Cassio Eduardo Lima de Paiva

Quantification of the Vertical Load Applied to the Pavement during Braking Maneuver of a Commercial Vehicle

AbstractFront and rear axle braking performance in a vehicle is a well-documented subject in reference to longitudinal forces, but this statement is not valid for vertical forces. To analyze this behavior, a commercial vehicle (equipped with an anti-lock braking system) was instrumented with strain gauges and one accelerometer for testing in a closed course proving ground located in Brazil. By using the vehicle’s springs as load cells, the vertical load applied to the pavement was evaluated during different braking maneuvers. Using a bidirectional ANOVA and a post hoc Tukey’s honest significant differences (HSD) test, statistical differences of the transferred load were evaluated between the vehicle’s initial speed and brake pedal position. Although it is recommended to repeat the same procedure on different vehicles and varying pavements surfaces, the results indicate that, unlike the brake pedal position, the vehicle’s initial speed does not influence the vertical load applied to the pavement during bra...

PROPOSTA DE METODOLOGIA PARA USAR MEDIÇÕES REAIS NA REALIMENTAÇÃO DE ANÁLISES DE ENGENHARIA AUXILIADA POR COMPUTADOR

Within a new scenario in the product development cycle, led by an integrated and synchronized task approach, the combination of testing and simulation is a visible tendency even on the design stage of a component. In this regard, the main objective of this paper is to propose a methodology of how real (physic) measurements can be used as feedback loops in computer-aided engineering (CAE) analysis. For this purpose, a truck’s suspension component, instrumented with strain gauges on its hotspots, has been tested in a proving ground track located in Brazil. Using an existing durability commercial software, virtual strain gauges have been placed in the component’s finite element model and, with the measured forces inputs, theirs results were compared to the correspondents gauges on the real component. In the case study presented, the proposed methodology reduced the average error of simulation damage results from 58% to 30% on the tested component. This result suggests that this method is capable to offer good enhancements in CAE virtual analysis, providing calibrated finite elements models and consequently more reliable virtual verifications. Keywords: Computer-aided engineering, Strain gauges, Finite element analysis, Real measurements. RESUMO Em um novo cenario no ciclo de desenvolvimento de produto, caracterizado por uma abordagem sincronizada de atividades, a combinacao de testes fisicos e simulacao e uma tendencia observada ate mesmo nas fases inicias de concepcao de um componente. Dessa forma, o objetivo deste trabalho e propor uma metodologia de como medicoes reais (fisicas) podem ser usadas na realimentacao de analises de engenharia auxiliada por computador. Para isto, um componente da suspensao de um caminhao, instrumentado em seus pontos de maior tensao mecânica, foi testado em um campo de provas situado no Brasil. Usando um software comercial de durabilidade, extensometros virtuais foram posicionados no modelo de elementos finitos do componente e, com as entradas de forcas medidas, os seus resultados foram comparados aos dos sensores reais, instrumentados na peca fisica. No estudo de caso apresentado, a metodologia proposta reduziu o erro medio dos resultados numericos de dano de 58% para 30% no componente testado. Tal resultado sugere que a metodologia proposta e capaz de proporcionar melhoras significativas em atividades de simulacao, disponibilizando modelos numericos calibrados e consequentemente verificacoes virtuais mais confiaveis. Palavras-chave: engenharia auxiliada por computador, extensometros, analise de elementos finitos, medicoes reais.