Level Chip Scale Package on Build up Micro via Printed Circuit Board," IEEE T Electron Pack 25 (1):51-58,2002.
[21] Yaomin Lin, Student Member, IEEE, Wenning Liu, Yifan Guo, and Frank G. Shi,Member, IEEE"Reliability Issues of Low-Cost Over molded Flip-Chip Packages", IEEE Transactions on Advanced Packaging, Vol 28, NO:1,pp 79-88, 2005.
[22] Pei-Haw Tsao; Chender Huang ; Mirng -Ji-Lii;B.Su; Nun Sian Tsai,"Underfill characterization for low-k dielectric / Cu interconnect IC Flip-Chip package reliability", 2004 Proceedings. 54th Electronic Components and Technology Conference (IEEE Cat. No.04CH37546), Vol 1, 2004.
[23]S. Shidore; V. Adams; T. T. Lee,"A study of compact thermal model topologiesin CFD for a flip chip plastic ball grid array package", IEEE Transactions on Components and Packaging Technologies, vol. 24, No. 2, pp 191-198, 2001.
[24]Chih-Kuang Yu; Chun-Kai Liu; Ming-Ji Dai; Sheng-Liang Kuo; Chung-Yen Hsu, " A thermoelectric cooler integrated with IHS on a FC-PBGA package", 2007 26th International Conference on Thermoelectrics, pp274-278, 2007.
[25] 趙仁輝, Parametric Anaysis of FC-PBGA Thermal Performance, 國立成功大學機械工程學系, 2017.