Understanding the aeroacoustic noise mechanisms and noise control techniques of roof rack systems

Aeroacoustics is one of the top NVH concerns in the automotive industry. HEV/EV have increased the challenge in rebalancing wind noise, and SDC is pushing sound quality requirements to be significantly more demanding than they are in conventional ICE manually-driven vehicles. The most severe aeroacoustic phenomena in ground vehicles are the ones with a tonal nature. Roof rack systems are directly exposed to the airflow and generate broadband noise and a discrete aeolian tone. Typical crossbar profiles are variations over an elliptical profile, i.e. not as blunt as a circular cylinder, neither as thin as a wing section, and a particular solution optimized for one profile will prove less effective in different designs. Therefore, the objective of this project is to investigate the noise mechanisms involved in elliptical crossbars through actual acoustic measurements taken on track. The first part of the project correlated exterior acoustic pressure and intensity measurements taken on track and in an aeroacoustic wind tunnel with the objective of assessing accuracy and repeatability. Exterior sound pressure on-track has demonstrated good accuracy in capturing both narrow and broadband noise effects, despite the uncontrolled background noise. The crossbar wake interaction with the roof plane was investigated through local flow visualization and reference aeroacoustic measurements. The second part of the project compared the noise generated by an elliptical cylinder with that generated by a circular cylinder and a NACA 0012 airfoil with the same thicknesses and at the same operational conditions. Results have shown that the elliptical crossbar noise characteristics have similarities when compared to those of blunt bodies at low Reynolds numbers and wing sections at higher speeds. Different leading and trailing edge geometries demonstrated that the trailing edge is the key contributor to the aeolian tone, while the leading edge affects primarily the broadband noise. Noise control techniques such as Angle of Attack and two and three-dimensional Boundary Layer Tripping (BLT) were investigated. Positive and negative incidence angles presented opposite trends towards noise reduction and have proven to be ineffective at higher speeds. 2D and 3D BLT did not eliminate the main tone but reduced its amplitude and bandwidth. 3D BLT techniques have demonstrated an advantage over 2D BLT. Innovative solutions such as Perforation and active Trailing Edge Blowing (TEB) were assessed. Both Perforation and TEB were effective in reducing the aeolian tone but presented side effects such as high frequency whistling, thus requiring further geometric optimization.