As the automotive industry moves towards greater intelligence, electric tailgate systems have seen widespread adoption, featuring remote control, obstacle detection, and intelligent opening functions that significantly enhance the user experience. The electric telescopic rod, as a key actuator, has drawn attention for its structural and transmission design. However, studies have shown that during actual operation, various noise issues arise with electric telescopic rods, affecting the sound quality and smoothness of the tailgate's opening and closing. This paper presents a noise detection and analysis study based on a dedicated testbench platform specifically developed for electric telescopic rods. The platform was designed to simulate the real-world opening and closing process of automotive tailgates, enabling a controlled environment for capturing and analyzing noise characteristics effectively. Using a microphone to capture noise signals, three main types of noise were identified: high-frequency electromagnetic noise, low-frequency impact noise, and periodic structural noise. The study accurately detects high-frequency electromagnetic noise using wavelet packet energy feature extraction techniques and identifies and localizes low-frequency impact noise using a sliding window dynamic threshold method. It also distinguishes this noise from testbench noise through frequency domain characteristics. Additionally, autocorrelation analysis was employed to detect and evaluate periodic noise. The experimental results demonstrate that these methods effectively identify and localize various noise sources, offering valuable data to support future noise source localization and acoustic optimization of electric telescopic rods.