Papers by Keyword: Variable Amplitude Loading

Paper TitlePage

Authors: Li Xiong Gu, Zhi Fang Liu, Zhong Yong Xu
Abstract: Almost all load bearing components usually experience variable amplitude loading (VAL) rather than constant amplitude loading (CAL) during their service lives. The present study aims at evaluating residual fatigue life under VAL by adopting a dynamical coefficient mechanics (DCM) model which we have reported. New formulas connecting the crack length with number of cycles and expressions for the FCG rate under VAL have been derived and were used to predict crack propagation. The ratios of predicted-to-experimental lives range from 1.00 to 1.04, which indicates that the results obtained from this DCM model are in good agreement with experimental data from published literatures and cover all stages of fatigue crack growth curve.
Authors: Vitor Anes, Luís G. Reis, Manuel de Freitas
Abstract: Generally, mechanical components or structures are subjected to random and a three-dimensional stress state; there are very few field loading paths which can be experimentally fully simulated in laboratory. Loading path parameters such as load sequence, stress level or proportionality/non-proportionality presences are unknown variables with unknown levels under random loading conditions which are impossible to modulate in laboratory because the load spectra is unknown. The load spectrum depends on numerous factors such as environmental, mechanical or user behavior. At design stages the fatigue life estimation is based on typical loading paths or typical loading spectra, however that assumption may be very different from the usage regime. From here it can be concluded that the random multiaxial fatigue issue is of utmost importance to monitoring the in-field damage accumulation. This work presents a proposal to estimate the accumulated damage resulted from multiaxial random loadings based on the SSF equivalent stress and SSF virtual cycle counting concept.
Authors: Kwang Hwan Oh, C.K. Jung, Y.C. Yang, Kyung Seop Han
Abstract: This study investigated acoustic emission behavior during fatigue crack growth test under constant and variable amplitude loading in 304 stainless steel. To describe the acoustic emission behavior, counts rate(dη/dn) was related with stress intensity factor range (SIFR, ΔK) in log-log plot. As a result of test, the relationship was represented a curve, which forms rise and fall behavior in counts rate as the SIFR increases. AE response to a single overload was sudden drop and slow recovery in counts rate, which was similar to crack growth retardation behavior. Under block loading, counts rate of each loading block was same as that of constant amplitude loading. Overall experimental results indicated that stress intensity factor controls the counts rate (dη/dn) as well as crack growth rate (da/dn) regardless of load range or crack length.
Authors: M. Abdul Razzaq, Kamal A. Ariffin, Ahmed El Shafie, Shahrum Abdullah, Z. Sajuri, N.A. Akeel
Abstract: Artificial intelligence (AI) techniques and in particular, adaptive neural networks (ANN) have been commonly used in order to Fatigue life prediction. The aim of this paper is to consider a new crack propagation principle based on simulating experimental tests on three point-bend (TPB) specimens, which allow predicting the fatigue life and fatigue crack growth rate (FCGR). An important part of this paper is estimation of FCG rate related to different load histories. The effects of different load histories on the crack growth life are obtained in different representative simulation and experiments.
Authors: Jorge Alberto Rodriguez Duran, Dionisio Jose Rodrigues da Costa
Abstract: An experimental device was constructed with the aim of testing various cylindrical V-notched specimens until fracture and under variable amplitude torsional loads. The specimens had different notch depths resulting in the same number of values for the stress concentration factor. Strain gages directly bonded at the specimens’ surface and using a slip ring system for their communication with the conditioner, allowed the measurement of the actual applied loads. The well-known rain flow cycle counting procedure was then applied on the scaled signal for identifying the frequency of the 64 classes of stress amplitudes and means. The traditional nominal stress-based approach was then evaluated as the most widely used tool for fatigue lifetime calculations. As the occurrence of stress amplitudes above the endurance fatigue limit tends to lower it, the Miner elementary method was used. The results show damage sum ranges between 0.5 and 6.4 with a mean value of 2.0. Despite the small size of the sample used in the present paper (only 13 tests), these significant deviations are in agreement with previous results reported by different researchers.
Authors: A. Rama Chandra Murthy, G.S. Palani, Nagesh R. Iyer
Abstract: This paper presents methodologies for damage tolerance evaluation of tubular T- and Y-joints by using linear elastic fracture mechanics (LEFM) principles. The damage tolerance evaluation is in terms of crack growth analysis and remaining life prediction of tubular joints. Stress intensity factor (SIF) for T-butt plates which can be used for computation of SIF for tubular joints has been evaluated as per BS: 7910. It is observed from the literature that the expressions given in BS: 7910 for computation of SIF have not been used for remaining life prediction of tubular joints. In this paper, these expressions have been used for analytical prediction of remaining life of tubular T- and Y- joints subjected to constant amplitude loading (CAL) and variable amplitude loading (VAL). Wheeler residual stress model has been employed to represent the retardation effects due to tensile overloads. It is observed that remaining life predicted for T- and Y-joints under CAL are found to be in good agreement with those of experimental values reported in the literature. In the case of VAL, it is observed that crack growth retardation increases with increase of OLR resulting in higher predicted remaining life. It has also been observed that the predicted remaining life is influenced by the number of OLs and occurrence of OL. Early occurrence of OL causes the higher remaining life compared to later OLs.
Authors: Pavel Pokorný, Luboš Náhlík, M. Ševčík, Pavel Hutař
Abstract: The paper deals with the effects of variable loading on residual fatigue life of the railway wheelset. The railway wheelsets can include some cracks created during manufacturing process or during previous operation. Therefore, it is important to know how the existing cracks will behave during further service of the train. The experiments show that the fatigue crack growth rate depends not only on size of the load amplitudes in a loading spectrum, but it depends also on the sequence of load amplitudes. Taking into account interaction effects of overloading cycles requires use a method that calculates increments of crack length in each cycle, i.e. cycle-by-cycle. One of such methods represents generalized Willenborg model. This model was used for residual fatigue life time estimations of railway wheelset and results obtained were compared with classical approach, which did not take into account interaction effects. Results obtained can be used for establishing of service intervals of railway wheelsets.
Authors: Thomas Müller, Manuela Sander
Abstract: The main focus of this investigation is to clarify the influence of variable amplitude loadings on subsurface crack initiation and crack growth. Therefore, differently reconstructed load sequences on the basis of a standardized load time history called FELIX are investigated with an R-ratio of -1. The major amount of cycles is situated beneath the fatigue strength. A new damage calculation approach considering inclusion sizes is presented. Thus, the stress amplitude in the S-N curve was normalized with a calculated fatigue limit σw(area), which is defined by Murakami. Afterward, the fatigue life depending on the inclusion size is calculated using a Palmgren/Miner rule. The largest inclusion in the measurement volume was determined using extreme value statistics. Fatigue lives for each investigated load sequence were calculated taking the scatter of inclusion sizes into account.
Showing 1 to 10 of 31 Paper Titles