Detection of defects in steel materials by vibration analysis

A variety of examination methods have been used for detecting seen and unseen defects in materials. In this work, detection of defects in beams is based upon the use of frequency analysis of longitudinal vibration data First of all, mathematical models and equations of motions for longitudinal vibration of the healthy and defected beams have been created in order to permit the analytical calculations of vibration characteristics (i.e. mode shapes and natural frequencies) of considered beams. During vibration analysis, the beam in question is considered as a multi-degree of freedom system by considering discrete mass model. In this model, the beam is divided into equal pieces and each piece is thought to be connected to each other by massless springs. After that, a computer programme is coded in Matlab which enables calculations of vibration characteristics of a beam for healthy and defected conditions and it is found that the results obtained from the use of discrete mass model converge to analytical results when the number of considered mass is selected reasonably large. To prove accuracy of the analytical results experimentally, a steel beam whose diameter and length are 20.1mm and 3m respectively is used and its free longitudinal vibrations for health and defected conditions are measured under free-free end conditions. It is found when the analytical and experimental results are compared that discrete mass model approximation can reliably be used for revealing the presence of fault in beams.