The semiconductor industry is experiencing significant growth in production scale and investment, driven by rising demand for generative AI, high-performance computing (HPC), high-bandwidth memory (HBM), and high-performance/high-density chips. As a result, precision inspection and yield management at the wafer stage have become critical challenges. Probe cards, essential for verifying a chip's electrical performance, play a vital role in yield management. However, during repetitive inspection processes, probe cards absorb heat from the underlying chuck, leading to probe tip-pad alignment errors that degrade contact reliability and measurement accuracy. This situation necessitates a quantitative evaluation system based on thermo-structural coupled analysis. Additionally, the modeling process for multiple interposers and interposer housings, along with the preprocessing of contact conditions, adds complexity due to the increasing number of contact surfaces. This complexity can result in convergence issues and reduced accuracy. To address these challenges, this study employs Ansys Parametric Design Language (APDL) to enhance interposer and housing modeling, as well as contact problem resolution. It introduces a novel meshing method that positions nodes at target coordinates using point clouds, providing an effective analysis approach applicable to large, high-density probe cards and thermo-structural problems involving numerous contacts.

Induction heating is a technology that uses heat generated by resistance when a high-frequency current is applied to a coil. An electric range using this is called an Induction Heating (IH) electric range. IH electric ranges are being widely applied in commercial products recently because they have higher thermal efficiency performances than other methods. The performance of a heating coil of an IH electric range greatly varies depending on the shape and number of coils. Thus, research on optimal coil shape and number according to product shape is required. Therefore, this study aimed to design an optimal heating coil at the set temperature of an electric range product. Target temperature was set to the temperature that a commercial stainless-steel container could withstand. The thickness of the coil copper wire, the number of windings, the applied voltage, and the frequency were set as design variables. A sensitivity analysis was performed to check the influence of each design variable on coil temperature. Based on this, optimal design was performed using the response surface method. Electromagnetic field-thermal analysis was performed with the designed coil and a very approximate result was obtained with a 0.07% error from the set target temperature.

Precision positioning stages are devices for precisely positioning objects according to required degrees of freedom and performance. Precision positioning stages are classified into serial and parallel mechanisms. Except for specific applications, the parallel mechanism is preferred. In serial mechanism, dynamic characteristics such as resonant frequency are clearly different from axis to axis and the first resonance frequency is distinctly low compared to the second. These make the control performance different for each axis and incurs limitation in control. In this study, the first and second resonant frequencies in a serial 2-DOF precision positioning stage were increased while maintaining their approximal value. Compliance analysis for the stage was performed by applying the matrix based method. A new concept of resonant frequency isotropy (RFI) was introduced and design optimization was performed in which first and second resonant frequencies almost coincided. This optimization allowed for the design of a serial 2-DOF precision positioning stage with enhanced first resonance frequency by 50.8% and RFI by 80.2% compared to the initial design. This paper is expected to increase the use of precision positioning stages based on serial mechanism and apply the concept of RFI to the positioning stages with more than 2-DOF.