AJ
A.A. Jensen
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
1 records found
1
Surface dimples are a well-established means of reducing drag on bluff bodies, most famously the golf ball, where tripping the boundary layer into turbulence delays the flow separation and narrows the wake, reportedly halving drag relative to a smooth sphere. Whether the same mechanism benefits a lifting surface is far less settled. An airfoil in attached flow has little separation to delay, so a dimple that helps a sphere can just as easily add drag to a wing. Existing studies on dimpled airfoils reflect this tension, reporting inconsistent benefits that depend strongly on dimple depth, size, and chordwise location, with no overwhelming consensus on where along the chord dimples should be placed. This thesis investigates how the controlled variation of dimple depth and chordwise placement affects the aerodynamic characteristics and flow-separation behaviour of a lifting surface across varying operating conditions, and develops a proof-of-concept prototype capable of mechanically varying dimple depth as a first step toward a fully adaptive surface.
A NACA 63-415 airfoil with chord c = 200 mm was tested with three chordwise dimple coverage families, mid (0.33c to 0.59c), aft (0.59c to 0.85c), and full (0.33c to 0.85c) suction-side coverage, each at three dimple depths, across three Reynolds numbers (1*10^5 , 2*10^5, and 5*10^5). Measurements spanned an angle-of-attack range of−10° to 20°, with 0° to 16° forming the most substantial part of the campaign. The campaign combined wake-rake and wall-pressure measurements, surface-pressure taps, a force balance, particle image velocimetry (PIV), and tuft flow visualisation across two wind-tunnel facilities, the Red Wind Tunnel at DTU and the Small Low-Turbulence Tunnel at TU Delft, supported by RANS and LES simulations.
The results show that dimple depth, chordwise location, and Reynolds number interact strongly, and no single fixed combination performs well across the full envelope tested. At the lowest Reynolds number, mid and full coverage produce substantial glide-ratio gains, up to 132% over the smooth baseline, driven primarily by drag reduction from delayed separation. At the highest Reynolds number, this reverses entirely. Mid coverage becomes uniformly detrimental, while aft coverage becomes the only layout to show any benefit, concentrated near stall. Dimple depth shows a similarly reversing, non-monotonic effect. The depth that gives the largest drag reduction at the lowest Reynolds number is the most harmful choice at the highest, and near stall this non-monotonicity appears even within a single Reynolds number, where a depth of 2.50 mm produces a genuine drag reduction while both a shallower (1.25 mm) and a deeper (5.00 mm) variant do not. PIV, surface-pressure, and wake-rake measurements on the standout mid configuration confirm the underlying mechanism directly. The dimples energise the near-wall boundary layer, suppress separation that is otherwise present on the smooth baseline, and narrow the wake to roughly half the width of its baseline counterpart, extending to a measurable stall-delay effect at a higher Reynolds number. RANS and LES reproduce the broad direction of these trends but not the mechanism itself in detail, consistent with the RANS model’s lack of a transition formulation.
These findings show that no single, fixed dimple geometry is optimal across a realistic operating envelope, motivating a surface that can adjust its own geometry in response to the flow condition. A proof-of-concept prototype was designed and built around a pneumatically actuated, depth-adjustable membrane, and structured-light scanning related each actuation setting to a physical dimple depth. Wind-tunnel testing at Re = 1*10^5 showed the prototype reproducing the drag behaviour of the static configurations in trend, with a peak drag reduction of approximately 22% relative to its own unactuated state, and reproducing the same suction-side wake narrowing at matched depth. The correspondence is qualitative rather than exact, and no lift benefit is observed, but a single airfoil is shown to select between aerodynamic states that individually resemble distinct static geometries, establishing the feasibility of an adaptive dimpled surface. ...
A NACA 63-415 airfoil with chord c = 200 mm was tested with three chordwise dimple coverage families, mid (0.33c to 0.59c), aft (0.59c to 0.85c), and full (0.33c to 0.85c) suction-side coverage, each at three dimple depths, across three Reynolds numbers (1*10^5 , 2*10^5, and 5*10^5). Measurements spanned an angle-of-attack range of−10° to 20°, with 0° to 16° forming the most substantial part of the campaign. The campaign combined wake-rake and wall-pressure measurements, surface-pressure taps, a force balance, particle image velocimetry (PIV), and tuft flow visualisation across two wind-tunnel facilities, the Red Wind Tunnel at DTU and the Small Low-Turbulence Tunnel at TU Delft, supported by RANS and LES simulations.
The results show that dimple depth, chordwise location, and Reynolds number interact strongly, and no single fixed combination performs well across the full envelope tested. At the lowest Reynolds number, mid and full coverage produce substantial glide-ratio gains, up to 132% over the smooth baseline, driven primarily by drag reduction from delayed separation. At the highest Reynolds number, this reverses entirely. Mid coverage becomes uniformly detrimental, while aft coverage becomes the only layout to show any benefit, concentrated near stall. Dimple depth shows a similarly reversing, non-monotonic effect. The depth that gives the largest drag reduction at the lowest Reynolds number is the most harmful choice at the highest, and near stall this non-monotonicity appears even within a single Reynolds number, where a depth of 2.50 mm produces a genuine drag reduction while both a shallower (1.25 mm) and a deeper (5.00 mm) variant do not. PIV, surface-pressure, and wake-rake measurements on the standout mid configuration confirm the underlying mechanism directly. The dimples energise the near-wall boundary layer, suppress separation that is otherwise present on the smooth baseline, and narrow the wake to roughly half the width of its baseline counterpart, extending to a measurable stall-delay effect at a higher Reynolds number. RANS and LES reproduce the broad direction of these trends but not the mechanism itself in detail, consistent with the RANS model’s lack of a transition formulation.
These findings show that no single, fixed dimple geometry is optimal across a realistic operating envelope, motivating a surface that can adjust its own geometry in response to the flow condition. A proof-of-concept prototype was designed and built around a pneumatically actuated, depth-adjustable membrane, and structured-light scanning related each actuation setting to a physical dimple depth. Wind-tunnel testing at Re = 1*10^5 showed the prototype reproducing the drag behaviour of the static configurations in trend, with a peak drag reduction of approximately 22% relative to its own unactuated state, and reproducing the same suction-side wake narrowing at matched depth. The correspondence is qualitative rather than exact, and no lift benefit is observed, but a single airfoil is shown to select between aerodynamic states that individually resemble distinct static geometries, establishing the feasibility of an adaptive dimpled surface. ...
Surface dimples are a well-established means of reducing drag on bluff bodies, most famously the golf ball, where tripping the boundary layer into turbulence delays the flow separation and narrows the wake, reportedly halving drag relative to a smooth sphere. Whether the same mechanism benefits a lifting surface is far less settled. An airfoil in attached flow has little separation to delay, so a dimple that helps a sphere can just as easily add drag to a wing. Existing studies on dimpled airfoils reflect this tension, reporting inconsistent benefits that depend strongly on dimple depth, size, and chordwise location, with no overwhelming consensus on where along the chord dimples should be placed. This thesis investigates how the controlled variation of dimple depth and chordwise placement affects the aerodynamic characteristics and flow-separation behaviour of a lifting surface across varying operating conditions, and develops a proof-of-concept prototype capable of mechanically varying dimple depth as a first step toward a fully adaptive surface.
A NACA 63-415 airfoil with chord c = 200 mm was tested with three chordwise dimple coverage families, mid (0.33c to 0.59c), aft (0.59c to 0.85c), and full (0.33c to 0.85c) suction-side coverage, each at three dimple depths, across three Reynolds numbers (1*10^5 , 2*10^5, and 5*10^5). Measurements spanned an angle-of-attack range of−10° to 20°, with 0° to 16° forming the most substantial part of the campaign. The campaign combined wake-rake and wall-pressure measurements, surface-pressure taps, a force balance, particle image velocimetry (PIV), and tuft flow visualisation across two wind-tunnel facilities, the Red Wind Tunnel at DTU and the Small Low-Turbulence Tunnel at TU Delft, supported by RANS and LES simulations.
The results show that dimple depth, chordwise location, and Reynolds number interact strongly, and no single fixed combination performs well across the full envelope tested. At the lowest Reynolds number, mid and full coverage produce substantial glide-ratio gains, up to 132% over the smooth baseline, driven primarily by drag reduction from delayed separation. At the highest Reynolds number, this reverses entirely. Mid coverage becomes uniformly detrimental, while aft coverage becomes the only layout to show any benefit, concentrated near stall. Dimple depth shows a similarly reversing, non-monotonic effect. The depth that gives the largest drag reduction at the lowest Reynolds number is the most harmful choice at the highest, and near stall this non-monotonicity appears even within a single Reynolds number, where a depth of 2.50 mm produces a genuine drag reduction while both a shallower (1.25 mm) and a deeper (5.00 mm) variant do not. PIV, surface-pressure, and wake-rake measurements on the standout mid configuration confirm the underlying mechanism directly. The dimples energise the near-wall boundary layer, suppress separation that is otherwise present on the smooth baseline, and narrow the wake to roughly half the width of its baseline counterpart, extending to a measurable stall-delay effect at a higher Reynolds number. RANS and LES reproduce the broad direction of these trends but not the mechanism itself in detail, consistent with the RANS model’s lack of a transition formulation.
These findings show that no single, fixed dimple geometry is optimal across a realistic operating envelope, motivating a surface that can adjust its own geometry in response to the flow condition. A proof-of-concept prototype was designed and built around a pneumatically actuated, depth-adjustable membrane, and structured-light scanning related each actuation setting to a physical dimple depth. Wind-tunnel testing at Re = 1*10^5 showed the prototype reproducing the drag behaviour of the static configurations in trend, with a peak drag reduction of approximately 22% relative to its own unactuated state, and reproducing the same suction-side wake narrowing at matched depth. The correspondence is qualitative rather than exact, and no lift benefit is observed, but a single airfoil is shown to select between aerodynamic states that individually resemble distinct static geometries, establishing the feasibility of an adaptive dimpled surface.
A NACA 63-415 airfoil with chord c = 200 mm was tested with three chordwise dimple coverage families, mid (0.33c to 0.59c), aft (0.59c to 0.85c), and full (0.33c to 0.85c) suction-side coverage, each at three dimple depths, across three Reynolds numbers (1*10^5 , 2*10^5, and 5*10^5). Measurements spanned an angle-of-attack range of−10° to 20°, with 0° to 16° forming the most substantial part of the campaign. The campaign combined wake-rake and wall-pressure measurements, surface-pressure taps, a force balance, particle image velocimetry (PIV), and tuft flow visualisation across two wind-tunnel facilities, the Red Wind Tunnel at DTU and the Small Low-Turbulence Tunnel at TU Delft, supported by RANS and LES simulations.
The results show that dimple depth, chordwise location, and Reynolds number interact strongly, and no single fixed combination performs well across the full envelope tested. At the lowest Reynolds number, mid and full coverage produce substantial glide-ratio gains, up to 132% over the smooth baseline, driven primarily by drag reduction from delayed separation. At the highest Reynolds number, this reverses entirely. Mid coverage becomes uniformly detrimental, while aft coverage becomes the only layout to show any benefit, concentrated near stall. Dimple depth shows a similarly reversing, non-monotonic effect. The depth that gives the largest drag reduction at the lowest Reynolds number is the most harmful choice at the highest, and near stall this non-monotonicity appears even within a single Reynolds number, where a depth of 2.50 mm produces a genuine drag reduction while both a shallower (1.25 mm) and a deeper (5.00 mm) variant do not. PIV, surface-pressure, and wake-rake measurements on the standout mid configuration confirm the underlying mechanism directly. The dimples energise the near-wall boundary layer, suppress separation that is otherwise present on the smooth baseline, and narrow the wake to roughly half the width of its baseline counterpart, extending to a measurable stall-delay effect at a higher Reynolds number. RANS and LES reproduce the broad direction of these trends but not the mechanism itself in detail, consistent with the RANS model’s lack of a transition formulation.
These findings show that no single, fixed dimple geometry is optimal across a realistic operating envelope, motivating a surface that can adjust its own geometry in response to the flow condition. A proof-of-concept prototype was designed and built around a pneumatically actuated, depth-adjustable membrane, and structured-light scanning related each actuation setting to a physical dimple depth. Wind-tunnel testing at Re = 1*10^5 showed the prototype reproducing the drag behaviour of the static configurations in trend, with a peak drag reduction of approximately 22% relative to its own unactuated state, and reproducing the same suction-side wake narrowing at matched depth. The correspondence is qualitative rather than exact, and no lift benefit is observed, but a single airfoil is shown to select between aerodynamic states that individually resemble distinct static geometries, establishing the feasibility of an adaptive dimpled surface.