Doranga Sushil, Schuldt Matthew, Khanal Mukunda
Department of Mechanical Engineering, Lamar University, Beaumont, TX 77710, USA.
Materials (Basel). 2022 Sep 7;15(18):6208. doi: 10.3390/ma15186208.
Predictive analysis of the life of an electronic package requires a sequence of processes involving: (i) development of a finite element (FE) model, (ii) correlation of the FE model using experimental data, and (iii) development of a local model using the correlated FE model. The life of the critical components is obtained from the local model and is usually compared to the experimental results. Although the specifics of such analyses are available in the literature, a comparison among them and against the same electronic package with different user printed circuit board (PCB) thicknesses does not exist. This study addresses the issues raised during the design phase/life analysis, by considering a particular package with a variable geometric thickness of the user PCB. In this paper, the effect of stiffening the user PCB on the fatigue life of a ball grid array (BGA), SAC305 solder joint is studied. The board stiffness was varied by changing the thickness of the PCB, while the size of the substrate, chips, and solder balls were kept constant. The test vehicle consisted of BGA chips soldered to a user PCB. The thickness of the user PCB was varied, but the surface area of the BGA chip remained identical. The test vehicle was then modeled using a finite element analysis tool (ANSYS). Using a global/local modeling approach, the modal parameters in the simulations were correlated with experimental data. The first resonance frequency dwell test was carried out in ANSYS, and the high-cycle fatigue life was estimated using the stress-life approach. Following the simulation, the test vehicle was subjected to resonance fatigue testing by exciting at the first mode resonance frequency, the mode with the most severe solder joint failure. The resistance of the solder joint during the experiment was monitored using a daisy-chain circuit, and the point of failure was further confirmed using the destructive evaluation technique. Both the experimental and simulation results showed that stiffening the board will significantly increase the fatigue life of the solder joint. Although the amplitude of the acceleration response of the test vehicle will be higher due to board stiffening, the increase in natural frequencies will significantly reduce the amplitude of relative displacement between the PCB and the substrate.
对电子封装寿命进行预测分析需要一系列过程,包括:(i)开发有限元(FE)模型;(ii)使用实验数据对FE模型进行关联;以及(iii)使用经关联的FE模型开发局部模型。关键组件的寿命从局部模型中获得,并且通常与实验结果进行比较。尽管此类分析的具体细节在文献中已有记载,但对它们之间以及针对具有不同用户印刷电路板(PCB)厚度的同一电子封装进行比较的情况并不存在。本研究通过考虑用户PCB几何厚度可变的特定封装,解决了设计阶段/寿命分析过程中出现的问题。本文研究了增强用户PCB对球栅阵列(BGA)、SAC305焊点疲劳寿命的影响。通过改变PCB的厚度来改变板的刚度,同时保持基板、芯片和焊球的尺寸不变。测试载体由焊接到用户PCB上的BGA芯片组成。用户PCB的厚度发生变化,但BGA芯片的表面积保持相同。然后使用有限元分析工具(ANSYS)对测试载体进行建模。采用全局/局部建模方法,将模拟中的模态参数与实验数据进行关联。在ANSYS中进行首次共振频率驻留测试,并使用应力寿命方法估算高周疲劳寿命。模拟之后,通过在第一模态共振频率(焊点失效最严重的模态)下激励,对测试载体进行共振疲劳测试。在实验过程中,使用菊花链电路监测焊点的电阻,并使用破坏性评估技术进一步确认失效点。实验和模拟结果均表明,增强板的刚度将显著提高焊点的疲劳寿命。尽管由于板的刚度增强,测试载体的加速度响应幅度会更高,但固有频率的增加将显著降低PCB与基板之间相对位移的幅度。