Document
NASA/TP-2002-210730
Flight Test Techniques Used to Evaluate
Performance Benefits During Formation
Flight
Ronald J. Ray, Brent R. Cobleigh, M. Jake Vachon, and Clinton St. John NASA Dryden Flight Research Center Edwards, California August 2002 The NASA STI Program Office...in Profile CONFERENCE PUBLICATION.
Since its founding, NASA has been dedicated to the advancement of aeronautics and space Collected papers from scientific and science. The NASA Scientific and Technical technical conferences, symposia, seminars, Information (STI) Program Office plays a key or other meetings sponsored or cosponsored part in helping NASA maintain this by NASA.
important role.
SPECIAL PUBLICATION. Scientific, technical, or historical information from The NASA STI Program Office is operated by NASA programs, projects, and mission, Langley Research Center, the lead center for often concerned with subjects having NASA's scientific and technical information.
substantial public interest.
The NASA STI Program Office provides access to the NASA STI Database, the largest collection TECHNICAL TRANSLATION. English- of aeronautical and space science STI in the language translations of foreign scientific world. The Program Office is also NASA's and technical material pertinent to institutional mechanism for disseminating the NASA's mission.
results of its research and development activities.
These results are published by NASA in the Specialized services that complement the STI NASA STI Report Series, which includes the Program Office's diverse offerings include following report types: creating custom thesauri, building customized databases, organizing and publishing research TECHNICAL PUBLICATION. Reports of results...even providing videos.
completed research or a major significant phase of research that present the results of For more information about the NASA STI NASA programs and include extensive data Program Office, see the following: or theoretical analysis. Includes compilations of significant scientific and technical data • Access the NASA STI Program Home Page and information deemed to be of continuing at http://www.sti.nasa.gov reference value. NASA's counterpart of peer-reviewed formal professional papers but • E-mail your question via the Internet to has less stringent limitations on manuscript help@sti.nasa.gov length and extent of graphic presentations.
• Fax your question to the NASA Access Help TECHNICAL MEMORANDUM. Scientific Desk at (301) 621-0134 and technical findings that are preliminary or of specialized interest, e.g., quick release • Telephone the NASA Access Help Desk at reports, working papers, and bibliographies (301) 621-0390 that contain minimal annotation. Does not contain extensive analysis.
Write to: NASA Access Help Desk CONTRACTOR REPORT. Scientific and NASA Center for AeroSpace Information 7121 Standard Drive technical findings by NASA-sponsored contractors and grantees. Hanover, MD 21076-1320 NASA/TP-2002-210730
Flight Test Techniques Used to Evaluate
Performance Benefits During Formation
Flight
Ronald J. Ray, Brent R. Cobleigh, M. Jake Vachon, and Clinton St. John NASA Dryden Flight Research Center Edwards', California National Aeronautics and Space Administration Dryden Flight Research Center Edwards, California 93523-0273 August 2002
NOTICE
Use of trade names or names of manufacturers in this document does not constitute an official endorsement of such products or manufacturers, either expressed or implied, by the National Aeronautics and Space Administration.
Available from the following: NASA Center for AeroSpace Information (CASI) National Technical Information Service (NTIS) 7121 Standard Drive 5285 Port Royal Road Hanover, MD 21076-1320 Springfield, VA 22161-2171 (301) 621-0390 (703) 487-4650 FLIGHT TEST TECHNIQUES USED TO EVALUATE PERFORMANCE BENEFITS DURING FORMATION FLIGHT Ronald J. Ray,* Brent R. Cobleigh, t M. Jake Vachon, $ and Clinton St. John § NASA Dryden Flight Research Center Edwards, California Abstract Nomenclature Acronyms The Autonomous Formation Flight research project has been implemented at the NASA Dryden Flight AFF Autonomous Formation Flight Research Center to demonstrate the benefits of ATC automatic throttle control formation flight and develop advanced technologies to facilitate exploiting these benefits. Two F/A-18 aircraft DPS digital performance simulation have been modified to precisely control and monitor GPS global positioning system relative position, and to determine performance of the HUD trailing airplane. Flight test maneuvers and analysis head-up display techniques have been developed to determine the IFT in-flight thrust performance advantages, including drag and fuel flow INS inertial navigation system reductions and improvements in range factor. By flying the trailing airplane through a matrix of lateral, NASA National Aeronautics and Space longitudinal, and vertical offset positions, a detailed map Administration of the performance benefits has been obtained at two flight conditions. Significant performance benefits have __mbols been obtained during this flight test phase. Drag acceleration along flightpath, g Ax W reductions of more than 20 percent and fuel flow reductions of more than 18 percent have been measured lateral acceleration (wind axis), g Ay W at flight conditions of Mach 0.56 and an altitude of normal acceleration perpendicular to Az W 25,000 ft. The results show favorable agreement with flightpath, g published theory and generic predictions. An F/A-18 coefficient of drag (C D = 2D/(Sp V 2) CD long-range cruise mission at Mach 0.8 and an altitude of 40,000 ft has been simulated in the optimum formation coefficient of induced drag CD i position and has demonstrated a M-percent fuel (CDi = C D-CDo ) reduction when compared with a controlled chase coefficient of zero-lift drag airplane of similar configuration. CD o coefficient of lift (C L = 2L/(S O V 2) Q *Aerospace Engineer.
D drag, lbf #Aerospace Engineer, Member AIAA.
:)Aerospace Engineer, Member AIAA.
excess thrust, lbf FEX §Aerospace Engineer, Member AIAA.
Copyright © 2002 by the American Institute of Aeronautics and FEDRAG engine throttle_lependent drag, lbf Astronautics, Inc. No copyright is asserted in the United States under Fa gross thrust, lbf Title 17, U.S. Code. The U.S. Government has a royalty-flee license to exercise all rights under the copyright claimed herein for FRAM ram drag, lbf Governmental purposes. All other rights axe reserved by the copyright owner.
g gravity constant Notice: the use of trade names or names of manufacturers in this document does not constitute an official endorsement of such GW gross weight, lbf products or manufacturers, either expressed or implied, by the National Aeronautics and Space Administration.
L lift, lbf American Institute of Aeronautics and Astronautics load factor perpendicular to flightpath, g reduction from the experimental proof-of-concept stage Nz w to a prototype demonstration within three years of P power, lbf-ft/sec commencing. The prototype demonstration was planned S to be accomplished using two highly instrumented, wing area, ft 2 NASA-owned F/A-18 aircraft (McDonnell Douglas V velocity, ft/sec Corporation, now The Boeing Company, St. Louis, W Missouri; and Northrop Corporation, now Northrop upwash velocity, ft/sec Grumman, Newbury Park, California) equipped with the WFT fuel flow rate, lbm/hr necessary research systems. A primary factor to verifying the drag reduction goal has been the X longitudinal separation, wingspan (37.5 ft) implementation and validation of the in-flight Y lateral separation, wingspan performance data system and the development of flight test techniques to analyze performance benefits during Z vertical separation, wingspan formation flight.
(_ angle of attack, (leg The AFF project goals and objectives originally were 7 flightpath angle, deg planned to be accomplished in four phases, with flight A change in parameter test beginning in the first quarter of the 2001 fiscal year and completing by the end of the 2003 fiscal year. The pitch angle, deg first phase demonstrated precision autonomous density, lbm/ft 3 P station-keeping. The second phase mapped the vortex effects of the leading airplane on the trailing airplane, Subscripts and evaluated the performance of the high-accuracy relative-position sensor system and datalink BL baseline (nonformation flight) communications. This report presents the results est estimated obtained from this phase.
FF formation flight The AFF project was canceled shortly after this lead leading airplane second phase because of budget constraints. The third phase contains the primary objective of the project and trail trailing airplane was intended to demonstrate the integrated system performance by achieving a sustained 10-percent fuel Introduction savings while under close formation flight. The results of this third phase have been envisioned to have The performance benefits of formation flight were commercial and military applications to cargo and known before man could even fly. Many bird species fly passenger transports and unmanned air vehicles.
in "V" formation to take advantage of the upwash field Although not all the autonomous control goals were generated by adjacent birds, resulting in less energy realized during this now-truncated AFF project, expended. 1Analytical studies and qualitative flight tests significant vehicle performance improvements have have shown this benefit is significant and can be been demostrated.
reproduced for a formation of aircraft.
Determining vehicle performance and test techniques Beukenberg and Hummel 2 flew two aircraft in to use while in formation flight has unique challenges.
formation using autopilots and measured values of For example, the localized upwash effects from the induced upwash velocity and aileron deflection to leading airplane made the angle-of-attack probes of the optimize position within the vortex. This simple test trailing airplane inaccurate during formation flight.
achieved an average relative power reduction of 10.24 Alternative measurement techniques had to be percent using limited instrumentation.
developed. An important aspect of obtaining the data A primary objective of the Autonomous Formation presented in this report was the accurate positioning of Flight (AFF) project at the NASA Dryden Flight the trailing airplane during formation flight to acquire Research Center (Edwards, California) has been to important aerodynamic data for development of the autonomous control system. To enable the pilots to validate the drag reduction concept and prediction tools in the flight environment for aircraft in formation. The manually fly the trailing formation position with project intended to advance the concept of AFF drag accuracy, a relative-position indicator was developed to AmericanInstituteof Aeronautics and Astronautics show position errors using the head-up display (HUD). The rotation of the original lift and drag values are In turn, flying with accuracy allowed for the systematic represented by L' and D'. Primary assumptions in this mapping of the drag and fuel consumption benefits. An derivation are that lift is much greater than drag additional challenge was to determine how the varying (L>>D), and the Ao¢ value is sufficiently small enough fuel weights of the leading and trailing aircraft affected the results.
that approximations in trigonometry can be used with a great level of accuracy. The theory also implies the This report describes the analysis methods and test magnitude of the resultant aerodynamic force techniques developed and employed to determine (_L2+D _) remains constant because the upwash only aircraft performance during the detailed mapping of the rotates this force: vortex, the second flight test phase of the AFF project. A summary of prediction theory and the techniques used to validate that theory is also provided. Sample results L2_--_++ D2 = L_ + D '2 are given to demonstrate the data quality and are (1) compared to predictions obtained from basic theory.
In reality, the external upwash field from the leading Basic Theory airplane is not uniform over the trailing airplane and does not simply rotate the aerodynamic force. Because The most common theory on formation flight states the rotation appears to be the predominant effect, most that drag reduction is actually attained because of a reports make this assumption and do not discuss the rotation of the lift vector that occurs while a trailing airplane is in the upwash field of the leading airplane. 3 5 second-order effects of how upwash affects the velocity Figure 1 shows how the baseline (nonformation flight) (V) and induced drag (CDi). Because of traditional values for lift and drag (L and D, respectively) rotate by bookkeeping methodology, the actual L and D values the change in angle of attack (Ace) because of the are maintained relative to the vehicle flightpath during upwash (W) effect of the trailing vortex of the leading formation flight. The term AD is used to represent the airplane while in formation flight.
' f °'°'_-- Resultant aerodynamic S D' AL
f °
V I
j TTtw
v V Aa 020205 Figure 1. Rotation of resultant aerodynamic force caused by upwash of the leading airplane.
American Institute of Aeronautics and Astronautics drag change caused bytherotation of theoriginal lift with a single "horseshoe" vortex. The assumption is that only the induced drag is affected by the upwash force from L to L'. The drag during formation flight, influence. Hoerner 6 has presented predictions using this DFF, iS obtained as follows: approach and assuming no lateral wingtip separation.
He shows a single trailing airplane is predicted to have a decrease in drag of 20 percent when flying level and one (2) DFF = D' cos(AoQ -AD wingspan aft of the leading airplane. Figure 2 (adapted from reference 5) shows a comprehensive prediction where map of the benefits of mutual induced drag between a given pair of aircraft using the horseshoe vortex model AD = sin(Aoc)L as a function of lateral and vertical spacing.
In a similar manner, the term AL is used to represent the lift change caused by the rotation of the drag force from D to D'. The lift during formation flight, LFF, iS obtained as follows: (3) LFF = L'cos(AoQ + AL /---AFF ............. /, flight data where 12 11 region_ AL = sin(AoQD In practice, the pilot will adjust the airplane pitch attitude and throttle setting while in formation to keep all the forces acting on the airplane in balance. That is, -.5 0 .5 1.0 the reduction in drag requires the pilot to reduce power Lateral separation (Y), wingspan to maintain speed, and the small increase in lift requires Wingtip overlap Separation the pilot to slightly pitch the aircraft nose forward or it 020206 will diverge from its flightpath. Because lift tends to be Figure 2. Predicted variation in mutual induced drag an order of magnitude greater than drag (L>>D), drag with aircraft position using the horseshoe vortex model.
is influenced significantly more by the rotation effect than lift is: sin (AoQL >> sin(Ao0D The mutual induced drag accounts for the changes in (4) induced drag of the leading and trailing aircraft. For AD>>AL positions where the trailing airplane is sufficiently aft of the leading airplane, its drag is not influenced. 5 The A considerable reduction in drag can be attained by a variation in mutual induced drag is thus only caused by small upwash angle while an insignificant increase in the change incurred on the trailing airplane. Because of lift simultaneously occurs.
this idiosyncrasy, the predicted values of variation in mutual induced drag shown in figure 2 are directly Because of the upwash effect while in formation comparable to the measured variation of induced drag of flight, the effective or aerodynamic velocity vector is no the trailing airplane presented in this report.
longer in the direction of the actual flightpath. This change makes the trailing airplane appear to A more detailed prediction method, based on a vortex aerodynamically be in a descent (relative to the wind), lattice theory that uses multiple quadrilateral vortices, requiring the pilot to reduce power to maintain altitude.
has been presented in references 4, 5, and 7. Blake 5 This characteristic is similar to how a glider soars provides a comparison of the mutual induced drag factor without power because of thermals or vertical updrafts as a function of lateral position using the horseshoe over a ridge.
vortex and the vortex lattice methods (fig. 3). The The simplest theoretical analysis for predicting drag predictions for the vortex lattice method are for two reduction while in formation flight replaces each wing rectangular wings with an elliptical lift distribution.
American Institute of Aeronautics and Astronautics P = DxV Percent power reduction = k P/PBL (6)
cff
= (D BLV - D FFV)/D BLV c t,- = 1-(DFF/DBL ) ci- -50 "o z_ Horseshoe vortex: Aircraft Description "o Vortex diameter/ Wingspan = 0.03 Two NASA Dryden F/A-18 research aircraft (fig. 4) -100 I I I I I I t I I I I = I I I , I I I -I .0 -.5 0 .5 1.0 were modified to support the AFF project. The F/A-18 Lateral separation (Y), wingspan aircraft is a supersonic, high-performance fighter powered by two F404-GE-400 (General Electric, Lynn, Wingtip overlap Separation 020207 Massachusetts) turbofan engines, each producing 16,000 lbf of thrust in the afterburner. Both engines are Figure 3. Mutual induced drag variation predicted by mounted close together in the aft fuselage. The aircraft various generic methods.
has a wingspan of 37 ft, 5 in. The fuselage is 56-ft long.
NASA Dryden aircraft tail numbers 845 and 847 were specified to be the leading and trailing airplanes, In addition to being dependent on relative position of respectively.
the two aircraft, the actual drag reduction values are highly dependent on the vortex strength of the leading airplane, which varies with lift or gross weight (GW).
As fuel is burned, the leading airplane requires less lift to maintain level flight conditions and produces a vortex of lesser strength. This weaker vortex reduces the upwash on the trailing airplane and its drag reduction potential.
Conversely, as the trailing airplane burns fuel, it is flying at a lower drag state (because of the lower lift) and therefore may show greater potential for drag reduction benefits because the percent drag reduction will be less as baseline drag (DBL) decreases: EC01-0328-12 Percent drag reduction Figure 4. The AFF aircraft in formation flight.
= AD/DBL (5) = (DBL - DFF)/DBL The trailing airplane (number 847) is a single-seat = 1 - (DFF/DBL) configuration that weighs 36,433 lb when fully loaded with 10,860 lb of internal fuel. This F/A-18A aircraft The effects of both leading and trailing aircraft burning has a published attack radius of 575 nmi, and its ferry fuel and consequently lowering weight might counteract range is listed at more than 2,000 nmi. The leading each other to some degree.
airplane (number 845) is a two-seat, preproduction (or full-scale development) aircraft that has been Engine fuel reduction is more directly associated to extensively modified to provide a flexible platform for the power reduction than drag reduction during advanced flight systems. Because of its two-seat formation flight. Because engine power change (AP) is configuration, the airplane carries less fuel and has a of significant interest, AP can easily be shown to be longer canopy. The airplane weighs 36,021 lb when fully loaded with 9,926 lb of fuel.
equal, in percent, to that of the drag change: American Institute of Aeronautics and Astronautics Instrumentation The engine manufacturer's aerothermodynamic in-flight thrust (IFT) computer model was used to determine thrust values for this study. 1° The model uses The trailing airplane has been specially instrumented two correlation techniques for determining ideal gross to obtain aircraft performance data and detailed thrust: area pressure, and mass flow temperature. The relative-position information while in formation. To mass flow temperature technique was chosen as the precisely map the vortex, the pilot was provided with an primary method for use in AFF performance calculation indicator on the HUD that showed the error between the because of its proven accuracy (on the order of 2 percent current relative position and the commanded relative for net thrust). 11 position. The current relative position was calculated using global positioning system (GPS) measurements Table 1 shows a summary of the input requirements on both aircraft. The leading airplane transmitted its for the IFT model for both thrust methods. A 20-probe GPS position, velocity, course over the ground, and GPS total pressure rake was used to determine the average time to the trailing airplane using a commercially turbine exit pressure. A volumetric fuel flow meter was available wireless modem. The instrumentation system installed to provide improved primary values for fuel on the trailing airplane time-correlated and differenced flow rate, WFT.
this data with its local GPS position measurements to obtain a 2-Hz relative-position estimate. 8 This estimate Table 1. F404 engine thrust model measurement inputs.
was extrapolated using the relative velocity to obtain a 10-Hz output. The 10-Hz relative position then was Area Mass flow Parameter differenced with one of 64 preprogrammed commanded pressure temperature positions that were selectable from a cockpit switch.
Nozzle area Primary Secondary Mach number Secondary Primary The resulting error signal was represented by two Fan rotor speed Secondary Secondary needle displays on the HUD. The vertical needle provided lateral position error; the horizontal needle Power lever angle Secondary Secondary provided vertical error. By maneuvering the airplane to Ambient pressure Primary Secondary center the two needles, the pilot was able to maintain a Turbine exit pressure Primary Secondary constant relative position. Earlier flight test experience with this technique 9 showed that an accuracy of 4 ft Fan inlet temperature Secondary Primary (2 standard deviations) is achievable when both GPS Primary fuel flow Secondary Primary systems are using a common satellite set and are time-synchronized. During the limited time that the two GPS units did not have a common satellite set, the A laser-mounted inertial navigation system (INS) was needles were programmed to disappear from the pilot used to obtain vehicle accelerations, attitudes, and rates.
display and the flight testing was delayed until common The aircraft airdata system was used to obtain Mach satellites were reestablished. The accuracy of the number and altitude values. Comparisons with the real-time, relative-position system was validated using leading airplane show no significant effect of the vortex postflight differential, carrier-phase GPS measurements.
on the airdata measurements of velocity, Mach number, The error was found to be 2.5 ft (2 standard deviations).
and altitude.
The position measurement system used in this report Flight Test Maneuvers is provided in units as a ratio of the F/A-18 wingspan (37.5 ft). The longitudinal separation, X, is defined such Specific maneuvers were flown to validate the thrust that X = 0 when both aircraft are aligned nose-to-nose.
and basic performance data. This validation was Because the F/A-18 aircraft is 56-ft long, X = 1.5 accomplished before AFF performance data were when the trailing airplane has zero separation distance gathered.
between its nose and the tail of the leading airplane. The lateral separation, E is defined such that Y = 0 when the Thrust Validation wingtips are aligned, and increasing overlap is represented by negative values of E Vertical separation, To verify the engine manufacturer's thrust model and Z, is defined such that Z = 0 when the wingtips are its required instrumentation were working properly, an aligned, and trailing airplane positions below the installed ground test was accomplished on the Universal leading airplane are negative values of Z. Horizontal Thrust Stand at the Air Force Flight Test AmericanInstituteof Aeronautics and Astronautics Center (Edwards, California) before firstflight. These performance. This technique consisted of flying to the tests were individually accomplished atvarious power prescribed formation position using the HUD needles settings foreach engine. Calculated gross thrust values for AY and AZ position and control room calls for compare favorably tothose measured ontheteststand, AX position. When in position, the pilot would hit the giving confidence the IFT model and required trigger button on the control stick to indicate the instrumentation were working properly. beginning of the maneuver. After approximately 30 sec, a control room call to "engage automatic throttle
To complete thecheckout of theIFT model and
control" (ATC) was made, indicating to the pilot to
instrumentation, bothsteady-state and dynamic engine
engage the ATC command button when the pilot thought test points were accomplished atvarious test conditions. conditions were stable. The use of ATC tended to
Thesteady-state datawereobtained at cruise flight
smooth out and limit throttle movement while holding
conditions, andtheresults fromthe in-flight thrust
aircraft position.
model were compared topredictions obtained from the
manufacturer's specification model. Throttle steps and
After approximately 20-sec more, a control room call
throttle frequency sweeps were used toassess theeffect
to "slide out" was made, for which the pilot laterally
unsteady engine operation hadontheIFTusing the
maneuvered out of the formation and immediately
techniques described inreference 12. Although theIFT
engaged the altitude-hold autopilot. This final condition
model wasdesigned for steady-state applications, the
was held for a minimum of 20 sec. Because the ATC
data gathered during theunsteady throttle testindicate
was still engaged, it automatically increased the throttle
reasonable accuracy forthethrottle rates used during
to maintain speed at the higher drag condition outside this phase ofthe AFF project.
the vortex. This maneuver provided both data at the desired formation position and a '°baseline" Basic Performance (nonformation) condition in one continuous ("back-to-back") data set for direct comparison.
Before conducting maneuvers for drag reduction, the Examples of the maneuver technique are presented in aircraft performance data were validated by conducting the results section of this report.
classical performance maneuvers and comparing the results to F/A-18 predictions. The basic maneuver, More than 400 data points were conducted during this initiated from stabilized cruise points, primarily consisted of a pushover-pullup (0-2 g) and a windup phase of the AFF project. A matrix of test points turn conducted at constant Mach number. These data consisting of a maximum of seven lateral and seven were compared to predictions to verify the vertical positions was flown at four longitudinal trailing positions: X = 2.0, 3.0, 4.4, and 6.6 (aft, nose-to-nose).
reasonableness of the drag data.
Formation Flight Performance Experimental Methods and Analysis Techniques Three maneuver techniques were evaluated for their suitability in obtaining the best quality of drag and fuel flow reduction data. The first technique evaluated An analysis technique was developed to consisted of flying several formation positions followed systematically and efficiently reduce the large volume of by an occasional dedicated baseline (nonvortex) performance data obtained on the AFF project. Figure 5 position. The correction for changing fuel weight was shows a summary of the performance data reduction found to be greater than desired. The second technique process.
was to fly at a given position in the vortex and acquire data. The throttle then was set to a fixed position, and Performance data were determined using classical the leading airplane quickly moved out of formation techniques. A summation of forces along the flightpath position. The change in velocity caused by the change in was used to determine drag, and a force balance drag then was measured. Large velocity changes were perpendicular to that was used to determine lift: measured using this technique as the trailing airplane responded to its change in drag, but failed to provide a direct comparison of drag in and out of the vortex at the D = cos ((Zest) F G -FRA M -FEDRAG -(G W x Axw ) same velocity.
(7) The third technique evaluated was found to be the L = (GW x Nzw ) - sin((Zest)F G most effective for determining formation flight AmericanInstituteof Aeronauticsand Astronautics Flight test database
I nginedata I I'Nsatal
Airdata computations: In-flight thrust model Gross weight, Mach number, FG' FRAM' FEDRAG ] I Wind-axiSAxw, AYw'aCcelerati°nSAzw altitude, and (Zest = 0trail - _'lead ((_est) FG - FRAM - FEDRAG - FEX Performa!ce model ___]D=cos CL, C D = GW * AXw Compare.,
l
i
C L, C D percent ACD, percent AWFT DPS-predicted performance Vortex effect = Vortex - Baseline 020208 Figure 5. The AFF performance data reduction process.
Three primary data reduction areas feed the (Zest -- 0trail - _/lead (8) performance model: the IFT model; airdata; and accelerations.
where The engine manufacturer's IFT model was used to 7lead = Olead- IXlead calculate thrust on the F404-GE-400 engines installed in the trailing F/A-18 airplane. The instrumentation This equation assumes the flightpath angle of the section describes the measurements used in this model.
trailing airplane was equal to that of the leading airplane The model calculated gross thrust (FG), ram drag ('}'trail = "}'lead), which generally is true during (FRAM), and engine throttle-dependent drag, (FEDRAG).
formation flight.
Gross thrust is the primary force the engine produces out the tallpipe, FRA M represents the force loss caused This technique was verified by comparing (Zest to the angle of attack measured by the production probes by the momentum of air entering the inlet, and during steady-state flight conditions outside of the FEDRAG accounts for the external drag forces influence of the vortex. After verification, the trailing associated with the engine nozzle and inlet spillage flow. The IFT model also accounts for bleed-air and airplane angle-of-attack probes were found to be off by as much as 1.5 ° during formation flight because of the horsepower extraction specific to the F/A-18 localized upwash influences of the leading airplane.
installation.
Because the leading airplane flew at steady-state The alrdata model computes gross weight using conditions (constant speed and altitude), "}'lead was empty weight, crew weight, and the remaining total always close to zero. The INS was used to obtain fuel. The model also provides Mach number, altitude, vehicle acceleration data. These data were corrected for and the calculation of estimated angle of attack. The rotation effects caused by the INS not being mounted trailing airplane angle-of-attack probe could not be used exactly on the center of gravity. The data then were during formation flight because of local influences of translated into the flightpath (wind-axis) coordinate the upwash field from the leading airplane. A technique system. Axial acceleration was used to compute vehicle was developed to estimate angle of attack (IXes t ) while excess thrust ( FEX ) • FEX = GW x Axw.
in the vortex by primarily using the inertial pitch angle (0) of the trailing vehicle and subtracting the flightpath To obtain drag reduction values, data obtained during angle (5') obtained from the leading airplane: formation flight (vortex) were compared with baseline American Institute of Aeronautics and Astronautics small did not include a baseline or slide-out maneuver.
(nonvortex) data. Some AFF test points did not include a slide-out maneuver to obtain baseline conditions. For For these cases, performance trend data based on gross these few points, baseline data were estimated based on weight from other slide-out points were used to estimate predictions and data trends in drag related to gross the baseline drag.
weight. The digital performance simulation (DPS) The uncertainty in the calculated coefficient of drag computer model provided predictions for a similarly (CD) is estimated to be on the order of 5 percent. The configured F/A-18 airplane. 13 This simple prediction flight test technique of comparing calculated CD in model was developed from previous flight testing and was used to evaluate the reasonableness of the baseline formation flight ( CDF F) to that of the baseline condition £ and D values.
(C D ) minimizes any bias effects. Several additional BL factors influenced the final uncertainty of the data; including variations in maneuver technique, air The CDi was calculated to compare with basic turbulence, stability of the vortex location, and the prediction theory. The in-flight CDi data were calculated using predicted values of parasite drag accuracy of the position measurement system. Another important factor was the effect fuel burn had on vortex obtained from wind-tunnel data and the following relationship: strength of the leading airplane and on the absolute drag value calculated for the trailing airplane. No corrections for trim drag effects were made. Time-averaging the C D = CDo + CDi (9) data and repeating several data points helped improve the overall quality of the results. However, the final An important element in the data analysis process was uncertainty is difficult to fully access because the factors determining the proper time period to use. The discussed above are difficult to quantify.
calculated drag plot was evaluated first to assure adequate data quality. Position data then were evaluated Results to assure the conditions closest to the desired position were used. The fuel flow data were reviewed last, and This report focuses on the AFF performance flight test the time period was adjusted to account for localized and analysis techniques developed and applied to get the variations in these data. The pilot (or ATC system, when AFF performance result. Only samples of the flight engaged) continuously pulsed the throttle to try and result are presented to illustrate this aspect. These maintain longitudinal position (or constant speed). This samples focus on one flight condition: Mach 0.56 at an pulsation resulted in cycles of increasing and decreasing altitude of 25,000 ft and a trailing position of X = 3.00.
fuel flow values. Because of the cyclic nature of the fuel Most of the AFF data were gathered at this test flow data, small adjustments (1-2 sec) in choosing the condition, and reference 15 provides a comprehensive time period could have large effects on the average fuel review of the drag and fuel flow results for all test flow value for a given time period. More reasonable conditions for this phase of the AFF project.
values of fuel flow were obtained by adjusting the time period to capture complete cycles of throttle movement.
Data quality varied during this phase of the testing This adjustment had little impact on the overall drag primarily because of the difficulty pilots incurred trying results.
to hold the specified relative position required to map the performance benefits, and because of atmospheric or A real-time drag reduction model was implemented in turbulence effects. Several positions were highly the control room using the manufacturer's IFT model to unstable, particularly those with large overlap in wing calculate thrust. Although not as sophisticated as the position. In some cases, the leading airplane wingtip postflight analysis model, the real-time drag model did vortex impinged directly on the trailing airplane tail.
provide sufficient information regarding the quality of Where possible, testing was repeated to try to improve performance data while the data were being obtained, data quality.
thus enhancing the efficiency of the data gathering. 14 Because of this capability, poor quality test points could Basic Performance be immediately repeated and other test points dropped Figure 6 shows a sample of the baseline data obtained when the previous point resulted in little or no from the dedicated baseline performance maneuvers performance benefit.
compared with those predicted from the DPS model. An To maximize the number of points obtained, some test intermittent problem with the INS accelerations was points where performance benefits were predicted to be discovered that caused a position bias in longitudinal American Institute of Aeronautics and Astronautics ..+ .... ira,, ................................................ INS data problem 'i " '__.2 ;:_ .... _"'" _ .............................................
before flight 731 :'_'-..>._._ _ _ _[_ _ :: o X 0a O z i '_/_redlctton Before flight 731 • Slide out .......................................................................................................................................................................................................................... After flight 731 -- 0 Baseline Slide out 28,000 29,000 30,000 31,000 32,000 33,000 34,000 Weight, Ib 020209 Figure 6. Sample comparison of baseline and slide-out performance data with predicted data from flight test conditions of Mach 0.56 and an altitude of 25,000 ft.
acceleration for some early test points. The INS was positioning and calculated performance data for the replaced after flight 731 and the problem was resolved. trailing airplane. Using the direct comparison with The resulting bias CD was limited to less than 5 percent baseline conditions the flight test maneuver technique of the total test points. Because CD was always biased provides, the data clearly show a large reduction in CD (approximately 10-percent) higher than actual during occurs while the airplane is in the vortex. Fuel flow also this anomaly, it resulted in conservative values of undergoes a significant change as the ATC responds to calculated percent drag reduction (approximately 0.9 of the drag change when the airplane transitions out of the actual). Many of these test points were repeated later in vortex. The cyclic nature of the fuel flow data is also the flight program, particularly those near maximum evident, particularly when using manual throttle control.
drag reduction. Except for this anomaly, the stabilized The fuel flow values for the leading airplane show no baseline test points compare favorably to the evidence the trailing airplane is influencing its predictions. These results give confidence to the validity performance.
of the performance values obtained from the AFF performance analysis code. Because of their simplicity, Data Ouality the DPS-predicted results were added to all performance data outputs (fig. 5) to assure the baseline The overall quality of each maneuver and the total maneuvers continued to yield reasonable results.
data varied because of atmospheric conditions, unstable vortex effects at some positions, and pilot technique or Figure 6 also shows excellent agreement between the experience. Turbulent flight conditions made it difficult slide-out performance data and the dedicated to obtain stabilized data at the prescribed conditions.
performance baseline points. Because of this agreement, Fortunately, most flight data were gathered on days very few dedicated baseline points were flown later in when calm weather existed at altitude. Certain positions the program.
in the vortex were very difficult to fly because the vortex impinged directly on the vertical tall or fuselage and the Drag Reduction nose wanted to wander back and forth. Some conditions Figure 7 shows a sample of an AFF test point and the completely discharged the airplane out of position.
calculated drag reduction results. The figure shows Because getting quality data at each condition was Mach number and fuel flow values for both aircraft, and desired, most unstable points were repeated (sometimes American Institute of Aeronautics and Astronautics Predicted baseline value U . ,.: .................... )..............
iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii ii iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii 400O _3000 o = 2500 c" O -50 -100 ,565 ,_ .564 _Lj_,_%_ /_FTmilingai;Piane ................................................................................................
E ,563 .5o2 ...... _,_ i o_, _=-..... i j. ........ _ .... .j,_", ........... i,'__"_'_,_ i_ ......
,561 _"_'_: _ Leading airplane .....
,560 9:24:50 9:25:00 9:25:10 9:25:220 9:25:30 9:25:40 9:25:50 9:26:00 9:26:10 Time, PDT 020210 Figure 7. Sample performance maneuver from flight test conditions of Mach 0.56 and an altitude of 25,000 ft showing drag reduction.
American Institute of Aeronautics and Astronautics more than once) to try and improve the results. After To help ascertain the overall quality of the final data gaining a feel for what to expect, the pilot usually did a set, a rating system was applied to the results obtained better job on the repeat maneuver. The real-time display from each test point. Based on the maneuver and data of aeroperformance data, in particular CD and the quality issues discussed above, a rating of "good," "adequate," or "poor" was applied. Typically, the test coefficient of lift (CL), greatly aided in assessing the quality of the data. points where large performance gains were obtained rated adequate to good. Test points that were unstable or difficult at which to fly were often rated poor. A few The pilots often had difficulty gauging when to turn random points rated poor because of extreme turbulent the ATC system on while in the vortex. Often the throttle conditions. Several data points in the region with the would immediately adjust up or down, moving the largest gains were repeated at least once, allowing the airplane out of the longitudinal tolerance requirements.
best quality of data to complete the final matrix of data.
The pilots or control room would monitor this and make Figure 7 shows an example of a data point rated good; a call to reacquire the X position. Good quality points figure 8 shows examples of adequate and poor ratings.
typically had more than 15 sec of ATC engaged while the airplane was in the desired vortex position, and maintained Mach number and altitude conditions during The problems with the region of the test point rated the slide-out segment to the baseline condition. The use poor are caused by two accumulative factors: excessive of ATC on these maneuvers clearly show the fuel dynamics of position data, and inconsistent drag values savings while in the vortex. Manual throttle use tended while the airplane is in the vortex. This maneuver to have significantly larger throttle transients while the illustrates how the pilot was able to "improve" pilot tried to maintain condition, often affecting the fuel technique during the test point and eventually obtain flow results (fig. 7). The drag values tend to be not very adequate data. Figure 8 also shows the significant drag sensitive to throttle change unless the amplitude or rate penalty that occurs when flying at positions with large is very high. wing overlap.
One challenge in evaluating the drag reduction data Table 2 shows a sample of the summary database was finding periods of time within a maneuver where showing drag and fuel flow reduction results, flight the pilot maintained proper positioning. For a conditions, and the relative positions. The position data 30-60-sec formation maneuver, sometimes only 10 sec show the variation in actual position from the target. The were adequately on the target conditions because longitudinal position was most difficult to achieve conditions were unstable. Also, 10 sec sometimes because of the lack of real-time feedback. The lateral elapsed after the slide out before the airplane settled and vertical data were more precise because of the information provided on the HUD.
down to a good baseline condition.
Table 2. Sample results data from flight test conditions of Mach 0.56, an altitude of 25,000 ft, X = 3.00 (aft), and Y = -0.125.
Measured average, Drag change, Fuel flow change, Flight conditions Target, wingspan Rated Flight Test wingspan percent percent data number point Mach Altitude, Weight, X Y Z X Y Z CD Trail Lead Trail quality number ft lbm Cdi (corrected) 727 TP14C 0.560 25,012 30,475 -3.0 -0.125-0.375 -3.05 -0.079-0.396 -6.7 -11.6 -7.6 -1.0 -6.6 Adequate 728 TP08D 0.563 25,018 31,221 -3.0 -0.125-0.25 -2.92 -0.09 -0.225 -11.0 -24.6 - 9.3 1.1 -10.3 Adequate -3.0-0.125-0.125-2.87-0.12-0.069-19.2-39.6-17.3 1.1 -18.3 Good 728 TP08E 0.563 25,024 30,839 -3.0-0.125 0.0 -2.81 -0.116-0.016-20.1 -42.9 -17.7-1.1 -16.6 Good 739 TP04E 0.563 25,029 32,337 -3.0-0.125 0.125-2.94 -0.102 0.120-10.5 -21.5 -6.2 3.0 -9.2 Good 728 TP09B 0.563 25,035 30,642 728 TP09C 0.564 25,041 30,461 -3.0-0.125 0.250-3.14 -0.126 0.214 -3.9 -8.2 -3.7 0.5 -4.2 Adequate -3.0-0.125 0.375-3.02 -0.157 0.372 1.0 2.2 1.9 0.0 1.9 Poor 728 TP09D 0.563 25,045 30,362 American Institute of Aeronautics and Astronautics conditions :-_ o 13 percent i Predicied baselihe value _iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_iii_ii_iiiiiiiiiiii_iiiiiiiiiiiiiiiiiiiiiii_iiiiiiiiiiiiiiiiiiiiiiiiii ..................................................... ) Tt ................... _ \+............... i_ _<_i_ _<<...................
O ul O ll.
-20 -40 -60 -8O ,580 ,575 ii ,.t_ i _l/i-- TTaiiing air_i_ne ii i i i ,570 e- e- Leading l# airplane _ Ii i _i_'_%'_J".,'%_ i ,/'_aj _;_ / ...... _'_"_',. j"-_.,F ......
,565 .560 10:55:00 10:55:10 10:55:20 10:55:30 10:55:40 10:55:50 10:56:00 10:56:10 10:56:20 10:56:30 10:56:40 10:56:50 Time, PDT 020211 Figure 8. Sample of test point illustrating poor and adequate data quality.
American Institute of Aeronautics and Astronautics Z = 0.37 Z = 0.25 Z = 0.13 Z=O ""C'"" Z = -0.06 --'C'-- Z =-0.13 -20 --_-- Z=-0,25 -- _} - - Z = -0.37 -25 -.5 -.4 -.3 -.2 -.1 0 .1 .2 .3 Lateral separation (Y), wingspan Figure 9. Summary of drag reduction results from flight test conditions of Mach 0.56, an altitude of 25,000 ft, and X=l.5 (aft).
The final performance and fuel savings values were increases were measured at some high wing overlap plotted as a function of lateral position for various target positions, verifying the importance of proper vertical positions (fig. 9). For this flight condition, a station-keeping to obtain the best results.
maximum of 20-percent drag reduction was calculated, with peak values at level and -13-percent vertical position (0<Z<0.13) and a lateral position of 10-20-percent wingtip overlap (-0.10 < Y < -0.20).
After discovering the peak drag reduction location, the airplane was flown at some additional test points at a vertical position of -6-percent, where more than _.2 21-percent drag reduction was measured.
g ",_ ,1 To enhance the interpretation of the results, the data were developed into Y-Z position contour plots for various longitudinal locations obtained at the two primary flight conditions. Figure 10 shows an example of a drag reduction contour plot obtained at Mach 0.56, an altitude of 25,000 ft, and X = 3.00.
_-.2 This contour plot represents the results from 42 actual test points. To complete the matrix of data required to calculate some contour plots, a small number of points --.3 near the corner positions were estimated from data trends and not actually flown. A bicubic spline was used -,3 -.2 -.1 0 .1 .2 to smooth the final contour plot data. Overall, the data Lateral separation (Y), wingspan 020213 indicate a large region of significant gains. The data are not symmetric about the peak position, and show increased sensitivity as the trailing airplane moved Figure 10. Example of drag reduction contour plot as a inboard (more wing overlap) of the peak position as function of Y-Z position from flight test conditions of opposed to outboard of this position. In fact, drag Mach 0.56, an altitude of 25,000 ft, and X=3.0 (aft).
American Institute of Aeronautics and Astronautics Fuel Reduction the rotation of the aerodynamic resultant force L2,_++D2, which can be assumed to remain constant To provide a one-to-one correspondence with the drag (fig. 1).
reduction data, fuel reduction values were also calculated for the same exact time periods used for the drag contour. In addition, the leading airplane was also Note the similarity in shape of the resultant angle plot evaluated for these periods to determine if a pattern of in figure 12 and the drag coefficient plot shown in fuel flow changes could be detected corresponding to figure7. This similarity is because drag is highly the trailing airplane position. Although no dependent on angle of attack, which changes similarly correspondence was determined, the leading airplane to how the resultant force angle does when the airplane did sometimes show fuel flow shifts corresponding to pitches down while the airplane is in formation flight.
changes in atmospheric conditions such as wind shear.
The force vector angle shows a rotation of 1.15-deg forward with respect to the horizon while the airplane is Figures 7 and 8 show this effect to some degree, and in formation flight.
how the autopilot or throttles of both aircraft try to maintain constant speed during the baseline test condition. When both the leading and trailing aircraft The induced drag reduction results (fig. 13) show have ATC engaged, they react similarly (in fuel flow excellent comparison to the simple horseshoe vortex changes) to the same atmospheric disturbance. The prediction model previously discussed. To improve the discovery of this result led to adding a correction to the overall quality of the CDi database for developing this trailing airplane fuel flow reduction based on changes in contour plot, the data were interpolated from 45 original the leading airplane as follows: data points to a finer grid (147 data points) following general trends and using extrapolation techniques to fill Percent A WFTtrai I (corrected) in missing data points along the outer edges. A bicubic (10) spline was again used to smooth the final contour plot = Percent AWFTtrai I - Percent AWFTlead data.
Table 2 shows the percent of fuel flow change for both The overall shape and magnitudes of the flight and the leading and trailing aircraft for a sample of data.
simple prediction model are very similar. The maximum Figure 11 shows the resulting contour plot using the flight-measured drag location is at a slightly lower corrected fuel flow values for the same matrix of data used to calculate the drag contour plot from flight vertical location, which is caused by the generic model conditions of Mach 0.56, an altitude of 25,000 ft, and assuming a planar wake. The size of the "sweet spot" X = 3.00 aft.
region (more than 25-percent CDi reduction) was calculated to be significantly larger in flight than in the The fuel flow reduction data trends are very similar in simple analytical model. This result is important, shape to the drag reduction data, averaging 2-3-percent less in overall magnitude. These results give confidence indicating that an AFF controller might not need to be as to the overall drag reduction values.
precise as predicted to achieve large benefits.
Validation of Basic Theoretical Predictions The flight results also show higher drag increases at large wingtip overlap than predicted by the generic Figure 12 shows the resultant force and corresponding theory, but this region is also the one where data quality angle (relative to the horizon) for the maneuver shown is worse because the points are more difficult at which in figure 7. Although the resultant force value varies to fly. Higher trim drag effects can also contribute to the more while the airplane is in formation because of large drag increases. The line of zero benefit is also located at a lower overlap position than in the simple aircraft dynamics, no significant change in overall theory. These results indicate substantially higher magnitude is seen compared to nonformation flight.
sensitivity to lateral positioning inboard of the sweet This result confirms the theoretical assumptions spot than predicted. Small changes in lateral positioning previously discussed that drag reduction is the result of in this region can result in large changes in benefits.
American Institute of Aeronautics and Astronautics _rpercen t . . . . .
.31111_ _ : ' ' : "_ .1 ......i......... ; ....... i......... :......... .......... i........
o _-.2 -.3 -.2 -.1 0 .1 .2 Lateral separation (Y), wingspan ozo2_4 Figure 11. Example of fuel flow reduction contour plot as a function of Y-Z position from flight test conditions of Mach 0.56, an altitude of 25,000 ft, and X=3.0 (aft).
35,000 97.5 ................... L_%_..i_j..!.ii._./_J...:V,J..:_.....S.....:>_._.,_._,,._ _,,_ ..............
96.5 30,000 i i ; ; i I _ Resultant aerodynamic i/_& l{_._lll.lllil jl. _, , .....
96.0 25,000 i i Slide out_(I I_ i _ I [ ............................................................................... .................. '?':e"n; I ......
95.5 20,000 .-¢ Resultant ii I ,_ A {'lip ', ', ,.,,,,o, P
..We :: i
95.0_= 15,000 94.5 10,000
I_ ,n,,ortax "I
5,000 94.0 9:24:50 9:25:00 9:25:10 9:25:20 9:25:30 9:25:40 9:25:50 9:26:00 9:26:10 Time, PDT 020215 Figure 12. Resultant force and angle values showing rotation of aerodynamic force vector while in formation flight.
American Institute of Aeronautics and Astronautics The induced drag flight results also show excellent F- 0 percent v4 comparison with the predictions for the vortex lattice : 4jl/" I _.
method (fig. 14). The horseshoe vortex model is also shown for comparison. The flight data represent a small range of vertical conditions at which the largest c /////_, --4 I _"_ reductions were calculated. The overall magnitude tends to correlate better with the horseshoe vortex predictions.
v I t S f Interestingly, on one side of the peak, the data match the o 0 vortex lattice predictions and on the other side, match -i _,_ ....
the horseshoe vortex. The data tend to diverge from both f /I methods as the wing overlap becomes greater. Again, ., / _ ........
this region was the most difficult in which to fly and in .-- which the vortex often impinged on the aircraft tail or fuselage. Overall, the data give confidence to the generic --=4 prediction models.
-,5 0 ,5 Lateral separation (¥), wingspan 020216 Long-Range Demonstration Flight (a) Measured induced drag change obtained from flight An F/A-18 cruise mission was simulated to data.
demonstrate the potential benefits of flying in the optimum formation position during extended periods.
.4 An independent chase airplane was also flown during C this mission to obtain fuel burn data for an F/A-18 Q.
airplane of similar configuration and weight. Both the m trailing and chasing aircraft were single-seat configurations of similar weight, and the leading airplane was a two-seat configuration. The independent ._o chase airplane had no data acquisition system installed.
Fuel tank readings for the chase airplane were periodically recorded during the mission (fig. 15), along with the telemetry data from the two formation aircraft.
O ,m The flight profile included flying in optimum formation position during the climb and descent -.4 -.5 0 .5 portions of the mission at altitudes greater than Lateral separation (Y), wingspan 15,000ft. The cruise condition (Mach 0.8 and an 020217 altitude of 40,000 ft) was chosen based on predictions of (b) Predicted mutual induced drag change using the best range factor for a single airplane.
horseshoe vortex model.
The results show significant fuel savings were Figure 13. Comparison of predicted with measured recorded for the trailing airplane despite significant induced drag change.
problems with the mission. Telemetry data were not available during part of the mission because the planned flight profile took the aircraft out of range. When The overall vertical sensitivity is less than predicted; and the overall shape of the region of most benefit is telemetry was reestablished, the speed brake was discovered to be partially deployed on the trailing more round than oval, as predicted for a generic wing.
The theory also predicts the maximum value is at a airplane. That anomaly was corrected for the remainder of the mission. Estimation from trends in measured fuel vertical position of wings coplanar (Z = 0). The longitudinal position had the least sensitivity to position during the mission (the dashed lines on figure 15) and is reviewed more extensively in reference 15. The determined that approximately 100 lb of fuel savings pilots were more easily able to gauge separation were not realized because of this problem. Also, the pilot flying the trailing position began to realize the distance as they moved in closer.
guidance needles were not accurate as the airplane flew farther away from Edwards Air Force Base (Edwards, American Institute of Aeronautics and Astronautics 5O
. i i ,i i i i'i'' i"i i'i ........i.................i.........i i i'
=o D.
r_ o -5O ............... Fiight data ........................... i _-Horseshoe vortex ............................................
........ _ ......... _ ......... i........................ i.................. i Vortex diameter/ i.................. i........ ......... i......... i.........
£_ Z=O_ • . : Z = -0.060 ..........i.................. i........... Wingspan = 0.03 .................... i........
Z =-0.125
!
i i -1 O0 -1.0 -.5 0 .5 1.0 Lateral separation (Y), wingspan 020218 Figure 14. Comparison ofAFF CDi change from flight test data with various generic prediction models.
Mission profile 40 kft and Mach 0.85 41 kft and Mach 0.84_ -- Formation flight _'_....
c ii_ii iii -- Lead - trail Chase - trail ,::Fq ................................................
800 t------ _ -- -- Estimated duration -- E I I speed brake was J.._ _, I !
700 I------- ---1-- -- partially deployed _ 500r-- 4-- ..,.-" : - , _5 400 1-_ ---4 __ 14percent__ - ^ I I . _,, _'_ _ _--Lo_s"t'~lO0 Ibm fuel u. 3uu J------ ---1 _ _ savings- 200 " I i 100 _ Y/-q I data out of range ;_---_ 9:00:00 9:15:00 9:30:00 9:45:00 10:00:00 10:15:00 10:30:00 10:45:00 Time, PDT 020219 Figure 15. Summary of measured fuel difference between aircraft during cruise demonstration formation flight.
American Institute of Aeronautics and Astronautics California). This problem was caused by an error maintains a constant magnitude while the airplane is in discovered in the position measurement calculation that formation flight.
was later corrected on all postflight flight data. The pilot Significant performance benefits were obtained continued the mission by flying using the experience during this flight test phase, in which pilots manually gained during the flight test program. Even with these flew the trailing airplane while in formation using problems, a 640-1bm(14-percent) fuel savings was position feedback information on the head-up display.
realized compared with the chase airplane, and more Drag reductions values of more than 20 percent and fuel than 700 lbm of savings over the duration of the flow reductions of more than 18 percent were measured formation compared with the leading airplane were at flight conditions of Mach 0.56 and an altitude of calculated. Independent checks of the fuel required to 25,000 ft. A long-range cruise mission was simulated in fill up each aircraft verified these readings to within 50 lb. the optimum formation position at conditions of Mach 0.8 and an altitude of 40,000 ft and demonstrated a 14-percent fuel flow reduction when compared with These results were converted to range improvement estimates that assumed continuation in formation at the the fuel flow of a controlled chase airplane of similar configuration.
cruise condition (Mach 0.85 and an altitude of 40,000 ft). These analyses resulted in an availability of References an estimated 110 nmi of additional range from the 640 lb of reduced fuel flow if the flight continued at the 1Hummel, D., "Aerodynamic Aspects of Formation cruise condition in the formation position.
Flight in Birds," Journal of Theoretical Biology, vol.
104, no. 3, Oct. 7, 1983, pp. 321-347.
Concluding Remarks 2Beukenberg, Markus and Dietrich Hummel, Flight test maneuvers and analysis techniques were "Aerodynamics, Performance and Control of Airplanes developed to determine the performance advantages of in Formation Flight" 17th ICAS Congress Proceedings, formation flight and validate published theory vol. 2,AIAA-A91-24301, Sept. 1990. pp. 1777-1794.
predictions. Two specially instrumented F/A-18 aircraft were flown through patterns of varying lateral, 3Hummel, D., "The Use of Aircraft Wakes to Achieve longitudinal, and vertical offset positions to obtain Power Reductions in Formation Flight," Proceedings of detailed maps of the performance benefits. A systematic the AGARD FDP Symposium on The Characterization approach to obtaining and evaluating the aircraft and Modification of Wakes from Lifting Vehicles in performance data was developed.
Fluid, Nov. 1996. pp. 36-1-36-13.
The most successful technique for obtaining accurate 4Wagner, Geno, Dave Jacques, Bill Blake, and Meir performance data consisted of flying the trailing Pachter, "An Analytical Study Of Drag Reduction in airplane at the desired formation test condition for Tight Formation Flight," AIAA-2001-4075, Aug. 2001.
approximately 30 sec, followed immediately by a "slide-out" maneuver to obtain a baseline 5Blake, William, and Dieter Multhopp, "Design, (nonformation) comparison condition. This technique Performance and Modeling Considerations for Close used the aircraft automatic pilot and automatic throttle Formation Flight," AIAA-98-4343, Aug. 1998.
features to maintain altitude and velocity during and following the slide-out maneuver. Each formation test 6Hoerner, Sighard F., Fluid-Dynamic Drag." Practical point was analyzed for drag and fuel flow benefits.
Information on Aerodynamic Drag and Hydrodynamic Resistance, Self-published work, Library of Congress Contour plots of the performance benefits were Card Number 64-19666, Washington, D.C., 1965.
obtained from a matrix of lateral and vertical positions at various nose-to-tail positions between the leading and 7Maskew, Brian, Formation Flying Benefits Based on trailing aircraft. The shape of the performance benefit Vortex Lattice Calculations, NASA-CR- 151974, 1977.
maps obtained in flight were found to have good 8Bever, Glenn, Peter Urschel and Curtis E. Hanson, agreement with published theory; and in some cases, the actual benefit magnitudes were greater than the generic Comparison of Relative Navigation Solutions Applied predictions. A simple analysis technique was also Between Two Aircraft, NASA/TM-2002-210728, July developed to validate basic theory assumptions that the 2002. Also presented at the AIAA First Technical Conference and Workshop on Unmanned Aerospace lift and drag resultant force vector rotates forward and American Institute of Aeronautics and Astronautics Vehicles, Systems, Technologies and Operations, Portsmouth, Virginia, 20-23 May 2002.
9Hanson, Curtis E., Jack Ryan, Michael J. Allen, and Steven R. Jacobson, "An Overview of Flight Test Results for a Formation Flight Autopilot," AIAA-2002-4755, Aug. 2002.
l°General Electric Co., "In-Flight Thrust Calculation Program," software program no. 83112, version 8-08-83, General Electric Co., 1983.
11Ray, Ronald J., Evaluation of Various Thrust Calculation Techniques on an F404 Engine, NASA TP-3001, 1990.
12Ray, Ronald J., Evaluating the Dynamic Response of In-Flight Thrust Calculation Techniques During Throttle Transients, NASA TM-4591, 1994.
13Orme, John S., Digital Performance Simulation Models of the F-15, F-16XL, F-18, F104, TACT F-111, X-29 and Hypersonic Research Vehicle, NASA TM-104244, 1992. (Distribution authorized to U.S.
Government agencies and their contractors.)
14Ray, R. J., J. W. Hick, and R. I. Alexander, Development of a Real-Time Aeroperformance Analysis Technique for the X-29A Advanced Technology Demonstrator, NASA TM-100432, 1988.
15Vachon, M. Jake, Ronald J. Ray, Kevin R. Walsh, Kimberly A. Ennix, "Measured Performance Benefits During the Autonomous Formation Flight Program," AIAA 2002-4491, August 2002.
2O AmericanInstituteof Aeronautics and Astronautics
REPORT DOCUMENTATION PAGE Form Approved
OMB No. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Re _orts, 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503.
1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE 3. REPORTTYPE AND DATES COVERED August 2002 Technical Publication 4. TITLE AND SUBTITLE 5. FUNDING NUMBERS Flight Test Techniques Used to Evaluate Performance Benefits During Formation Flight 706 35 00 E8 28 00 AFF 6. AUTHOR(S) Ronald J. Ray, Brent R. Cobleigh, M. Jake Vachon, and Clinton St. John 8. PERFORMING ORGANIZATION 7. PERFORMING ORGANIZATION NAME(S)AND ADDRESS(ES) REPORT NUMBER NASA Dryden Flight Research Center RO. Box 273 H-2500 Edwards, California 93523-0273 10. SPONSORING/MONITORING 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) AGENCY REPORTNUMBER National Aeronautics and Space Administration NASA/TP-2002-210730 Washington, DC 20546-0001 11. SUPPLEMENTARY NOTES Also presented at the AIAA Atmospheric Flight Mechanics Conference and Exhibit, Monterey, CA, August 5-8, 2002.
12a. DISTRIBUTION/AVAILABILITY STATEMENT 12b. DISTRIBUTION CODE Unclassified--Unlimited Subject Category 05 This report is available at http://www.dfrc.nasa.gov/DTRS/ 13. ABSTRACT (Maximum 200 words) Previous investigations into formation flight have shown the possibility for significant fuel savings through drag reduction. Using two F/A-18 aircraft, NASA Dryden Flight Research Center has investigated flying aircraft in autonomous formation. Positioning the trailing airplane for best drag reduction requires investigation of the wingtip vortex effects induced by the leading airplane. A full accounting of the vortex effect on the trailing airplane is desired to validate vortex-effect prediction methods and provide a database for the design of a formation flight autopilot. A recent flight phase has mapped the complete wingtip vortex effects at two flight conditions with the trailing airplane at varying distances behind the leading one. Force and moment data at Mach 0.56 and an altitude of 25,000 ft and Mach 0.86 and an altitude of 36,000 ft have been obtained with 20, 55, 110, and 190 ft of longitudinal distance between the aircraft. The moments induced by the vortex on the trailing airplane were well within the pilot's ability to control. This report discusses the data analysis methods and vortex-induced effects on moments and side force. An assessment of the impact of the nonlinear vortex effects on the design of a formation autopilot is offered.
14. SUBJECTTERMS 15. NUMBER OF PAGES Aircraft performance, drag reduction, drag measurement, performance tests, fuel 16. PRICE CODE consumption 17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACT OF REPORT OF THIS PAGE OF ABSTRACT Unclassified Unclassified Unclassified Unlimited NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89) Prescribed byANSI Std Z39 18 298 102