A flapper valve is made from a hardened and tempered high strength strip steel which opens and shuts as it is subjected to very high cyclic loads. Steel strip of which flapper valves are made from can   encounter a surface defect which are anticipated to influence fatigue life negatively. In this study, the influence of surface scratches on fatigue life of flapper valve strip was investigated. The analysis was carried out by using thirty samples that were blanked out of eight different steel strips in the transverse direction. Of these samples, fifteen of them had scratches on the surface and fifteen did not, all these samples were fatigue tested by constant amplitude method. An S-N curve was plotted based upon the values and results from the fatigue test, considering the curve as the nerve center in relation with fractrographic studies using the Scanning electron microscope. Therefore this master thesis work aims to explain the influence of scratches on fatigue life of flapper valve strip and suggest future improvements based on the findings.

In the last two decades, the emphasis in aircraft maintenance has been on developing online structural health monitoring systems to replace conventional non destructive inspection techniques which require considerable down-time, human effort and cost. Vibration based damage detection is one of the most promising techniques for implementation in Structural Health Monitoring (SHM). In vibration based methods, the presence of damage is detected by monitoring changes in one of the dynamic parameters of the structure, resonant frequencies, modeshapes or damping characteristics. Compared to modeshape based methods, frequency based methods have the advantage that measurements need to be taken only at a single location. Previous developments on frequency based techniques have relied on Finite Element Model updating; analytical techniques have hitherto been restricted to beams due to the complexity in developing equations for cracked two dimensional structures. In this thesis the analytical approach using an energy formulation is extended to plates with through-thickness cracks, where modeshapes from either numerical modelling or experimental measurements can be employed to determine the energy of vibration. It is demonstrated that by using a hybrid approach, incorporating experimentally measured modeshapes along with measured changes in frequencies, the damage parameters can be estimated without resorting to theoretical modelling or numerical analysis. The inverse problem of finding the crack location, size and orientation from measured changes in frequencies is addressed using minimisation techniques. The forward problem and the inverse algorithm is first validated using numerical simulation and experimental testing of beams with edge cracks and centre cracks. The application of the methodology to the two dimensional case is then validated by numerical simulation and experimental modal analysis of plates with through thickness cracks. A statistical procedure is developed for determination of the 90/95 probability of crack detection and the minimum detectable crack size in both cases. It is demonstrated that the measurement of frequency changes can be successfully employed to detect and assess the location and size of cracks in beams and plates, using modeshapes from theory, Finite Element Analysis.

Thesis (Ph. D.)--Electrical and Computer Engineering, Georgia Institute of Technology, 2008. / Committee Chair: Michaels, Jennifer; Committee Member: Habetler, Thomas; Committee Member: Jacobs, Laurence; Committee Member: Michaels, Thomas; Committee Member: Vachtsevanos, George.

Esta tese tem por objetivo avaliar a incerteza de medição no ensaio de fadiga do aço inoxidável AISI 316L em corpos de prova instrumentados ensaiados em temperatura ambiente em três níveis de rugosidade. Para isso, foi utilizado o método matemático de Monte Carlo para calcular o valor da incerteza com base nas variáveis de entrada. Para atingir esse objetivo foram realizados diversos ensaios de fadiga utilizando um número de corpos de prova suficientes para garantir confiabilidade estatística conforme norma ASTM E739-10 (ASTM, 2010), a uma frequência de ensaio de 20 Hz e razão de carregamento de 0,1. Nesta etapa 36 corpos de prova foram divididos em três grupos com valores de tensão 480, 500 e 520 MPa com três níveis de rugosidade distintos. Para o desenvolvimento do modelo matemático, cada fonte de incerteza teve distribuição de probabilidade determinada com base no comportamento estatístico característico de cada variável. Sendo assim, a partir de um modelo matemático desenvolvido foi possível determinar incerteza de medição do valor da tensão corrigida e do n° de ciclos, com uso do software Matlab®. Os resultados de incerteza de medição foram comparados aos valores obtidos pelo método Kragten, mostrando valores semelhantes entre os dois métodos matemáticos. / This thesis aims to evaluate the measurement uncertainty in the fatigue test of AISI 316L stainless steel in instrumented specimens tested at room temperature in three levels of roughness. For this, the mathematical method of Monte Carlo was used to evaluate the value of uncertainty based on the input variables. To achieve this goal several fatigue tests using a number of specimens sufficient to ensure statistical reliability according to ASTM E739-10 (ASTM, 2010) at test frequency of 20 Hz and stress ratio of 0.1 were tested. In this step 36 specimens were divided into three stress values of 480, 500 and 520 MPa with three different levels of roughness. For the mathematical development, each source of uncertainty probability distribution was initially determined based on the statistical behavior characteristic of each variable. A mathematical model was derived to determine the uncertainty of the stress value and number of cycles, using the Matlab® software. The combined uncertainty of results were compared to values obtained by Kragten method, showing similar values between the two mathematical methods.

Esta tese tem por objetivo avaliar a incerteza de medição no ensaio de fadiga do aço inoxidável AISI 316L em corpos de prova instrumentados ensaiados em temperatura ambiente em três níveis de rugosidade. Para isso, foi utilizado o método matemático de Monte Carlo para calcular o valor da incerteza com base nas variáveis de entrada. Para atingir esse objetivo foram realizados diversos ensaios de fadiga utilizando um número de corpos de prova suficientes para garantir confiabilidade estatística conforme norma ASTM E739-10 (ASTM, 2010), a uma frequência de ensaio de 20 Hz e razão de carregamento de 0,1. Nesta etapa 36 corpos de prova foram divididos em três grupos com valores de tensão 480, 500 e 520 MPa com três níveis de rugosidade distintos. Para o desenvolvimento do modelo matemático, cada fonte de incerteza teve distribuição de probabilidade determinada com base no comportamento estatístico característico de cada variável. Sendo assim, a partir de um modelo matemático desenvolvido foi possível determinar incerteza de medição do valor da tensão corrigida e do n° de ciclos, com uso do software Matlab®. Os resultados de incerteza de medição foram comparados aos valores obtidos pelo método Kragten, mostrando valores semelhantes entre os dois métodos matemáticos. / This thesis aims to evaluate the measurement uncertainty in the fatigue test of AISI 316L stainless steel in instrumented specimens tested at room temperature in three levels of roughness. For this, the mathematical method of Monte Carlo was used to evaluate the value of uncertainty based on the input variables. To achieve this goal several fatigue tests using a number of specimens sufficient to ensure statistical reliability according to ASTM E739-10 (ASTM, 2010) at test frequency of 20 Hz and stress ratio of 0.1 were tested. In this step 36 specimens were divided into three stress values of 480, 500 and 520 MPa with three different levels of roughness. For the mathematical development, each source of uncertainty probability distribution was initially determined based on the statistical behavior characteristic of each variable. A mathematical model was derived to determine the uncertainty of the stress value and number of cycles, using the Matlab® software. The combined uncertainty of results were compared to values obtained by Kragten method, showing similar values between the two mathematical methods.

abstract: Fatigue is a degradation process of materials that would lead to failure when materials are subjected to cyclic loadings. During past centuries, various of approaches have been proposed and utilized to help researchers understand the underlying theories of fatigue behavior of materials, as well as design engineering structures so that catastrophic disasters that arise from fatigue failure could be avoided. The stress-life approach is the most classical way that academia applies to analyze fatigue data, which correlates the fatigue lifetime with stress amplitudes during cyclic loadings. Fracture mechanics approach is another well-established way, by which people regard the cyclic stress intensity factor as the driving force during fatigue crack nucleation and propagation, and numerous models (such as the well-known Paris’ law) are developed by researchers. The significant drawback of currently widely-used fatigue analysis approaches, nevertheless, is that they are all cycle-based, limiting researchers from digging into sub-cycle regime and acquiring real-time fatigue behavior data. The missing of such data further impedes academia from validating hypotheses that are related to real-time observations of fatigue crack nucleation and growth, thus the existence of various phenomena, such as crack closure, remains controversial. In this thesis, both classical stress-life approach and fracture-mechanics-based approach are utilized to study the fatigue behavior of alloys. Distinctive material characterization instruments are harnessed to help collect and interpret key data during fatigue crack growth. Specifically, an investigation on the sub-cycle fatigue crack growth behavior is enabled by in-situ SEM mechanical testing, and a non-uniform growth mechanism within one loading cycle is confirmed by direct observation as well as image interpretation. Predictions based on proposed experimental procedure and observations show good match with cycle-based data from references, which indicates the credibility of proposed methodology and model, as well as their capability of being applied to a wide range of materials. / Dissertation/Thesis / Masters Thesis Materials Science and Engineering 2018

Esta tese tem por objetivo avaliar a incerteza de medição no ensaio de fadiga do aço inoxidável AISI 316L em corpos de prova instrumentados ensaiados em temperatura ambiente em três níveis de rugosidade. Para isso, foi utilizado o método matemático de Monte Carlo para calcular o valor da incerteza com base nas variáveis de entrada. Para atingir esse objetivo foram realizados diversos ensaios de fadiga utilizando um número de corpos de prova suficientes para garantir confiabilidade estatística conforme norma ASTM E739-10 (ASTM, 2010), a uma frequência de ensaio de 20 Hz e razão de carregamento de 0,1. Nesta etapa 36 corpos de prova foram divididos em três grupos com valores de tensão 480, 500 e 520 MPa com três níveis de rugosidade distintos. Para o desenvolvimento do modelo matemático, cada fonte de incerteza teve distribuição de probabilidade determinada com base no comportamento estatístico característico de cada variável. Sendo assim, a partir de um modelo matemático desenvolvido foi possível determinar incerteza de medição do valor da tensão corrigida e do n° de ciclos, com uso do software Matlab®. Os resultados de incerteza de medição foram comparados aos valores obtidos pelo método Kragten, mostrando valores semelhantes entre os dois métodos matemáticos. / This thesis aims to evaluate the measurement uncertainty in the fatigue test of AISI 316L stainless steel in instrumented specimens tested at room temperature in three levels of roughness. For this, the mathematical method of Monte Carlo was used to evaluate the value of uncertainty based on the input variables. To achieve this goal several fatigue tests using a number of specimens sufficient to ensure statistical reliability according to ASTM E739-10 (ASTM, 2010) at test frequency of 20 Hz and stress ratio of 0.1 were tested. In this step 36 specimens were divided into three stress values of 480, 500 and 520 MPa with three different levels of roughness. For the mathematical development, each source of uncertainty probability distribution was initially determined based on the statistical behavior characteristic of each variable. A mathematical model was derived to determine the uncertainty of the stress value and number of cycles, using the Matlab® software. The combined uncertainty of results were compared to values obtained by Kragten method, showing similar values between the two mathematical methods.

No description available.

A dissertation submitted to the Faculty of Engineering, University of the Witwatersrand, in fulfilment of the requirements for the degree of Master of Science in Engineering, Johannesburg, 1998 / This research examines fatigue in WC-Co, both under compressive and tensile loading conditions. A new macro-mechanism for compression fatigue crack propagation is put forward, which contradicts existing data on compression fatigue cracks as being self-limiting. Evidence of this macro-mechanism is presented in the form of final crack length versus number of cycles data, and micrographs of the compression fatigue cracks. A finite element study of the stress distribution in the WC-Co microstructure during compression fatigue loading has been developed. This model verifies possible methods of compression fatigue crack initiation. Examination of tensile fatigue and fast: fracture surfaces is used to show that fatigue is a separate mechanism to fast fracture in WC-Co. Characteristic features of the fatigue fracture surface are presented. A possible fatigue crack propagation mechanism is also presented. Finally, fatigue crack growth rate data in the form of the Paris equation is presented for WC-Co grades T6 and G6. / MT2017

No description available.