Drag Crisis Flow Past a Circular Cylinder Elizabeth

Drag Crisis: Flow Past a Circular Cylinder Elizabeth Browne, Kenneth Passmore Mentor: Ashis Nandy College of Engineering and Technology Northern New Mexico College Abstract Using a subsonic wind tunnel, flow past a single circular cylinder with and without artificial means of boundary layer tripping has been investigated. Aerodynamic characteristics and resulting fluid forces at different flow regimes were studied with a special attention in the critical Reynolds number range (Re). A 100 -mm-diameter cylinder model was used and the Reynolds number was varied from ~ 4 x 104 to ~2. 1 x 105. Mean surface pressure distribution over the circumference of the cylinder at the mid-span section was measured using manometers and from these results the pressure drag and lift coefficients were calculated. The phenomenon of sudden drop of drag coefficient in the critical flow state or “drag crisis” was studied by varying the angular position of two symmetrically placed, 0. 5 mm-diameter tripwires from 37. 50 to 82. 50 (from front stagnation point). Motivation Flow past bluff bodies has been researched for decades, primarily due to numerous applications, including: fluid - structure interactions, aerodynamics of both sports balls and automobiles, and hydrodynamic forces on submerged bodies. The aerodynamic and hydrodynamic behavior in these applications depend mainly on the boundary layer nature and dynamics. Circular cylinders, due to their geometric simplicity, are used extensively as a model to study these boundary layer dynamics and their implications on the aerodynamic properties, such as drag and lift forces Background Research The real flow around a cylinder differs from the ideal inviscid (frictionless) flow mainly due to the presence of the boundary layer. The flow pattern and characteristics depend on the Reynolds number. At subcritical Re (~ < 2 x 105), the laminar boundary layer flow separates at around 80 o (from the front stagnation point) creating a wide wake region, resulting in larger drag. At higher Re (~ 3 - 5 x 105) the boundary layer becomes turbulent and due to higher energy it can follow the adverse pressure gradient around the cylinder a little bit longer and separates later (~ 120 o) causing narrower wake and less drag coefficient 1. Experimental Methods Experiments were conducted in Aerolab Educational Wind Tunnel (EWT) which includes instrumentation, data acquisition and probes. The data acquisition software (Aeroware) is able collect and plot a variety of data, including: pressure, velocity, time, and frequency. There is also an integral multi-manometer with 24 vertical tubes for measuring pressure (Shown below). Results and Discussion Smooth cylinder results (without tripwire) Important Tunnel Specification: Test section dimension 12 in x 24 in Airspeed Range Turbulence level 10 mph - 145+ mph (4 m/s - 65+ m/s) < 0. 2 % Subsonic wind tunnel The Cp distribution at various Re values are quite similar for the smooth cylinder without the tripwires. The back pressure is found to be quite low (~ -2. 5 to -2. 0) and the separation point is close to the shoulder of the cylinder (~ 90 o). The flow on both sides remain more or less symmetric, resulting in very little lift (low CL values). The drag coefficient remains more or less constant at ~1. 75 which is considerably higher than the value found in literature 1. This could be attributed to the high blockage ratio (d/h ~ 33%) and the resulting interference effect from the tunnel walls. A smooth aluminum cylinder model (4 in or 10. 2 cm diameter) was mounted vertically which spans the test section of the tunnel. The cylinder consists of 24 flush mounted and equispaced (15 o apart) pressure taps around the circumference at the mid height of the cylinder, which are connected to the manometer tubes with flexible tubing. The blockage ratio d/h is equal to 33%. Pressure Cylinder with trip wire Cylinder with tripwires and comparison with the smooth cylinder Multimanometer during a test run Experiments were conducted first with the smooth cylinder model at varying flow speed (or Re). From manometer readings, the pressure coefficients, Cp around the cylinder is found using: Where, Pi : pressure at a pressure tap ρ: air density : Freestream pressure : Freestream velocity The pressure coefficients are then integrated numerically over the cylinder circumference to compute the resulting drag coefficient (only pressure drag is computed as for bluff bodies at high Reynolds number the skin friction drag is only ~ 5% of the total drag) §Overall, the CD values are considerably lower (specially at higher values of Re) for cases with the tripwires as compared to the no trip (smooth cylinder) case. §In general, for cases with tripwires, as the Re value is increased, the separation points move to a higher angle causing narrower wake region and correspondingly lower CD. Almost no such effect is observed for the no trip case. Thus boundary layer tripping and critical flow is induced for cases with tripwires, but only subcritical flow is observed for the smooth cylinder. §At low values of Re (e. g. Re ~ 6. 3 x 104), tripwires are more effective at a relatively higher angle (compare the back pressure for the 82. 5 degree case with the no trip case) §As the Re value is increased, the most effective location for the tripwires is shifted towards lower angle (e. g. at Re ~ 1. 4 x 105, 52. 5 degree case shows the highest back pressure and correspondingly the lowest CD, whereas at Re ~ 1. 9 x 105, the highest back pressure and correspondingly the lowest CD occurs for the 37. 5 degree case) References Where, ϴi is the angle of the ith pressure port from the front stagnation point and ∆ϴi is the spacing between two pressure ports Subcritical flow and Laminar boundary layer separation at ~80 o Supercritical flow and Turbulent boundary layer separation at ~120 o Boundary layer tripping were attempted at a lower Re value by using two symmetrically placed trip wires of diameter 0. 5 mm. The angles used were 37. 5 o, 52. 5 o, 67. 5 o and 82. 5 o. Pressure measurements were carried out at varying Re and compared with the no trip case. CD values are obtained in the same manner as described above. 1. E. Achenbach. Distribution of local pressure and skin friction around a circular cylinder in cross-flow up to Re = 5 x 106. J. Fluid Mech. vol. 34, part 4, pp. 625 -639. Great Britain. 1968. 2. Md. Mahbub Alam, Y. Zhou, J. M. Zhao, O. Flamand, and O. Boujard. Classification of the tripped cylinder wake and bi-stable phenomenon. International Journal of Heat and Fluid Flow. Vol. 31. Issue 4. pp. 545560. 2010. 3. Elmar Achenbach. Influence of surface roughness on the cross-flow arounda circular cylinder. J Fluid Mech. vol. 46, part 2, pp 321 -335. Great Britain. 1971. 4. Suresh Behara and Sanjay Mittal. Transition of the boundary layer on a circular cylinder in the presence of a trip. Journal of Fluids and Structures. Vol. 27, Issues 5 -6. pp. 702 -715. 2011. 5. Yunus A. Cengel and John M. Cimbala. Fluid Mechanics: Fundamentals and Applications, Third Edition. Mc. Graw Hill. 2014. 6. William J. Devenport and Aurelien Borgoltz. Experiment 3: Flow Past a Circular Cylinder. Virginia Tech. Web. 2016. 7. Education Design Division. Educational Wind Tunnel: Operations Manual. Aero. Ware. Aero. Lab. 2017. Acknowledgements Cylinder with Trip wires For more information contact: Dr. Ashis Nandy: ashis@nnmc. edu This work was partly supported by the EDUCERE grant (NSF IUSE Program).