The formation of scour patterns at bridge piers is driven by the forces at the boundary of the water flow. In most experimental scour studies, indirect processes have been applied to estimate the shear stress using measured velocity profiles. The estimations are based on theoretical models and associated assumptions. However, the turbulence flow fields and boundary layer in the pier-scour region are very complex and lead to low-fidelity results. In addition, available turbulence models cannot account accurately for the bed roughness effect.

Direct measurement of the boundary shear stress, normal stress, and their fluctuations are attractive alternatives. However, most direct-measurement shear sensors are bulky in size or not compatible to fluid flow.

A sensor has been developed that consists of a floating plate with folded beam support and an optical grid on the back, combined with a high-resolution optical position probe. The folded beam support makes the floating plate more flexible in the sensing direction within a small footprint, while maintaining high stiffness in the other directions. The floating plate converts the shear force to displacement, and the optical probe detects the plate’s position with nanometer resolution by sensing the pattern of the diffraction field of the grid through a glass window. This configuration makes the sensor compatible with liquid flow applications.

Most shear boundary fluid sensors using a direct measurement method include a floating plate and a position sensor. The plate moves under the shear force. To obtain high sensitivity, the floating part of the plate is supported with a structure flexible in the sensing direction and stiff in other directions. The structure could be in plane with the plate or out of plane. The in-plane support structure has an advantage to be fabricated by micromachining technology. The flexible support requires long beams and results in a large footprint. This approach applied a folded beam support to the floating plate design. The folded beams allow a much smaller footprint to achieve the same flexibility as the conventional configuration.

The breadboard sensor was tested in a water tank testbed. The testbed includes a “false bottom” plate and a running conveyor belt. The false bottom plate simulates the riverbed, and provides an opening for attaching the sensor mounting plate and an air pocket for separating the encoder from the water. The running belt is placed parallel to the direction of measurement, and is placed on a structure different than the false bottom for vibration isolation. The belt was running at controlled speed of up to 2 m/s in both forward and opposite directions, about 38 mm above the false bottom plate where the sensor was mounted. An average position change is about 620 nm, corresponding to a change in direction of the boundary shear stress. The fluctuation is believed to be caused by the turbulence of the flow.

Another flexure with a 3D configuration was proposed as well. The 3D version was designed to allow for a plate to rotate about an axis defined by three sets of flexures. The plate is mounted on a beam that passes through this axis and extends on the other side of the axis where it has a free end. An optical or capacitive sensor would read the displacement of the free end of the beam, which would be used with the flexure’s stiffness to calculate the shear force on the sensor plate. This configuration could use the horizontal blade to separate the test media, such as water, from the optical or capacitive sensor, but would present more manufacturing and packaging challenges.

This work was done by Xiaoqi Bao, Mircea Badescu, Stewart Sherrit, Yoseph Bar-Cohen, Shyh-Shiuh Lih, Zensheu Chang, and Brian P. Trease of Caltech; Kornel Kerenyi of the Federal Highway Administration; and Scott E. Widholm and Patrick N. Ostlund of Cal Poly Pomona for NASA’s Jet Propulsion Laboratory. NPO-47812



This Brief includes a Technical Support Package (TSP).
Document cover
Sensor for Boundary Shear Stress in Fluid Flow

(reference NPO-47812) is currently available for download from the TSP library.

Don't have an account? Sign up here.