J10
Dynamics & Vibration 2

Back to overview

08:35
conference time (CEST, Berlin)
Explicit Dynamic Analysis of Wafer Stage Cable Slab of EUV Lithography System
27/10/2021 08:35 conference time (CEST, Berlin)
Room: J
O. Khodko (ASML Netherlands B.V., NLD)
O. Khodko (ASML Netherlands B.V., NLD)
To produce microchips modern optical lithography systems are used. In an attempt to follow Moore’s law ASML is using extreme ultraviolet (EUV) light in their machines, which has a wavelength of only 13.5 nanometers. In EUV systems the wafer is rapidly moving inside a vacuum chamber where a blueprint of the chip is projected onto a silicon wafer with nanometer accuracy. To ensure that the wafer is at the right position at the right time, a position module is used. It’s connected to the base frame of the machine by means of a flexible connection called cable slab, which consists of cables for power and data, and hoses for transport of fluids and gasses. Because of high accelerations of the position module the cable slab behaves very dynamically. As a result, large deformations occur which might cause the cable slab to hit other parts inside the machine. Insufficient clamping force may also lead to slip between cables and brackets. This could result in damage or wear of the cable slab. Additionally, due to this dynamical behavior disturbance forces occur on the position module which negatively influence the positioning of the wafer. This article focuses on the prediction of the dynamic behavior of wafer stage cable slab in order to overcome existing issues and potentially minimize disturbance forces acting on the position module. To simulate complex non-linear dynamic behavior of cable slab at high operational speeds Altair RADIOSS explicit solver was used. For the propose of correct representation of cable slab’s behavior during different dynamic load cases, pre-loading of a system was performed by application of folding motion and gravity acceleration. Dynamic effects in these quasi-static load cases were minimized by using slow dynamic computation and energy discrete relaxation approaches to converge simulation towards static equilibrium. Following pre-loading steps, dynamic analysis of cable slab under various operating conditions was conducted. Proposed model allows fully describe the stress-stain state of cable slab at any given time and track its volume consumption during simultaneous movement of position module in two perpendicular directions, thus predict potential volume conflicts with surrounding parts. Knowledge about magnitude of contact forces at the interfaces allows to predict a wear of contacting parts. Disturbance forces on position module due to dynamic motion of cable slab also were investigated. Simulation shows a good level of correlation with experimental results obtained on test rig.
Semiconductor industry, microchips, cable slab, explicit analysis, RADIOSS, quasi-static load, dynamic load
08:55
conference time (CEST, Berlin)
NVH Optimization of Refrigerator Tubings for Structure Borne Noise Reduction Through Numerical Simulations
27/10/2021 08:55 conference time (CEST, Berlin)
Room: J
A. Jadhav, M. Kikale, A. Shedage, S. Paradhe (Whirlpool Corporation, IND); P. Kosowski (Whirlpool Corporation, POL)
A. Jadhav, M. Kikale, A. Shedage, S. Paradhe (Whirlpool Corporation, IND); P. Kosowski (Whirlpool Corporation, POL)
The recent stipulated guidelines on the noise levels in few regions of the world has fostered competition for quieter products, across various sectors from automobile to home appliances. Also, the customer expectations and preferences have encouraged the demand for quieter products. Domestic refrigerators are no exceptions to these advancements. The refrigerators have different components like compressor and fans which are responsible for noise generations. The compressor is one of the important and primary sources for vibration and noise generation, because of its intense inherent excitations. Many factors are influencing the compressor excitations such as type of the refrigerant, amount of the charge, compressor construction, operating speed etc. The suction and discharge tube (condenser tube) are very vulnerable to these compressor excitations as they are brazed directly to the compressor, unlike the baseplate which is generally equipped with isolation. Noise radiation from the condenser tube can add to the overall noise level of the refrigerator. Hence condenser tubes need to be routed with specific shapes for the noise reduction without altering core performance of the refrigerator. The study entails assessment of the condenser tube through series of testing or Computer Aided Engineering (CAE) simulations. In current work, Experimental Modal Testing (EMT) was performed and Frequency Response Function (FRF) were measured on the condenser tube at component and system level. A Finite Element Model (FEM) was developed using commercial software code and the FRF simulation results were validated with the test results to a high degree of accuracy, through appropriate representation of mass, local and global stiffness of the system. It was seen that variations in the interfacing conditions such as tube-plate connectivity, tube-plate thickness and compressor shell thickness were having significant impact on dynamic behavior of the condenser tube. The validated FE modelling strategy was further leveraged for the optimization of the tube shapes to improve the Noise Transfer Function (NTF) and has been demonstrated as a case study.
Refrigerator Tubing, Structure Borne Noise, Normal Modes Analysis, Experimental Validation, Frequency Response Function (FRF), Noise Transfer Function (NTF), Optimization
09:15
conference time (CEST, Berlin)
Why We Do ‘System Modeling’ for Geared Machine
27/10/2021 09:15 conference time (CEST, Berlin)
Room: J
J. Seo (FunctionBay, Inc, KOR)
J. Seo (FunctionBay, Inc, KOR)
Recently the requirement of system analysis is increasing drastically especially for dynamic simulation, where the driving system with high-speed electric motor is widely used. A rotating gear in a driving system generates specific frequency caused by gear and its pair, and it is called gear mesh frequency (GMF). This GMF is likely to cause system resonance, especially when driving speed varies. To find some problematic resonance in a system, numerical system modeling and simulation for the whole system is needed. In the meantime, electric motors may also be excessively vibrating at certain speed zone connected with gearboxes, so that, upfront design review of an entire assembly is more essential to avoid undesired system resonance. This presentation introduces a process of system modeling in terms of a driving system using two engineering software - RecurDyn and KISSsoft, and shows NVH result of a combined assembly. RecurDyn is a dedicate software for dynamic simulation and KISSsoft is an engineering software for gear design. By defining a gearbox system containing multiple gears and bearings using above software, GMF characteristics can be viewed and compared between simulation results and theoretical values. Since gears accelerate with linear or nonlinear slope profiles, system’s specific GMF as well as side frequency along with speed change was seen and checked. And, to derive system vibration characteristics from the system both from views of time-domain and frequency-domain at the same time, the gearbox was designed and defined with two different FEM methods, one is nonlinear flexible modeling for shafts and the other is linear flexible modeling for a housing. For higher reliability and accuracy of simulation result, an electric motor and an inverter model were also applied to the gearbox as an extra boundary condition. Torque ripple by electric motor is one of important excitation factors to the system vibration characteristic of gearbox. In order for understanding specific resonance in a complex system, a waterfall diagram (Campbell diagram) was used.
Driving system, Multibody dynamics, System modeling
09:35
conference time (CEST, Berlin)
Model Based Analysis of Overhead Crane and Inverted Pendulum
27/10/2021 09:35 conference time (CEST, Berlin)
Room: J
S. Yoshida (Shonan R&D, Inc., JPN)
S. Yoshida (Shonan R&D, Inc., JPN)
Because miscellaneous expenses such as an entrance fee, the facilities fee for use, the quay fee for use, the cargo-handling machine fee for use incur for coming alongside the pier in the quay, a large ship has to raise cargo work efficiency to the maximum. In this context, the container is required to stop its shaking, when the cart stops. When a cart of overhead crane suddenly stops, the cargo would swing for a while. On the other hand, by operating a cart in a procedure of the Full-Accelerate => Half-Decelerate => Half-Accelerate => Full-Decelerate, the cargo would not swing when the cart stops. This situation can be found by solving the following state-space representation, which is comprised of a cart position, cart speed, a pendulum angle of inclination, cart driving force. X ̇=AX+BU u- External Force (N) y ̇=v ϵ- Mass Ratio m_2⁄((m_1+m_2 ) ) (m_1=10(kg)&&m_2=1(kg) ) v ̇=ϵθ+u y- Cart Position (m) θ ̇=q v- Cart Velocity (m/s) q ̇=-θ-u θ- Pendulum Angle (rad) q- Pendulum Angular Velocity (rad/sec) Meanwhile, GEKKO Optimization Suite is a system seeking the solutions of the algebraic equation with non-linear Solver represented by (IPOPT, APOPT, BPOPT, SNOPT, MINOS) developed in Brigham Young University, U.S.A. After the calculation of GEKKO Optimization Suite, the optimal results are obtained, in which the pendulum stands still when the cart stops. Next, the external force history, which is obtained from GEKKO Optimization Suite is tried to replicate by OpenModelica modeling. As a result, the multiplication of cart mass and the cart acceleration obtained from OpenModelica accorded for a driving force history obtained from GEKKO Optimization Suite well. Finally, the same trend is obtained in the case of the inverted pendulum either, with the following state-space representation, of which configuration is almost same with that of overhead crane. X ̇=AX+BU u- External Force (N) y ̇=v ϵ- Mass Ratio m_2⁄((m_1+m_2 ) ) (m_1=10(kg)&&m_2=1(kg) ) v ̇=-ϵθ+u y- Cart Position (m) θ ̇=q v- Cart Velocity (m/s) q ̇=θ-u θ- Pendulum Angle (rad) q- Pendulum Angular Velocity (rad/sec)
State space representation, Optimization, OpenModelica, GEKKO Optimization Suite, Python, IPOPT
×

[TITLE]

[LISTING

[ABSTRACT]

[DATE]

[ROOM]

[KEYWORDS]