Overrunning spring clutches are widely used as essential transmission devices, and the occurrence of abnormal noise can lead to a decline in their performance. This study investigates the dynamic aspects of abnormal noise in engineering applications, including its causes, influencing factors, and suppression methods. Audio processing algorithms are employed to analyze the audio associated with abnormal noise, and the Fourier Motion Blur algorithm is applied to process video images of the springs. By combining the motion blur curve with the noise spectrum curve, the source of the abnormal noise is identified as friction-induced vibrations in the spring. Theoretical modeling and calculations are carried out from a dynamic perspective to validate that the phenomenon of abnormal noise in the clutch is a result of self-excited friction vibration caused by the stick–slip phenomenon. Based on theoretical analysis and practical engineering, surface texturing is added to the center shaft of the spring seat, optimizing the system as an overdamped system to suppress self-vibration. Utilizing CFD simulation analysis, the simulation results are used to improve the texturing parameters and further optimize the texturing shape, resulting in an optimal parallelogram surface texture structure. Experimental validation confirms that the improved overrunning spring clutch completely eliminates abnormal noise during overrunning operation. Therefore, this paper contributes to the understanding of the dynamic issues associated with abnormal noise in overrunning spring clutches, confirming that the mechanism for abnormal noise generation is friction-induced self-excitation vibration, and demonstrating that surface texture optimization methods effectively suppress the occurrence of abnormal noise.