Roderick Galbraith

A Modified Dynamic Stall Model for Low Mach Numbers

The Leishman–Beddoes dynamic stall model is a popular model that has been widely applied in both helicopter and wind turbine aerodynamics. This model has been specially refined and tuned for helicopter applications, where the Mach number is usually above 0.3. However, experimental results and analyses at the University of Glasgow have suggested that the original Leishman–Beddoes model reconstructs the unsteady airloads at low Mach numbers less well than at higher Mach numbers. This is particularly so for stall onset and the return from the fully stalled state. In this paper, a modified dynamic stall model that adapts the Leishman–Beddoes dynamic stall model for lower Mach numbers is proposed. The main modifications include a new stall-onset indication, a new return modeling from stalled state, a revised chordwise force, and dynamic vortex modeling. The comparisons to the Glasgow University dynamic stall database showed that the modified model is capable of giving improved reconstructions of unsteady aerofoil data in low Mach numbers.

Control of Rotorcraft Retreating Blade Stall Using Air-Jet Vortex Generators

A series of low-speed wind tunnel tests was carried out on an oscillating airfoil fitted with two rows of air-jet vortex generators (AJVGs). The airfoil used had an Royal Aircraft Establishment 9645 section, and the two spanwise arrays of AJVGs were located at x/c = 0.12 and 0.62. The devices and their distributions were chosen to assess their ability to modify/control dynamic stall, the goal being to enhance the aerodynamic performance of helicopter rotors on the retreating blade side of the disk. The model was pitched about the quarter chord with a reduced frequency k of 0.1 in a sinusoidal motion defined by a = 15 + 10sin ωt deg. The measured data indicate that, for continuous blowing from the front row of AJVGs with a momentum blowing coefficient C μ greater than 0.008, modifications to the stalling process are encouraging. In particular, the pitching moment behavior exhibits delayed stall and there is a marked reduction in the normal force hysteresis.

Prediction of Dynamic Stall Onset for Oscillatory Low-Speed Airfoils

This research presents some common features of oscillatory airfoils, and the method for indicating dynamic stall onset for the unsteady process. Under deep stall conditions, the stall-onset angle in oscillation is independent of the mean angle of the oscillatory motion, and by combining the reduced frequency and the amplitude of the oscillatory motion, the equivalent reduced pitch rate is an analog of this motion to the constant reduced pitch rate of the ramp-up motion. By correlating with the measured data, and with the ramp-up results, the equivalent reduced pitch rate can be defined as a representation for the oscillatory motion. Accordingly, the triple-parameter problem of an oscillation (mean angle, reduced frequency, and amplitude) degrades into the single-parameter problem (equivalent reduced pitch rate). Based on these foundations, an extension of the stall-onset criterion is then made for oscillatory airfoils: a method of extracting the stall-onset parameters directly from oscillatory test data, and an indication of stall onset for the oscillatory airfoils. The results from the new proposed method have shown the consistency with the data of Glasgow University and the public data.

Parametric Study of an Air-Jet Vortex Generator Configuration to Control Rotorcraft Retreating Blade Stall

A series of low-speed wind tunnel tests were carried out on an oscillating airfoil fitted with two rows of air-jet vortex generators (AJVGs). The airfoil used had an RAE 9645 section and the two spanwise arrays of AJVGs were located at x/c=0.12 and 0.62. The devices and their distribution were chosen to assess their ability to modify/control dynamic stall; the goal being to enhance the aerodynamic performance of helicopter rotors on the retreating blade side of the disc. The model was pitched about the quarter chord with a reduced frequency (k) of 0.1 in a sinusoidal motion defined by a=15 o +10 o sinω t. The measured data indicate that, for continuous blowing from the front row of AJVGs with a momentum blowing coefficient (Cµ) greater than 0.008, modifications to the stalling process are encouraging. In particular, the pitching moment behavior exhibits delayed stall and there is a marked reduction in the normal force hysteresis.

An Examination of Key Aerodynamic Modeling Issues Raised by the NREL Blind Comparison

This paper describes some recent work undertaken in the aftermath of the ‘blind comparison’ with the NREL Unsteady Aerodynamics Experimental data collected in the NASA Ames wind tunnel. The data set collected in the NASA Ames tunnel represents a unique opportunity for aerodynamic modelers to enhance the capability of prediction schemes by comparison with ‘clean’ aerodynamic data. In this paper, the sensitivity of the results predicted by the Glasgow University prescribed wake model (HAWTDAWG) to a range of parameters including the blade section aerodynamic data is examined with reference to the measured data. In addition, specific modeling considerations highlighted by the measured data set are also discussed.Copyright

Analysis of Model Rotor Blade Pressures During Parallel Interaction with Twin Vortices

This paper presents and provides analysis of unsteady surface pressures measured on a model rotor blade as the blade experienced near parallel blade vortex interaction with a twin vortex system. To provide a basis for analysis, the vortex system was characterized by hot-wire measurements made in the interaction plane but in the absence of the rotor. The unsteady pressure response resulting from a single vortex interaction is then presented to provide a frame of reference for the twin vortex results. A series of twin vortex interaction cases are then presented and analyzed. It is shown that the unsteady blade pressures and forces are very sensitive to the inclination angle and separation distance of the vortex pair. When the vortex cores lie almost parallel to the blade chord, the interaction is characterized by a two-stage response associated with the sequential passage of the two cores. Conversely, when the cores lie on a plane that is almost perpendicular to the blade chord, the response is similar to that of a single vortex interaction. In all cases, the normal force response is consistent with the distribution of vertical velocity in the flow field of the vortex system. The pitching moment response, on the other hand, depends on the localized suction associated with the vortex cores as they traverse the blade chord.

A wind-tunnel based study of helicopter tail rotor blade vortex interaction

The interaction of a helicopter tail rotor blade with the tip vortex system from the main rotor is a significant source of noise and, in some flight states, can produce marked reductions in control effectiveness. This paper describes a series of wind-tunnel tests to simulate tail rotor blade vortex interaction with a view to providing data for the development and validation of numerical simulations of the phenomenon. In the experiments, which were carried out in the Argyll wind-tunnel of Glasgow University, a single-bladed rotor located in the tunnel’s contraction was used to generate the tip vortex which travelled downstream into the working section where it interacted with a model tail rotor. The tail rotor was instrumented with miniature pressure transducers that measured the aerodynamic response during the interaction. The results suggest that the rotor blade vortex interaction is similar in form to that measured at much higher spatial resolution on a fixed, non-rotating blade.

The effect of edge profile on delta wing flow

This paper presents flow measurements on four delta wing configurations which are differentiated by their leading edge profiles; sharp-edged, small, medium, and large radius. The experiments were performed as a part of the European Vortex Flow Experiment-2 campaign. Tests were conducted at speeds of 20.63 m/s and 41.23 m/s representing Reynolds numbers of 1 × 106 and 2 × 106, respectively. In this paper, oil flow visualization data are presented for the four wings together with particle image velocimetry results for the large radius wing. The study has identified interesting features of the interrelationship between the conventional leading edge primary vortex and the occurrence and development of the inner vortex on the round-edged delta wings. The effects of Reynolds number, angle of attack, and leading-edge radii on both vortex systems are discussed in detail.

Advancement of aerofoil section dynamic stall synthesis methods for rotor design

Dynamic stall is a complex process encountered by an aerofoil in the unsteady flow environment such as a helicopter rotor in forward flight as well as a fixed wing aircraft in manoeuvres and in other unsteady situations. The onset of dynamic stall effectively determines the flight envelope of the helicopter. Significant effort are being made to develop CFD to capture the dynamic stall behaviour, however traditional engineering models based on lifting line theory still offer fast turnround and broad understanding required for the rotor design process. This paper describes a new engineering model for dynamic stall, developed originally for wind turbine application at a typical Mach number of 0·12. The new dynamic stall model, with a better definition of stall onset, is based on improvements made to Beddoes’ original trailing edge stall model. This paper will describe and demonstrate the improvements in identifying both the stall-onset and the pitching moment break at high pitch rates, when being applied to a generic rotor aerofoil RAE 9651 at M = 0·3. Further validation against oscillatory tests and other Mach numbers are still required. However the study has provided sufficient confidence for it to be employed in a rotor analysis code. NOMENCLATURE a co-ordinate of pitching axis in terms of half chord (a = x/b) b half chord (b = c/2) c chord length CC chordwise force coefficient THE AERONAUTICAL JOURNAL MAY 2012 VOLUME 116 NO 1179 521 Paper No. 3706. Manuscript received 15 March 2011, revised version received 27 July 2011, second revised version received 28 November 2011, accepted 15 December 2011. Paper originally presented at the HeliJapan 2010, Tokyo, Japan, 1-3 November 2010 CD drag coefficient Cm pitching moment coefficient ΔCmv pitching moment due to vortex CN normal force coefficient CNmin normal force coefficient local minimum before full re attachment ΔCNv additional normal force due to vortex lift Cp pressure coefficient |Cp| rise pressure coefficient suction rise at the leading edge (normally 2·5% chord) f separation location in terms of chord length f ′, f ′′ delayed separation function of f M Mach number r reduced pitch rate (r = cα/2V) s,τ non-dimensional time (s = τ = 2Vt/c) t time parameter Tf delay constant for separation point due to dynamic effect Tp delay constant of attaining same CN in a delayed leading-edge-pressure under unsteady condition Tr delay constant for reattachment process Tv time constant of vortex travelling over chord TVL vortex passage time constant Tα delay time constant for angle-of-attack due to dynamic effect V free stream velocity Vx shape function of normal force due to vortex wa flow speed on the aerofoil surface perpendicular to chord x co ordinate along chord centred at half chord x* dimensionless value of x in terms of half-chord (x = x/b) za flow displacement perpendicular to chord α angle-of-attack or incidence α0 angle-of-attack of zero normal force or mean angle-of-attack of oscillation of aerofoil α1 breakpoint of separation αcr critical stall-onset angle-of-attack αds stall-onset angle-of-attack αds0 constant critical stall-onset angle-of-attack αmin angle-of-attack at CNmin αmin0 angle-of attack at CNmin for static test αss static stall-onset angle-of-attack θ co-ordinate transformation, θ = Cos(x/b) κ reduced frequency (κ = ωc /2V) ηm forcing for circulatory pitching moment ηn forcing for circulatory normal force λ coefficient of least square fit λm forcing for non-circulatory pitching moment λn forcing for non-circulatory normal force τv non-dimensional time during vortex passage ω circular frequency of oscillation of aerofoil Δ a step change in forcing or in time 522 THE AERONAUTICAL JOURNAL MAY 2012

The effect of leading edge radius on a 65° delta wing

delta wing model with three sets of leading edges; sharp, medium radius and large radius. The model was originally tested with sharp and medium radius edges as part of the International Vortex flow Experiment 2 (VFE -2). Recently, the model was tested again at 1x10 6 and 2x10 6 Reynolds numbers in the Glasgow University 2.65 x 2.04 meter Argyll Wind Tunnel with large and small radius leading edges. During the experiments, oil flow visualization and load cell measurements were carried out. In this paper, the results from this study are presented along with the earlier results to examine three different effects; leading edge bluntness, Reynolds number and angle of attack. It is shown that the leading edge shape has a significant effect on the flow topology that is also manifest in the forces and moments on the wing.