PSI - Issue 33

Aprianur Fajri et al. / Procedia Structural Integrity 33 (2021) 11–18 Fajri et al. / Structural Integrity Procedia 00 (2019) 000–000

12 2

1. Introduction Empirical load (maximum static load) and dynamic load are important things that must be considered when designing mechanical structures (Tasdemir and Nohut, 2012). Dynamic loads that are under static yield strength can cause failure if repeated for a long time (Browell and Hancq, 2006). This phenomenon is often called fatigue failure. Fatigue is a process of permanent structural change (progressive and localized) under conditions that produce stress and strain fluctuations whose values are below the ultimate tensile strength. (Bishara et al., 2018). Fatigue failure often occurs suddenly without any prior warning so that it can lead to disaster (Chen, 2019). Estimating fatigue life accurately is very important to do where various variables must be taken into calculation to minimize the losses that may be incurred (Kamal and Rahman, 2018). Experimental methods by conducting direct testing can be used to determine fatigue life in a structure. This method has the advantage that the results obtained are very accurate and close to the real value on the original conditions. Besides, the process will take a very long time and developing costs will be very expensive because the process requires prototyping and testing repeatedly. Direct testing is less effective when used to test objects with very complex geometries and large dimensions. The type of load that can be applied to the structure is also limited to certain directions, while in actual conditions the load received can be sourced from several directions. Finite Element Method (FEM – see in Prabowo et al. (2019, 2020); Ridwan et al. (2020); and Dabit et al. (2020)) based simulation comes as a solution to solve this problem. The characteristics of the design will be studied first using Computer-Aided Engineering (CAE – see in Ary et al. (2020); and Caesar et al. (2020)) software before being tested directly. Identification and visualization of failure mechanisms can be carried out using FEM-based simulations, especially in areas with high stress concentrations (Kahoul et al., 2019). Simulations using the FEM approach can reduce development time, reduce development costs, simplify design optimization, and improve product quality. This method is the best numerical simulation tool for solving problems related to mechanical failure (Cao et al. (2016); Prabowo et al. (2018); and Sedmak (2018)). However, benchmarking with previous studies should be carried out as validation to maintain accuracy in identifying fatigue life. 2. Finite Element Calculation The Finite Element Method (FEM) is a numerical approach used to solve mathematical equations with certain boundary conditions. In principle, when applied to solve a problem, a space model will be divided into several parts of the domain called finite element, this process is called meshing. Each of these elements will be searched for distribution of its value using the principle of interpolation and extrapolation. Doing fatigue analysis using stress life approach, then each element should be searched for stress value, then compared with fatigue data on S-N Curve. Several types of stress can be used as a basis for predicting fatigue life, namely normal stress ( S x or S y ) and shear stress ( S xy ). These three types of normal stress can be summed up with The von-Mises equation (Benasciutti et al., 2016). This type of stress is called equivalent stress von-Mises ( S eqv ) as in Eq. (1). (1) This stress component is used in this study because it includes all other stress components. Then, for the simulation of fatigue due to cyclic load need to be searched the magnitude of maximum ( S max ) and minimum ( S min ) stresses, then searched the ratio using Eq. (2). min max S R S = (2) Fatigue data in the form of S-N Curve which becomes input variables are generated from laboratory test results. Usually this data is taken at zero mean or R = -1, which means that the tensile and bending stresses have the same ratio. If the fatigue data will be used to analyze the problem under zero-based loading conditions (R = 0 or R = ∞) or with a certain ratio, it is necessary to correct the mean stress that occurs. There are several theories that can be 2 2 2 - 3 Seqv Sx Sy SxSy Sxy   