THE AMERICAN SOCIETY OF MECHANICAL ENGINEERS 345 E. 47th St., New York, N.Y

Transcription

1 THE AMERICAN SOCIETY OF MECHANICAL ENGINEERS 345 E. 47th St., New York, N.Y The Society shall not be responsible for statements or opinions advanced in papers or discussion at meetings of the Society or of its Divisions or Sections, or printed In its publications. Discussion is printed only If the paper is published In an ASME Journal. Papers are available from ASME for 15 months after the meeting. Printed in U.S.A. Copyright 1994 by ASME 94-GT 408 THE FLIGHT TESTING OF NATURAL AND HYBRID LAMINAR FLOW NACELLES Brian Barry and Simon J. Parke Advanced Engineering, Aerospace Group Rolls-Royce plc Derby, United Kingdom Nicholas W. Bown Dept. of Engineering Science University of Oxford Oxford, United Kingdom Hansgeorg Riedel and Martin Sitzmann Institution for Design Aerodynamics Fluid Dynamics, DLR Braunschweig, Germany III g ABSTRACT The achievement of large areas of laminar flow over aircraft engine nacelles offers significant savings in aircraft fuel consumption. Based upon current engine configurations nett sfc benefits of up to 2% are possible. In addition the engine nacelle is ideally suited to the early inclusion of laminar flow technology,being relatively self contained with the possibility of application to existing airframes. In September 1992 a European Consortium managed by Rolls-Royce including MTU and DLR began flight testing of a natural laminar flow nacelle. This programme was later extended by R-R and DLR to flight test a hybrid laminar flow nacelle featuring boundary layer suction and insect contamination protection. The tests evaluated the effects of flight and engine environment, boundary layer transition phenomena, suction system operation and insect contamination avoidance strategies. This paper describes the global conclusions from these flight tests which are a significant milestone leading to the future application of laminar flow technology to engine nacelles. 1.0 INTRODUCTION Aerodynamic drag is responsible for a substantial portion of the operating cost of a civil transport aircraft: estimates for typical projects suggest that the achievement of laminar flow over all external surfaces could reduce the total drag by 30%. Contributions made by the powerplant are typically 5-6% of aircraft drag, with 60-70% of the total external drag of a conventional nacelle in cruise flight resulting from the skin friction of the turbulent boundary layers. For such a nacelle the point of laminar-turbulent boundary layer transition is located in the first few percent of nacelle length. Achievement of laminar flow over the nacelles of a medium sized aircraft powered by two turbofan engines could reduce drag by about 2%. yielding cost savings of close to $50K per annum (current fuel prices and an assumed utilisation of 3,500 flying hours per annum). This economic incentive to reduce nacelle drag may be expected to increase in future as ultra high bypass ratio (UHBR) turbofan engines are introduced with nacelles having increased wetted areas relative to the wings beneath which they are mounted. It is now 90 years since Ludwieg Prandtl presented his epoch making paper on "Fluid motion with very small friction" to the 3rd International Mathematics Congress in Heidelberg. Since that time there have been many papers reporting both laboratory and full scale flight programmes furthering the understanding of boundary layer control as a means of reducing drag and increasing lift. One is mindful of the pioneering work done in Germany in the 1930s and 40s, and of the developments in the UK, France, the USA and elsewhere in the twenty years or so following WW2. Early German workers at Braunschweig and Gottingen showed that substantial reductions in skin friction are possible by maintaining the boundary layer thickness below the critical limit at which transition occurs, and that the most effective way to do this is to suck the boundary layer from the surface through a distribution of small holes: they did not however have the technology to produce a viable system. British. French and American workers each flew full scale demonstrators in the '50s and '60s. In the UK, for instance, the RAE completed a series of tests in 1957(1] in which one wing of a Vampire was sleeved: this demonstrated that full chord two dimensional flow could be achieved with what was considered to be a practical construction method. The importance of surface smoothness and waviness were appreciated, as was the need to achieve a carefully tailored distribution of porosity, the latter demanding holes a few thousandths of an inch in diameter with well formed entry and exit profiles. In varying degrees all of these experimenters identified the practical constraints of hole and surface contamination by dirt and insects, and the difficulties of maintaining adequate surface finish in a real working environment. It has only been with the comparatively recent developments in CFD design methods, composite structures, and economic laser drilling techniques that it has been possible to bring together all of the technologies required to produce a viable commercial design. Presented at the International Gas Turbine and Aeroengine Congress end Exposition The Hague, Netherlands June 13-16, 1994

2 The greater part of past research into laminar flow technology has been aimed at aircraft wing surfaces; only recently has attention been turned to other surfaces such as fins and nacelles. The concept of a Natural Laminar Flow (NLF) nacelle shaped to produce a favourable pressure gradient over much of its surface, and hence to maintain laminar stability, was investigated by Younghans and Lahti for General Electric [2] in windtunnel tests. A similar nacelle concept was tested by Barber et al [3] for Pratt and Whitney. The GE work then developed into a joint GE/NASA flight test programme [4], [5]. This series of tests was conducted on a low speed Grumman OV-113 Mohawk reconnaissance aircraft with an NLF, through-flow, nacelle attached under-wing. Tests were performed on a number of NLF nacelle profiles and the effects of noise on transition were systematically investigated by the provision of external and internal sound sources. Lamb et al [6] performed windtunnel tests of NLF, through-flow, nacelle models installed on an under-wing subsonic transport aircraft model and demonstrated that NLF nacelles do not suffer an interference drag penalty relative to a conventional nacelle. In Europe, Redeker et al [7] performed a feasibility study showing tangible benefits from a UHBR (Superfan) installation incorporating an NLF nacelle, whilst Mullender et al described 2D windturmel tests of NLF and LFC nacelle concepts and associated transition prediction methods [8], [9]. Although NLF technology is conceptually attractive in low flight Mach number conditions, the technology becomes more difficult to apply, without incurring a damaging wave drag penalty, as the cruise Mach number increases. Overcoming this difficulty has led to the development of Hybrid Laminar Flow (HLF) technology. In this the aerodynamic shape is modified to give a short region of adverse pressure gradient in which suction is used to maintain laminar stability. This sucked region is then followed by a more substantial unsucked region incorporating a favourable pressure gradient; overall the peak Mach number, and hence wave drag, can be reduced relative to an equivalent NLF design without sacrificing the potential to reduce skin friction drag. In the present programme we have set out to design, manufacture and flight test an NLF nacelle, together with an HLF modification to this basic design and to compare their respective aerodynamic performances with that of a DATUM nacelle having conventional aerodynamic shape but a similar composite structure. The nacelles have been designed using modem CFD methods in combination with manufacturing techniques which are similar to those which might be adopted for a production nacelle. We have taken advantage of the extensive flight test experience developed over many years at DLR Braunschweig, and most recently embodied in studies of laminar flow wing technology using their Domier 228 and VFW 614 aircraft. We have also exploited the expertise in liquid crystal and thin film technology developed at Oxford University to study boundary layer transition and heat transfer on gas turbine blades. The aim has been to gain an understanding of both the aerostructural constraints which may impact on the design of a full scale nacelle, and of the practical issues such as dirt accretion, insect impact, noise, and structural vibration which might inhibit the successful application of boundary layer control to a powerplant in airline revenue service. 2.0 NACELLE DESIGN The DLR flight test aircraft VFW614/ATTAS is the vehicle which has been used for this series of nacelle flight demonstrations. The ATTAS (Advanced Technologies Testing Aircraft System) aircraft is a modification of the basic VFW614 originally intended by DLR to be operated mainly for developing and evaluating the advanced technologies required for flight mechanics, flight control and guidance. With the growing interest in the application of laminar flow technology, this test vehicle has been used recently also for aerodynamic investigations into laminar flow over wings [10]. The basic VFW6I4, certified in Germany in August 1974, is a40 seater short-haul transport aircraft fitted with two Rolls-Royce/SNECMA M45H Mk 501 turbofan engines in an above-wing installation. The production standard nacelles are axisymmetric. The M45 engine was a joint Rolls-Royce/SNECMA project to produce a small advanced turbofan engine for which development started in late It developed into a twin spool turbofan with a bypass ratio of 3:1. Type approval was obtained in August 1974 with 35.0 kn take-off thrust; the M45H Mk 501 version of the engine was certified at 32.4 IcN nominal thrust on the VFW NLF Nacelle Aerodynamic Design Methodology The objective of the aerodynamic design has been to create a nacelle geometry that possessed good aerodynamic properties at the maximum cruise condition, this being the design point, whilst also having satisfactory properties at the off-design conditions in the flight envelope such as take-off and landing. When referring to the NLF nacelle on the VFW614/ATTAS it should be understood that this relates only to the fan cowl. The aerodynamic design and analysis methods employed have been essentially identical to those described in [7] for an isolated nacelle. The methodology involves, in the first stage, the aerodynamic design of the isolated NLF nacelle which is then followed by the design of the installed NLF nacelle. With the geometric contour of the conventional (DATUM) nacelle serving as origin, the design process outlined in [7] was followed leading to the NLF nacelle NLF1. Fig 1 depicts a comparative plot of the aero line of both these nacelles together with the distributions of the static pressure coefficient Cp on the external nacelle surface; it demonstrates the typical difference in Cp-distribution between a conventional and a natural laminar flow nacelle. The flight condition associated with fig 1 is Moo = 0.65, H = 21000ft and cc = 0, which effectively corresponds to the specified maximum cruise condition for the basic VFW 614. It is to be expected that when this NLF nacelle is installed on the VFW6 14, it will be subjected to a three-dimensional flow in flight. In particular mounting the engines above the wing fie on the wing suction side) leads to strong compressibility effects in the vicinity of the NLF nacelle, especially as this nacelle has a larger maximum diameter than the DATUM. In fact Euler calculations for the VFW 614 wing-fuselage configuration have shown that the local flow acceleration on the wing suction side gives rise to a marked increase in the local flow Mach number, MI, at the engine intake plane compared with the freestream 2

3 cp = Moo H(ft) U 1.5 FIGURE 1: DISTRIBUTION OF Op ON THE ISOLATED DATUM AND NLF1 NACELLES Mach number Moo; when Moo = 0.65 the corresponding intake Mach number is M1 = Translating these flow conditions to the isolated NLF nacelle NLF I, and assuming an incidence = 1.5 to take account of local flow effects on the nacelle, an iterative process was used to modify the aero lines until the revised aero Imes designated as nacelle NLF2 emerged. This nacelle possessed the properties required to maintain laminar boundary layer flow over an extensive part of its external surface in an isolated configuration at flow conditions Moo =0.77 and a = It was also found to have satisfactory off-design properties. The aerodynamic design of the installed NLF nacelle was based on a 3D Euler analysis of the wing-fuselage-nacelle configuration. The aim was to make those geometric modifications to nacelle NLE2 necessary for it to perform satisfactorily in the 3D environment encountered when installed on the VFW614. A constraint imposed on the aerodynamic design process was that, like the DATUM nacelle, the resulting nacelle geometry should remain axisymmetric. The 3D Euler analysis of installed NLF2 for the specified maximum cruise condition of the basic VF1V614 (Moo = 0.65, a = 1.5, H = 21009ft) yielded as a predominant feature the presence of a shock wave in the gulley between the wing and the nacelle at = 180, cp being the meridianal angle measured from mc = 0 ) in a clockwise direction looking into the engine intake. The associated local Mach number near the fan cowl was Mn = 1.5 and the highest local Mach number on the wing was Mn = A comparison with the corresponding results for the installed DATUM nacelle at the same flow conditions (except for incidence, this being a = 1.0 ) shows the highest local Mach number on the fan cowl to be Mn = 1.25 (cp = 180 ) and on the wing to be Mn = These local flow conditions were subsequently used as a guide in the aerodynamic redesign of nacelle NLF2. One way of reducing the compressibility effects on nacelle NLF2 was to reduce its maximum diameter. As there was a limit to the amount by which the maximum diameter could be reduced before the N L F3 C FIGURE 2: DISTRIBUTION OF Op ON THE INSTALLED NLF3 NACELLE profile merged into that of the DATUM nacelle, see fig 1, a further possibility lay in a reduction of the maximum cruise Mach number. In an iterative process both means were applied to the nacelle NLF2, ultimately yielding nacelle NLF3. For the newly designed maximum cruise condition of Mee = 0.63, a = and H = 21000ft the highest local Mach number on the surface of nacelle NLF3 = 180 ) was found to be Mn = 1.3 and on the wing Mn = 1.1. At the same freestream conditions for the DATUM nacelle the corresponding values where Mn = 1.2 on the fan cowl and Mn = 1.0 on the wing. Following this result the conclusion was drawn that nacelle NLF3 could be regarded as possessing those aerodynamic properties necessary for performing satisfactorily in the 3D environment imposed when installed on the VFW614/ATTAS. The 3D character of the flow field is evident from the distribution of the static pressure coefficient Cp on the installed nacelle NLF3 at its design point, as depicted in fig 2. The nacelle was manufactured to strict specifications, in particular covering surface finish and waviness; these are summarised in fig 3. In order to achieve the required finish and stiffness the nacelle was fabricated as two longitudinally hinged carbon fibre doors. This allowed easy access to the instrumentation and equipment mounted on the inner barrel, and also permitted the elimination of access panels and other joints which would have introduced steps or gaps into the surface. The design and manufacture by Hurel Dubois (UK) Ltd used and developed techniques directly applicable to a production component. 3

4 Manufactured surface Prescribed surface HLF NLF M N(1t) a 0 79 Surface waviness parameter definitions entail step height (mime's) 100 so N 70 to so 4 Conventional nacelle FILF PROFILE to NLF PROFILE WI 'Distance from nabeee Hoot rt.n Aero-smoothness criteria 2D steps o (mows) , L (INCHES) Surface waviness criteria Circumferential ---- Class II Conventional and datum criteria Class I NLF nacelle FIGURES: ILLUSTRATING SURFACE WAVINESS AND ROUGHNESS CRITERIA 2.2 HLF Nacelle Design The VFW614/ATTAS flight test aircraft was again the vehicle used for the demonstration. An HLF panel was designed to be accommodated within the NLF3 nacelle geometry; this panel also included a liquid transpiration insect anti-contamination system [ I 1] located in the highlight zone, upstream of the boundary layer suction surface. 1.7 FIGURE 4: ILLUSTRATING THE AEROLINE VARIATION FOR THE HLF AND NLF PROFILES AND THE RESPECTIVE PRESSURE DISTRIBUTION AT CRUISE The objective of the HLF design was to produce a section of nacelle geometry with a cruise pressure profile having HLF characteristics; that is to say a discrete region of adverse pressure gradient, requiring suction to maintain a laminar boundary layer, followed by a more extensive region characterised by a favourable pressure gradient and therefore able to sustain a naturally laminar boundary layer. The location of the required suction zone had to be far enough aft of the leading edge to allow an adequate length of insect anti-contamination transpiration to be irtcorporated. Additionally, in common with the NLF nacelle, the HLF profile was required to exhibit acceptable performance characteristics in the low speed off-design regions of the flight envelope. The manufacturing specifications for the HLF nacelle were as strict as those for the NLF nacelle. The waviness and smoothness criteria were identical. As with the NLF nacelle, the HLF design was created initially as an isolated component and its installed performance subsequently adjusted to be acceptable. To minimise the required modifications following installation, the isolated HLF nacelle was designed using conditions which partially simulated the installed configuration but were deduced from those pertaining to the previously designed NLF3 nacelle. Thus, using the geometric contour of the NLF nacelle as the starting point, a 21:7 potential flow re-design code was used to create geometry modifications giving the required pressure profile. As the geometric iterations progressed the resulting pressure profiles were assessed for their suitability using an integral boundary layer code. This code was used to assess the location of laminar to turbulent transition without suction and to determine the amount and distribution of the suction flow required to delay the transition location beyond the adverse pressure gradient, through the 4

5 TOO 00 INBOARD I OUTBOARD = 75 9 = 90 9 = 105 Pw = PB Tw TF = Pt/Tt = A LC wall static pressure. base pressure wall temperature thin film gauge array total pressure/temperature static pressure accelerometer probe microphone liquid crystal strip LENGTH OF SUCTION ZONE FIGURE 5: ILLUSTRATING THE MERIDIANAL VARIATION IN PRESSURE DISTRIBUTION AND THE CORRESPONDING SUCTION VELOCITY REQUIRED favourable pressure gradient, and hence to approximately 60% nacelle length. Fig 4 shows a comparative plot of the aero lines for the NLF3 and HLF nacelles together with the Mach number distributions on their external surfaces; the figure highlights typical differences between NLF and FILF nacelles though it must be stressed that the design was constrained by the requirement that the HLF panel be built into the existing NLF3 nacelle: consequently the nacelle maximum diameter, and the peak afterbody Mach number, have not been reduced; such reductions would probably both be possible for an unconstrained design. The flight condition associated with fig 3 is M00 = 0.56, H = 21000ft and a = 0 and corresponds to the nominal cruise conditions for the VFW6I4. Having designed the HLF geometry for isolated conditions it was necessary to examine the installed performance of the nacelle. In common with NLF3, the installed performance was investigated using a 3D Euler analysis of the wing-fuselage-nacelle configuration. The aims of the analysis were, first, to determine the optimum meridianal location of the HLF panel within the NLF geometry, with a view to minimising any influence of the strong ovenving 3D installation effects and, second, to determine the strength of the installation effects at the optimum location, thus allowing the design to be tailored to meet these effects. The circumferential width of the panel was chosen to ensure an adequate margin on either side of a trailing edge rake, hence allowing the effects of the turbulent wedges developed at the HLF to NLF FIGURE 6: DATUM NACELLE INSTRUMENTATION geometric blend region, close to the leading edge of the nacelle, to be ignored in the rake data analysis. The panel width was designed to be approximately lam wide, which corresponded to a 55 0 circumferential arc. From the installed analysis the optimum panel location was determined to be centred on the trailing edge rake, mounted at 9 = 79. The panel therefore extended from 9 = 50 to 9 = 108. Fig 5 shows the predicted circumferential pressure distribution over the panel. Again, the integral boundary layer code was used to determine the minimum suction flow rates required to delay the transition location at a range of different meridianal positions on the panel, these variations also being illustrated in fig 5, and define the optimum cruise suction flow distribution. Having determined the cruise suction surface pressure distribution and the ideal suction rates, it was necessary to design a suction system that would closely achieve the required performance. The system was designed to provide close to the ideal suction rates, whilst maintaining the constraints of no recirculation at minimum suction velocity, an acceptable maximum suction velocity (avoiding the risk of transition 5

6 ANGULAR DISTANCE r. lilt! I I z I OUTBOARD LEADING EDGE. 0 CI- 0: 21 Co: OUTBOARD ANGULAR DISTANCE I I I LEADING EDGE F'. o_ 1 0 ; ; A0901 PS PS ". A0902 0' 00' K o AP903i I PRUT TRT PRDT e"i 2?3 1 ; i; P6120 P6080 PB030 Internal Instrumentation External Surface Instrumentation Notation PS Suction Compartment Pressure PRU Upstream Restrictor Pressure PRO Downstream Restrictor Pressure TA Restrictor Temperalwe A Accelerometer KO Kulite Transducer TF Thin Film PW Surface Pressure Tappings TW Surface Platinum Resistance Thermometers PB Base Pressure Tappings FIGURE 7: ILLUSTRATING LOCATIONS OF HLF INSTRUMENTATION by over suction), and an acceptable pressure drop through the suction surface to allow a wide range of suction rates within the limitations of the suction pumping system. Suction was maintained over the required area in as homogeneous manner as possible. This was achieved using a distribution of microscopic laser drilled holes. The pitch of the holes governed the pressure drop through the surface. The size of the drilled holes were nominally 0.002" in diameter and the pitch varied from 10 to 18 diameters. The acceptable variation of the hole spacing governed the range of pressure drops available, and this, in combination with limitations on the amount of over suction, determined the number of suction manifolds. Matching the required suction rates to the axial variation in surface pressure distribution resulted in the selection of 8 manifolds. Circumferential variations in the surface porosity were required to allow for the predicted variation in surface pressure distribution, the predicted pressure drops being derived from [12]. 6

7 FIGURE 8: THIN FILM GAUGE FIGURE 9: LIQUID CRYSTAL HEATER PAD The suction flow was maintained by an ejector pump powered by HP compressor bleed air. The flows from each compartment, and the total flow, were measured using calibrated orifice plates. They were regulated using motor driven needle valves; the flight test team controlled the valves from the cabin. It has long been recognised that one critical requirement for achieving a practical laminar flow design is resolution of the insect contamination problem. Results from the NLF flight trials reported herein clearly demonstrated this requirement. A liquid transpiration anti-contamination system developed by AS&T [llj and based upon the TKS anti-icing system, was incorporated into the leading edge of the HLF panel. The length of the protected region was determined by observations (from the previous NLF nacelle flight test) of the inlet streamtube stagnation point movements during the take-off roll and climb out, and of the "burst velocity" above which experience showed insects smeared on the nacelle highlight zone. The system was designed to operate over a temperature range between 10 0 and 50 C; at temperatures below this it was expected that insect contamination would not be a significant concern. The fluid flow rate was governed by the fluid viscosity and the pump speed: a wide range of flow rates could be examined enabling the minimum requirement to be determined. 3.0 INSTRUMENTATION The flight test nacelles (DATUM, NLF and HLF) were all extensively instrumented. The instrumentation is summarised in figs 6 and 7 for the DATUM and HLF nacelles respectively. The NLF and DATUM systems were identical except for some changes in axial location to account for the later NLF transition. The instrumentation standard for the HLF nacelle was similar, but confined to a smaller circumferential region of the nacelle; it included a distribution of suction and anti-contamination systems transducers. Right test instrumentation also included conventional and infra-red cameras installed both in the cabin and in an outboard wing mounted pod. 3.1 Nacelle Mounted Instrumentation The instrumentation on all three nacelles included axial rows of surface static pressure tappings; axial rows of flush mounted platinum resistance thermometers; accelerometers mounted on the trailing edge of the nacelle; Kulite pressure transducers and probe microphones mounted flush to the nacelle surface. Total pressure/temperature (Pt/ft) and static pressure rakes were mounted close to the cold nozzle lip. The rakes were interchangeable, allowing both radial pressure distribution to be measured in each meridianal plane, albeit on successive flights. On the HLF nacelle only, additional instrumentation included suction manifold pressure tappings; static pressure tapping at key points in the piping system,; the massflow from each manifold, together with the total flow using calibrated orifice plates. The suction control valve positions were recorded together with the pump speed and fluid pressures within the anti-contamination system, these enabling fluid flow rates to be determined. 3.2 Laminar/turbulent Boundary Layer Flow Transition Instrumentation Infra-red cameras were used for global observations of the transition location. Each of the fan cowls was fitted with an array of hot film gauges (HFGs) containing up to thirteen gauges, fig 8. These were manufactured specifically for the flight tests by the Department of Engineering Science at the University of Oxford. The gauges were arranged so that the first three would yield the wavelength, speed and direction of Tollmien-Schlichting (TS) waves, the remaining in-line gauges being used for transition detection and turbulent spot analysis. A hot wire probe and hot film wedge probe were mounted from the 7

8 FIGURE 10: LAMINAR FLOW NACELLE IN FLIGHT fuselage downstream of the main access door to enable freestream conditions to be monitored. Two liquid crystal heater pads consisting of uni-directional carbon fibres were inlaid into the surface of the NLF cowls in meridianal planes q = 120 and 240, see fig 9. Each pad had twelve Platinum Resistance Thermometers (PRTs) positioned just under the surface. The pads were sprayed with several coats of encapsulated, thermochromic liquid crystals. Each liquid crystal had colour play in a narrow band of approximately I.5 C width and three liquid crystals with colour bands at 5 C intervals were used. An ultra-violet filter was also sprayed on the surface which was then polished with fine grade paper to remove any surface roughness. The liquid crystal colour play was recorded with video and SLR cameras mounted in the fuselage and within electrically heated containers in the wing pod. Knowing the temperature distribution along the nacelle from the PRTs and liquid crystals and the corresponding input heat flux, the heat transfer coefficient can be determined for a given flight condition [13]. 4.0 FLIGHT TEST RESULTS 4.1 NLF Nacelle The first series of flight trials using the NLF nacelle took place in Sept Initial flights established that the nacelle met all of its operational requirements. Instrumentation evaluation flights followed; at an early stage it became clear that significant amounts of laminar flow existed on the nacelle throughout the flight envelope. A successful two week test programme was then concluded. Further flight tests of the NLF nacelle took place in April These included two tests in which the boundary layer was artificially tripped to initiate turbulent flow immediately downstream of the intake highlight. The DATUM nacelle was also flown during this phase of the flight programme. During the two flight test series there were 34 NLF and DATUM nacelle flights taking 65.5 hours of flying and covering 672 test conditions. The test points covered:- Reynolds numbers up to 7.6 million/metre Mach numbers Altitudes 0-25,000 feet Engine Speeds Flight idle 100% NI Max Yaw angles + 3 Curves at constant Reynolds numbers Curves at constant Mach number Curves at constant altitude There were 2 flights with the DATUM nacelle, and 32 flights with the NLF.nacelle (including 2 with a tripped boundary layer). Steady test conditions were held, typically, for 2-3 minutes, though a limited number of test points were held for up to 10 minutes to allow extended investigations using the liquid crystals. A photographic illustration of the NLF flight configuration is shown on fig 10. Fig 11 illustrates the experimentally achieved Cp distribution on the NLF nacelle at cruise conditions, this being compared with the predictions from the installed Euler code. The distribution shown were located in the outboard of the nacelle; it should be noted that the accuracy of the prediction reduced at locations closer to the pylon as the pylon was not represented in the Euler model. Fig 12 illustrates the inboard and outboard infra-red images of the NLF nacelle at a cruise condition; the transition line has been highlighted. This is located at the change between the lighter and darker shaded regions of the infra-red image. The extent of laminar flow at this condition is typical, with approximately 60% occurring on the outboard side, and between 20% and 60% on the inboard side. 8

9 cp = 30 I 9=140 I FLIGHT TEST DATA EULER CALCULATION: WING FUSELAGE NACELLE NOTE: FOR THE EULER CALCULATIONS NO PYLON WAS REPRESENTED FIGURE 11: ILLUSTRATES THE ACHIEVED Cp DISTRUBUTION ON THE NLF NACELLE AT CRUISE CONDITIONS Note that care must be taken when evaluating of the infra-red images as the sub-surface nacelle structure produces a shading band which can also be seen in the images. The location of the transition regions corresponds closely with the predictions of both an integral boundary layer code, and a differential boundary layer code run in conjunction with a linear stability method. The HFG and liquid crystal pads produced excellent data, both sets of instrumentation illustrating clearly the start and length of the transition region. Fig 13 shows the development of turbulent spots into fully turbulent flows recorded by the outboards HFGs; analysis of this EIFG data showed that it is was possible to identify the T-S waves which had been amplified and caused transition. Early indications are that the amplified frequencies which have been measured agree well with analysis using the linear stability method. Fig 14 is an enhanced picture of the liquid crystals temperature contours on the heater pad; the temperature gradient present at the point of laminar turbulent transition causes the liquid crystal isotherms to group together, facilitating location of transition. Analysis has shown [13] that the experimentally determined heat transfer distribution can be modelled closely and that a simple finite difference boundary layer method gives reasonable agreement; the method leads to a good estimate of the skin friction coefficients. An estimate of the drag reduction attributable to the delayed transition can be obtained using the method of Squire and Young to analyse the cold nozzle plane rake measurements; this approach has shown that a 30% drag reduction is achieved in meridianal planes which are free of installation effects. During some flights significant numbers of insect strikes occurred. These clearly degraded the NLF nacelle laminar flow performers as illustrated in fig 15. These results emphasise the requirement for a viable anti-contamination system if laminar flow technology is to be operationally acceptable in a service environment. 4.2 HLF Nacelle Right trials of the HLF nacelle took place in September Initial flights again explored the safe working envelope of the nacelle and engine, the HLF nacelle also fully satisfied all operational requirements. A successful 4 week flight programme followed during which both the suction and anti-contamination systems were operated successfully. There were 14 flights, totalling some 28 hours of flying and 300 separate test conditions. The test points covered. Reynolds numbers up to 7.6 million/meter Mach numbers Altitudes feet Engine speeds Right idle-100% NI Max Yaw angles ±3 curves of constant Reynolds number 9

10 Data Point: GA038 FLAW flo. Data Point: GI038A FLOW Date: :12:58 Date: :14:45 OUTBOARD INBOARD SIRUCIURE. S/40WING DUE TO DIFFERING REA PROPERTIES:: " Data Point: GA038 Data Point: GI038A Date: :12:58 Date: :14:45 FIGURE 12: INFRA-RED IMAGES OF THE LAMINAR FLOW NACELLE AT CRUISE CONDITIONS 1 0

11 LAMINAR [ p Or I. 45% x/c Ural. 48V0 X/C 51 1Y0 x/c killoladuo,,j,l Ike OW 54CY0 X/C. 1 1\111003/400,V164 T o. ow 56.5% x/c. lira Cos) 59.5% x/c: curves of constant Mach number curves of constant altitude suction system on/off Insect anti-contamination system on/off during take off/landing A photographic illustration of the flight configuration is shown in fig 16. Fig 17 illustrates the infra-red image of the HLF panel at cruise conditions with and without suction; the transition line has again been highlighted. As expected the application of suction delays the onset of transition to the 60% chord position seen during the NLF flight trials. Because the installed aerodynamic effects on the nacelle were over predicted the range of surface porosity applied over compensated for the circumferential surface pressure variation. This resulted in a higher than expected total suction flow. The detailed variation in local suction rates has yet to be analysed; comprehensive surface pressure measurements from the flight programme coupled with post test flow measurement should allow the detailed distribution of suction flows to be determined. All of the instrumentation performed well, the transition location being identified clearly by the infra-red system, the HFGs and PRTs. The insect contamination system was operated successfully throughout the take-off and landing stages of 4 flights. Fig 18 illustrates the effectiveness of the system, the protected region being clear of detrimental insect strikes, as opposed to the significant number of strikes occurring on the unprotected region. It should be noted that although, after landing, there were some insects on the protected zone, these were picked up during the approach, were intact, and the debris was judged to be sufficiently mobile to be washed from the nacelle surface in any subsequent flight. The application of Hybrid Laminar Flow technology at flight Mach numbers as high as Moo = 0.82 will required careful aerodynamic design to prevent the early onset of wave drag. Design studies completed in Europe [15) have illustrated the feasibility of high speed HLF designs at both cruise and off design conditions. These studies are expected to culminate in large scale, high speed and low speed wind tunnel tests of an optimised HLF nacelle design. TURBULENT rm. fail 620/o x/c; Ta ct-) FIGURE 13: ILLUSTRATING THE TRANSITION DEVELOPMENT AS SHOWN BY THE OUTBOARD THIN FILM GAUGES 11

12 Direction of flight Note the grouping of colour bands at the transition location FIGURE 14: SHOWING THE DEFINITION OF THE TRANSITION LOCATION BY THE LIQUID CRYSTALS AT CRUISE FIGURE 15: TURBULENT WEDGES RESULTING FROM INSECT CONTAMINATION 12

13 FIGURE 16: HYBRID LAMINAR FLOW NACELLE INSTALLED ON VFW 614 SUCTION OFF SUCTION ON FIGURE 17: INFRA-RED IMAGES OF THE HYBRID LAMINAR FLOW NACELLE WITH AND WITHOUT SUCTION AT CRUISE CONDITIONS 13

14 PROTECTED REGION 3 Strikes invaluable cooperation of Hurel Dubois (UK), AS&T Ltd, and Deutsche Airbus is acknowledged gratefully. In addition the authors wish to thank Professor T Jones, Dr T Cain (Oxford University), Dr K H Horstmann (DLR), Dr P Lucking (MW), C Millar, P Shipley, Li Rodgers and C W Freeman (RR) for their significant technical contributions to the programme. Thanks are also due to the aircrew and groundstaff at DLR. The authors accept the responsibility for the ideas presented in this paper which do not necessarily represent the policies of their sponsoring organisations UNPROTECTED 20 Strikes FIGURE 18: SHOWING THE SUCCESSFUL OPERATION CONCLUSIONS A total of 93 1/2 hours of flying demonstrated clearly that the extension of laminar boundary layer flow to 60% of nacelle length in a fully installed engine environment is achievable both by natural means and by the use of suction. Laminar flow to 60% of nacelle length was achieved by natural and hybrid means over a large range of flight conditions. Successful application of a wide range of complementary instrumentation has given a good understanding of the extent of laminar flow and the location and the length of transition. This data has added confidence in the methods employed to define the present nacelles. Potential receptivity problems which could have affected the extent of laminar flow, such as vibration, noise, surface finish, roughness, and heat flux arising from solar heating had no apparent effect on the results. The methods of nacelle production which were employed achieved required manufacturing tolerances for laminar flow and could be transferred readily to a quantity production environment. Hybrid laminar flow will be required to satisfy the operational requirements of a high cruise Mach number nacelle configuration. Initial indications are that the constraints imposed on an operationally viable system in an airline environment, such as insect contamination, can be addressed. Further work is required to produce a fully engineered system suitable for routine airline operations. ACKNOWLEDGEMENTS The authors wish to thank Rolls-Royce plc, DLR and MTU for permission to publish this paper. Thanks are also due to MTU and the UK department of Trade and Industry for their fmancial support. The REFERENCE [1] Head, MR., Johnson, D., Coxon, M., "Flight experiments on boundary layers for low drag ARC," R & M no 3025 (1955). [2] Younghans, J.L., Lahti, D.J., "Experimental studies on Natural Laminar Flow Nacelles" AIAA (1984) [3] Barber, Ti., Ives, D.C., Nelson, D.P., Miller, R.M.,."Computional Design & Validation Test of Advanced Concept Subsonic Inlets." AIAA [4] Obara, C.J., Hastings, E.C., Schoenster, JA., Parrott, T.L., Holmes, B.J., "Natural Laminar How Hight Experiments on a Turbine Engine Nacelle Fairing." A1AA [5] Obara, C.J., Dodbele, S.S., Hastings, E.C., Schoenster, J.A., "Flight Research on Natural Laminar Flow Nacelles: A progress report." AIAA [6] Lamb, M., RE, R.I., "Natural Laminar Flow Nacelle for Transport Aircraft." N NASA CP 2397 (1985) [7] Redeker, G., Redespiel, R., Horstmann, K.H., "Feasibility study on the Design of a Laminar Flow nacelle." AJAA [8] Mullender, AJ., Bergin, AL., Poll, D.I.A., "The application of Laminar flow to aero engine nacelles Proc. Royal Aero. Soc. Boundary layer transition and Control, Cambridge (1991) [9] Mullender, A.J., Poll, D.I.A., "A combined experimental and theoretical study of laminar flow control with particular relevant to aero engine nacelles Proceedings, First European Forum on Laminar Flow Technology" , Hamburg (1992) [10] Horstmann, K.H., "Flight tests with a Natural How Glove on a Transport Aircraft." AIAA CP907 (1990) [II] Humphreys, B., "Contamination avoidance for laminar flow surfaces. (1992). Proceedings first European forum on Laminar flow technology." [12] Bieler, H., Priest, J., "HLFC for Commercial Aircraft First ELFIN test results (1992). Proceedings first European forum on Laminar flow technology." [13] Bown, NM., Cain T.M., Jones TN., "In Flight Heat transfer measurements on an Aero-Engine nacelle. Submitted to ASME for publications 1993." [14] Govindarajan, R., Narasimma, R., "The Role of Residual Non-turbulent Disturbances on Transition Onset in Two-Dimensional Boundary Layers", Trans ASME, J Fluids Eng, Vol 113, p.147 [15] Shipley, RP., Lecordix, It., "The design optimisation of a Hybrid Laminar flow nacelle." Godard, It., Rossow, C.C., Aerodays 93, Second Community Aeronautics RID Conference, Naples (1993) 14