It is common knowledge that a bolt is apt to loosen due to slippage between the contact surfaces of joined parts. Loosening tests using real parts enable precise scrutiny of real phenomena under the influence of multiple factors such as slip distance, surface roughness, and coefficient friction. However, estimating the influence of the individual factors is very difficult because the friction forces of real contact surfaces are compiled based on variations in friction coefficients, meaning friction is not stable. Therefore, the effects of factors were investigated using Finite Element Analysis (FEA) to control friction coefficients. The procedures were as follows. Assuming a joined structure consisting of a bolt, nut, and washer, bolt axial tension was generated through constant movement of a washer in the bolt’s axial direction, following which the washer was constantly vibrated in one direction transverse to the bolt axis. This vibration generated displacements equivalent to the degree of slippage between the two clamped parts. During vibration, the rotating angles of the bolt and the contact pressure of the threads and bearing surfaces were calculated. The results were as follows. The vibrating displacements of a washer have considerable influence on the rotational loosening of a bolt. In cases where there was only minor displacement of the washer vibrations, the rotational loosening angle rapidly decreased, although the loosening did not cease completely. Therefore, the magnitude of what is called “critical slip” was not confirmed under the conditions of this study. In addition, the friction coefficient has a significant influence on the rotational loosening of a bolt. When the respective friction coefficient values of the threads and bearing surfaces are not balanced, rotational loosening cannot continue. Surface roughness readily affects contact pressure, so it tends to make the contact pressure localized. In particular, high-pressure areas were affected by several projections set on the threads. However, under those conditions the rotational loosening did not differ greatly from the results of the fine surface models subject to the same vibrating amplitude and friction coefficient. Consequently, the localized contact pressure had little evident effect on loosening. Above all, FEA reproduced the loosening of the bolt, and the reference made in this analysis is useful.
