DEVELOPMENT AND FLIGHT TESTING OF THE HL-10 LIFTING BODY Robert W. Kempel*

PRC Inc. P.O. Box 273

Edwards, California 93523-0273 and

Weneth D. Painter* National Test Pilot School

P.O. Box 719 Mojave, California 93502

We have been convinced of the feasibility of a lifting entry, horizontal-landing spacecraft since we flew the M2-F1 seven years ago. . . . On the basis of our own experience, we cannot discuss the practicality of the pro- posed launch, boost, and orbit operations, nor can we assess the status of required technology in such critical areas as materials, structures, and thermal protection systems. . . . If all the other NASA centers, in conjunction with the Department of Defense and industry, can get the [space] shuttle off the ground, into orbit, and ensure that it survives the entry, we at the Flight Research Center can guarantee that it can be flown to the destination and landed safely (reference 3).

-Milton 0. Thompson

Abstract

The Horizontal Lander 10 lifting body successfully completed 37 flights, achieved the highest Mach number and altitude of this class of vehicle, and contributed to the technology base used to develop the space shuttle and future generations of lifting bodies. Design, development, and flight testing of this low-speed, air-launched, rocket- powered lifting body were part of an unprecedented effort by NASA and the Northrop Corporation. This paper describes the evolution of the HL-10 lifting body from the- oretical design, through development, to selection as one of two low-speed flight vehicles chosen for fabrication and piloted flight testing. Interesting and unusual events which occurred during the program and flight tests, review of sig- nificant problems encountered during the first flight, and discussion of how these problems were solved are pre- sented. In addition, impressions of the pilots who flew the HL-10 lifting body are given.

Nomenclature

AGL above ground level, ft AOA angle of attack, deg D drag, Ib FRC Flight Research Center, Edwards, California

&' acceleration due to gravity, 32.174 ft/sec2 h altitude, ft or m HL-10 Horizontal Lander 10

' ~ e r o s ~ a c e engineer. ~ o ~ ~ r i g h t ~ This paper is declared a work of the U.S. Govemmenr and

is not subject to copyright protection in the United States.

L lift, lb L/D lift-to-drag ratio LaRC Langley Research Center, Hampton, Virginia M Mach number MSL mean sea level MLRV Manned Lifting Reentry Vehicle NACA National Advisory Committee for Aeronautics NASA National Aeronautics and Space

Administration RC radio controlled SAS stability augmentation system USAF United States Air Force X- experimental

Background

A significant percentage of the entire planet's popula- tion has seen the space shuttle launch and its gliding return to Earth from orbital missions. Before these events could occur, significant amounts of preparation had to be com- pleted. A large part of this preparation included the suc- cessful demonstration of unpowered landings by a new class of vehicle. This paper describes the conception, design, development, and flight testing of a wingless experimental aircraft: the Horizontal Lander 10 (HL-10).

Commonly referred to as a "lifting reentry" or "lifting body" vehicle, the HL-10 contributed significantly to the development of the terminal gliding and horizontal land- ing technique currently used by the space shuttle. In the early 1950's. the concept of lifting reentry from suborbital

or orbital space flight evolved at the National Advisory Committee for Aeronautics (NACA),* Ames Aeronautical Laboratory (Ames), Moffett Field, California, through the efforts of two engineers: Messrs. H. Julian "Harvey" Allen and Alfred Eggers. Their work with the reentry survival of ballistic missile nose cones revealed that if the nose of a missile were blunt, then the reentry energy would rapidly dissipate through the large shock wave. Conversely, a sharp-nosed missile would absorb a great deal of energy in the form of heat through skin friction. Blunt-nosed vehi- cles were also more likely than sharp-nosed vehicles to survive reentry. The sharp-nosed vehicles may suffer severe damage from heating. In addition for a ballistic reentry vehicle, the maximum deceleration loads would be on the order of 8.5 g.l Using a blunt, 30" half-cone wing- less reentry configuration resulted in a high-lift-high-drag configuration. This configuration would result in maxi- mum deceleration loads on the order of 2 g or less and would accommodate aerodynamic controls. This configu- ration was also capable of a lateral reentry path deviation of approximately k230 miles and a longitudinal variation of approximately 700 miles.

Advantages of a blunt half-cone or wingless reentry vehicle configuration concept are numerous. Simply stated, lifting reentry would be achieved by flying from space to a conventional horizontal landing using such vehicles as a blunt half-cone body; a wingless body; or a vehicle with a delta planform, for example, the space shut- tle. This approach would take advantage of the ability of these vehicles to generate body lift and thus fly. These lift- ing bodies would have significant glide capability down- range (the dircction of its orbital track) as well as cross range (the direction across its orbital track) because of their ability to produce aerodynamic lift (L) during reentry.

Space capsules, on the other hand, reenter Earth's atmo- sphere on a ballistic trajectory and decelerate rapidly because of high-aerodynamic drag (D). Capsules can pro- duce small amounts of lift and large amounts of drag. Consequently, capsules are subject to high reentry forces because of rapid deceleration and have little or no maneu- vering ability. Figure l2 represents the hypothetical orbital track of a spacecraft following a launch from NASA Kcnncdy Space Center, Florida. The small triangular area labeled Mercury, Gemini, and Apollo, off the southeast coast of the United States, represents a typical landing footprint of a capsule vehicle. By contrast, the lifting body landing footprint for a hypersonic Mach number greater than or equal to 5 (M 2 5) lift-to-drag ratio ( L P ) of approximately 1.5 includes the entire western United States and parts of Mexico. This difference represents a significant improvement over a capsule.

* The NACA was the predecessor of the National Aeronautics and Space Administration (NASA).

The L/D is very important to an airplane, particularly one without power. This ratio is a direct measurement of how far an unpowered glider, sailplane, or airplane can glide. The higher the number, the farther the glide. Every- body wants to reduce drag. At an L P = 20 and at a given airspeed, for every 20 ft forward the sailplane moves, it will sink 1 ft for a glide ratio of 20 to 1 and have a glide angle of less than 3" nose down. Lifting bodies are vehi- cles with subsonic L P of approximately 3 to 4 at the best.

Program Concept The HL-10 was not the first lifting body tested at the NASA Flight Research Center (FRC), Edwards, Califor- nia.** In the spring of 1963, the Dryden M2-F1, a blunt half-cone configuration, was the first lifting body tested (Fig. 2). Lifting bodies were envisioned to be a new manned research program by Mr. R. Dale Reed, FRC, an innovative engineer and private pilot. Mr. Reed reviewed the plan for the Apollo mission to the Moon and return to Earth and found that the plan called for a ballistic reentry capsule. In addition, the program planners considered a lifting reentry vehicle configuration as still too risky, although at this time the Saturn booster had provided suffi- cient thrust and reliability.

Mr. Reed reasoned that if a lifting body could be built which would demonstrate a horizontal landing, then such a demonstration would build confidence within NASA that this class of vehicle could be used to great advantage. Having long been interested in work at NASA Ames with the M2 lifting body under the direction of Mr. Eggers, Ames Deputy Director, Mr. Reed contacted him and pro- posed the idea of building a large-scale, piloted, demon- strator lifting body vehicle. Mr. Eggers thought that the idea was good and told Mr. Reed to pursue it further. Mr. Reed proceeded, on his own, to build a small, free-flight M2 model which was towed aloft by a large radio- controlled (RC) model and released. His wife, Mrs. Donna Reed, took some 8-mm home movies of successful glide flights and landings.

Next, Mr. Reed approached Mr. Milton 0. Thompson, an FRC X-15 test pilot, and got him interested in the con- cept. Mr. Thompson agreed to fly such an unusual configu- ration if wind tunnel tests validated the design, even though it had the gliding characteristics of a well-polished brick. Armed with the movies and other presentation mate- rials, Messrs. Reed and Thompson briefed Messrs. Paul F. Bikle, FRC Director, and Eggers. With Mr. Thompson's assurance that he was a proponent of this concept, Messrs. Bikle and Eggers agreed on the spot.

With a modest budget and some dedicated volunteer help, a small team headed by Mr. Victor Horton. FRC, was established. Other team members included Messrs. Reed,

** In 1956, the NACA High-speed Flight Station, Edwards, California, was renamed the NASA Flight Research Center. In turn, the FRC was renamed the NASA Dryden Flight Research Center in 1976.

Dick Eldredge, and Richard Klein. This team enlisted the aid of Mr. Gus Briegleb, a well-known glider builder and operator of the nearby El Mirage Dry Lake glider port, and the design and construction of the M2-F1 was launched.

The M2 was basically a 13' blunt half-cone that was flat on top and round on the bottom. What resulted was a rather unusual creation which was nicknamed "The Flying Bathtub." This vehicle consisted of a steel tube primary structure covered with plywood, cockpit with minimal instruments, control surfaces, and landing gear (Fig. 2). The entire M2-F1 program was completed for less than $30,000, an unheard of sum even in those days. Reference 3 presents an excellent review of the overall lifting body programs at FRC. The M2-F1 program was successfully completed in August 1964.

With the initial successes of the M2-F1 program, Messrs. Bikle, Thompson, and Reed traveled to NASA Headquarters, Washington, D.C., with their presentation materials and proposed a follow-on program which called for the design and construction of two heavyweight alumi- num M2 vehicles. One of these vehicles would be reserved as a backup. This proposal called for the vehicle to be carried aloft and launched using the NASA B-52 (Boeing Aircraft Company, Seattle, Washington) aircraft. This B-52 had been structurally modified and configured for the launching of the joint NASA and United States Air Force (USAF) X-15 hypersonic research aircraft. While at NASA Headquarters, it was proposed to the FRC people that the Langley Research Center (LaRC), Hampton, Vir- ginia, HL-10 be included in a flight test program as the second candidate configuration. The FRC group agreed with this proposal. As a result, NASA Headquarters approved the program and funding for the construction of two heavyweight vehicles: the Ames M2-F2 and the LaRC HL- 10.

Configuration Design Evolution

Many hypersonic configuration studies were conducted to evaluate various candidate aerodynamic shapes within the Aero-Physics Division at LaRC. These studies indi- cated that as an advantage a reentry vehicle with negative camber and a flat bottom, rather than a blunt half-cone, may provide higher trimmed L/D over the angle-of-attack (AOA) range. In 1957 during theoretical trade-off studies and wind tunnel experiments, this negative-camber con- cept was used in developing a configuration stable about its three axes. A flat lower lifting surface was retained for improved hypersonic lifting capability. This vehicle was first referred to as a "manned lifting reentry vehicle" (MLRV). It w8s found that a vehicle with the combination of a nose tilted up at an angle of 20, an aerodynamic flap, and a flat bottom would be stable about the pitch, roll, and yaw axes. In addition, such a vehicle would trim at an AOA up to approximately 52O at a lift coefficient in excess of 0.6. This vehicle configuration, now referred to as a "lifting body," would also retain higher trimmed L/D over

lower AOA when compared with a vehicle with a 0' nose tilt. Advantages of a lifting body over a capsule included relatively high-lift-high-drag ratio characteristics as com- pared with zero-lift-high-drag. In addition, lifting bodies could achieve the specified goals because of the ability to be maneuvered during orbital reentry and in the terminal landing flight phase similar to conventional airplanes. The ability to control roll and pitch axes (to control the direc- tion and magnitude of the lift vector and hence the flight- path) was also considered a great advantage. This control was to be achieved by the use of either reaction jets, aero- dynamic control surfaces, or both.

In 1962, a negatively cambered lifting body configura- tion emerged and was designated the HL-10. Camber, the curved part of a wing surface, was upside-down. Figure 3 shows an original 1962 sketch of an MLRV known as the HL-10. The established mission goals were as follows:

1. Hypersonic L/D without elevon deflection of approximately 1

2. High trimmed lift at hypersonic speeds

3. Subsonic L/D of approximately 4 4. High volumemc efficiency, 12-person capability

5. Acceptable body shape at all speeds

6. Statically stable and controllable at all speeds 7. Launch vehicle compatibility

8. Low heating rates and loads comparison 9. Low AOA for a given subsonic lift coefficient

10. Reduced subsonic flow separation

The symmetrical configuration met 5 of these 10 mis- sion goals. On the other hand, the negative-camber config- uration met 9 of the 10. The negative-camber configuration did not meet goal 9: low AOA for a given subsonic lift coefficient.

Configuration Development

With NASA Headquarters approval for a heavyweight lifting body program, the FRC team compiled the require- ments and specifications for the two lifting body vehicles. With the statement of work completed, the NASA Request for Proposal, PR-2694, was issued to various manufactur- ers for the design and construction of two heavyweight, aluminum-structured lifting bodies consisting of the M2- F2 and the HL-10. On April 13,1964, the Northrop Corpo- ration, Hawthorne, California, submitted proposal NB 64- 31, titled "Design and Fabrication of Two (2) Research Lifting Body Vehicles, M2/HL-10," to NASA. This corpo- ration was subsequently awarded contract NAS4-603 to design and build the two vehicles.

Meanwhile, wind tunnel tests at LaRC revealed that the basic configuration trimmed subsonic L/D was only

slightly in excess of 3. This finding was considerably less than the established goal of a subsonic L/D of 4. In addi- tion, negative values of directional stability existed at low supersonic speeds and at some AOA. To rectify this situa- tion and to increase subsonic L P , an ejectable tip fin scheme was briefly considered; however, the ejection of tip fins during the final phase of a mission was considered unacceptable. From wind tunnel results, a tip fin configuration was developed that included changes in the tip fin shape which resulted in increased area, toe-in angle, and rollout angle that provided the required subsonic trimmed maximum L/D. In addition, simple two-position flaps were added to the trailing edge of the tip fins and upper elevon to vary the base area and, consequently, the subsonic base drag. Closing these flaps would minimize base drag. This modification also reduced the directional stability problem. This change was now required to be incorporated into the design specification.

On February 3, 1965, almost 10 months following con- tract award, a meeting was scheduled at FRC to present the modified tip fin and two-position flap proposals. Sev- eral top engineers from LaRC presented the proposal to add six control surfaces to the HL-10. These additions would include two-position surfaces. These surfaces con- sisted of two elevator flaps located on the upper surface of the elevon and tip fin flaps. These additional surfaces meant that a design change and modification to the exist- ing contractual agreement was now required. The change was made but did not have overwhelming support from the FRC team at the time. Later in the program, however, the change was viewed as one of the best decisions made. This modification allowed a simplified flight control design and permitted the vehicle to fly from subsonic to supersonic speeds with less trim change in the pilot's con- trol stick position. Figures 4(a) and 4(b) show a side-view comparison of the basic and final configurations. The enlarged center and tip fin modifications on the final con- figuration are obvious in these figures.

The delivery of the M2-F2 occurred on June 15, 1965, and of the HL-10 occurred on January 18, 1966. Follow- ing delivery of these vehicles to FRC, the next phase of the lifting body program began. This phase involved installing and testing the extensive flight test instrumenta- tion, vehicle systems, and subsystems and lasted approxi- mately 1 year for each of these vehicles. The first heavyweight lifting body flight, the M2-F2, took place on July 12, 1966, 2 years following contract award, with a glide flight from an altitude of 45,000 f t with Mr. Thomp- son at the controls.

Development of the two heavy lifting bodies required an unprecedented NASA and contractor effort. Each of the involved organizations contributed talents and resources to the fullest. The program was based on the ideals of innovation, initiative, and simplicity above all where pos- sible. Unneeded management, unnecessary paperwork,

and red tape were eliminated. Engineers and technicians from FRC worked with their contractor counterparts at the Northrop Corporation facility. The result was a superior end product with no cost overruns or significant schedule delays. When the final cost was totaled, these vehicles cost $1.2 million each; an unheard of price, even in 1965, for a new research aircraft.

Vehicle Description

The final flight configuration (Fig. 5) was a single-place vehicle with a relatively conventional 1960's aircraft cock- pit and instrument panel (Fig. 6). This configuration con- sisted of a negatively cambered airfoil with a 74' sweep- back delta planform with three aft vertical fins. The vehi- cle length was 21.17 ft. Figure 7 shows the critical dimen- sions and physical characteristics. Vehicle launch weight, with propellants, was 10,009 Ib. Landing weight was 6,473 lb. Center of gravity ranged from 53.14 percent of the body length for the launch weight configuration to 51.82 percent for the landing condition.

Rocket power was provided to boost the vehicle to test Mach numbers and altitudes. The rocket motor was an upgraded, off-the-shelf item which had been used during earlier programs at FRC. This rocket motor consisted of a four-chambered XLR- 11 RM- 13 which produced 2120 lb of thrustfchamber at 265 lb/in2 chamber pressure. Individ- ual chambers could be operated for thrust modulation to achieve the desired flight test conditions. Liquid oxygen was used as the oxidizer, and a water-alcohol mixture was used as the fuel. Total propellant weight was 3536 lb. The oxidizer and fuel were delivered to the chambers by a tur- bopump driven by decomposed hydrogen peroxide. Typi- cal rocket motor burntime lasted approximately 90 to 100 sec at maximum thrust using four chambers.

No new rocket motors were used during this program. These rocket motors had previously been used in the early X-15 program. At that time, at least one XLR-11 RM-13 motor had been loaned to a museum for display. As a result, the FRC team had to obtain the return of this motor before using it in this program.

In addition to the XLR-11 RM-13 primary rocket engine, two small hydrogen peroxide landing rockets were installed in the vehicle. These rockets were capable of producing 500 lb of thrust each for 30 sec and were pro- vided in the event that a landing approach needed to be extended. These rockets were never used as an aid to land- ing. The only time these rockets were used was for experi- mentation purposes during the last phase of the program.

The pilots on the FRC team required that the design specification include speed brakes to provide added drag, on demand, much like an inverse throttle to vary the land- ing pattern parameters. In addition, the weight of these brakes was minimal, and they did not require fuel. These requirements were later specified for the space shuttle.

Aerodynamic control was provided by the primary con- trol surfaces, elevons, and rudder. Symmetric deflection of the elevons provided pitch control, and differential deflec- tion provided roll control. A split rudder on the center ver- tical fin provided yaw control and speed brake. Pitch, roll, and yaw damping were provided through the limited authority stability augmentation system (SAS) to the elevons and rudder. Trim was provided by the elevons for pitch and roll and by the rudder for yaw or directional trim.

The limited authority SAS provided angular rate feed- back about all three axes for damping augmentation oper- ating through servoactuators. The pilot could select SAS gains via switches on the SAS control box. This box was located on the left-hand console. This system was all ana- log (electrical wires connected to resistors, capacitors, and operational amplifiers) with an electromechanical inter- face and was relatively simple when compared with the digital computer-based flight control systems of today.

Secondary movable surfaces were located on the inboard and outboard trailing edges of the tip fins and the upper surface of the elevons. These electric motor- actuated surfaces were two-position flaps to either a closed position, the subsonic configuration, or opened for the transonic configuration (Figs. 8(a) and 8(b)).

The landing gear consisted of off-the-shelf parts from several airplanes. Main gear wheels, tires, brakes, and gear and door toggle locks were T-38 hardware. The main gear shock strut consisted of F-5A hardware. The nose gear shock strut, wheels, and tires were comprised of T-39 hardware. Main landing gear and nose gear were pneumat- ically actuated and had extension times of approximately 1.2 and 1.5 sec, respectively. Once lowered, this gear could not be retracted while airborne.

Flight Mission

The HL-10 was carried aloft by the NASA B-52 launch aircraft (Figs. 9(a) and 9(b)). This aircraft had been modi- fied earlier specifically to launch the X- 15 hypersonic air- planes from a right-wing pylon. To carry and launch the lifting bodies, a special adapter was constructed and fitted to the B-52 wing pylon. Use of this airplane was compli- cated because the X- 15 program was still operational.

A typical HL-10 mission involved a launch at an alti- tude of 45,000 ft and at Mach 0.65. Eleven glide flights preceded the powered flights, so the pilots could become familiar with the vehicle handling qualities and aerodynamic characteristics, and the vehicle systems could be checked out. Figure 10 shows the ground track of flights in the terminal approach and landing pattern. Launch point for the powered flights was located south- west of the glide flight launch point by approximately 40 miles. During flight, ground radar tracked the vehicle and provided mission control with ground track and alti-

tude information. Deviations from planned profiles, such as high- or low-energy states, were radioed to the pilot for the appropriate corrective action. The low-key point shown on the ground track occurred at an altitude of approximately 20,000 ft. At this point, research data acqui- sition was terminated, and the full attention of the pilot was given to the landing approach pattern. The landing approach technique used in these lifting body programs was similar to the one developed and used in the X-15 program.

The average glide flight lasted 4.2 min, and powered flights lasted approximately 6.7 min. The average rate of descent in gliding flight approached 11,000 ft/min. One pilot indicated that if a brick were dropped from the B-52 at the same time that a lifting body were launched, the lift- ing body would beat the brick to the ground.

Most lifting body landings occurred on Rogers Dry Lake, runway 18, Edwards, California. This runway was approximately 4.5 miles or 24,000 ft long. The final approach and landing flare were accomplished by estab- lishing a preflare aim point (Fig. 11) during the 270 to 300 kn final approach (approximately 4000 ft above ground level). The unpowered approach and landing of the HL-10 was relatively typical of each of the lifting bodies. Land- ing occurred in three parts: final approach, flare, and post- flare deceleration (Figs. 12(a), 12(b), and 12(c)). The final approach was typically done at 300 kn at a flightpath angle of 16 to 18" nose down. Flare initiation usually initiated at 300 to 270 kn and 1000 ft above ground level. The flare was done at approximately 1.5 g to bring the vehicle to a relatively level flight attitude at an altitude of approxi- mately 100 ft. At this altitude, the speed of the vehicle had decreased to approximately 220 to 240 kn, and the landing gear was lowered. The postflare deceleration was made with touchdown between 155 to 223 kn (Fig. 13). Once the landing gear was down, maximum L/D was reduced by approximately 25 percent. To the pilots, however, the land- ing speeds were no problem. One advantage to the high speeds was that the handling qualities were improved.

Flight Testing

Before every lifting body free flight, a series of captive manned check flights was planned to evaluate the lifting body systems and subsystems. The HL-10 completed two captive flights in late 1966.

The HL-10 team convinced Mr. Bikle as well as the rest of NASA and USAF management that all the necessary engineering, systems, and mechanical work on the airplane, piloted simulations, paperwork, and briefings were completed. On December 21, 1966, the HL-I0 was placed beneath the right wing of the B-52 aircraft and lifted into position. Next, the HL-10 was attached to the B- 52 aircraft, and preflight checks were completed. Early the next day all preparations for the first free flight were com- pleted. Mr. Bruce Peterson (Fig. 14) took his place in the

cockpit, and the crew strapped him in. The takeoff of the B-52 aircraft was smooth, and the prelaunch HL-10 checks were satisfactory. Everything was now ready. The flight plan called for a launch point approximately 3 miles east of the eastern shoreline of Rogers Dry Lake abeam of the landing lakebed runway 18. Launch heading was to be to the north with two left turns. The launch point was almost hrectly over the USAF Rocket Propulsion Test Site, Edwards, California. This ground track looked simi- lar to a left-hand pattern with the launch on the down- wind leg, a base leg, a turn to final approach, and a final approach to runway 18.

Launch from the B-52 aircraft occurred at an altitude of 45,000 ft and an airspeed of 170 kn. This launch was simi- lar to simulator predictions. Airplane trim was much as expected although pilot Peterson sensed what he described as a high-frequency buffet in pitch and some in roll. Later this problem was identified as a limit cycle. As speed increased, the oscillation became noticeably worse. In addition as the first turn was completed, the pilot noticed that the pitch stick sensitivity was excessively high. This stick sensitivity resulted in too much pitching motion as a result of a relatively small movement of the pitch stick by the pilot. As the flight progressed, the high-frequency limit cycle increased in amplitude, and excess sensitivity in the longitudinal stick became even more obvious. Difficulties in the roll axis were masked by the pitch problems. The landing was completed somewhat prematurely because of the sensitive control problem. The landing flare was initi- ated at approximately 320 kn with touchdown at approxi- mately 280 kn or approximately 30 kn faster than anticipated. Flight time was 189 sec (3 min and 9 sec) from launch (45,000 ft mean sea level (MSL)) to touch- down (2,300 ft MSL). Average descent rate was almost 14,000 ft/min.

Mr. Peterson was greatly concerned with the pitch sen- sitivity and limit cycles. Figure 15 shows the flight control system limit cycle time histories for the first flight of the HL-10 lifting body. The amplitude of this sensitivity and limit cycles became larger as a function of the vehicle air- speed and system gain setting. The pitch gain was adjusted several times in an attempt to alleviate the problem. The limit cycle was a 2.75-Hz oscillation feeding through the SAS. The problem was primarily in the pitch axis, was most severe during the last thud of the flight, and contin- ued to exist despite the fact that the pitch SAS gain was reduced from 0.6 to 0.2 degldeglsec. Toward the end of the flight, the pitch limit cycle oscillation magnitude was approximately 0.4 g peak-to-peak at 2.75 Hz.

The first flight was very disappointing. The FRC team found the results quite poor when compared with the pre- flight simulations and analyses. Fortunately, FRC manage- ment was patient. In addition, the pending holiday season provided the team with some time to work.

Following the holidays, initial discussions seemed to lead the team to the conclusion that if the stick sensitivity were fixed and the SAS gains were decreased, then another flight could probably be tried. On the other hand, one lone dissenter in the group was not convinced that team had entirely understood all the problems. Mr. Weneth D. Painter, FRC, continued to analyze the flight data and argue against another attempt even though the project pilot, Mr. Peterson, had convinced Mr. Bikle that a second flight should be attempted.

The team initiated an in-depth, unified analysis of the flight data at the beginning of 1967 with a fresh perspec- tive. Each member knew what the job was and expended maximum effort to understand exactly what happened on that first flight and to fix the problems, whatever they were. This analysis resulted in identification and correc- tion of the following problems:

Large amplitude limit cycles in the SAS because of a 2.75-Hz elevon oscillation feeding through the SAS.

Extreme sensitivity in the longitudinal stick arising from the high pitch stick gearing of 6.92" of elevon travel per inch of stick travel which resulted in large vehicle motions because of small stick deflections. Lack of longitudinal- and lateral4uectional control during portions of the flight.

As previously stated, the first two problems were immedi- ately identified even before touchdown. The third problem was more illusive, and its identification and resolution are discussed later in this section.

The first problem, large amplitude limit cycles, was apparently caused by higher elevon control effectiveness than had been predicted during wind tunnel simulations and by feedback of a 2.75-Hz limit cycle oscillation through the SAS. This problem was partially solved by modifying the structural resonance, 22-Hz mode, lead-lag filter which had been installed before the first flight. This modification consisted of a notch filter and a lead-lag net- work in the SAS electronics. The problem was totally solved with the combination of installing the structural notch filter with a center, or notch, frequency of 22 Hz and of using decreased SAS gains.

The second problem, stick sensitivity, was solved by a relatively simple gearing modification of the longitudinal stick. This type of problem was easy to miss when prepa- rations for flight were completed on a fixed-base engineer- ing simulator.

A third, more illusive, problem was not really apparent to the pilot or test team during the initial postflight analy- sis. This problem affected controllability of the vehicle at some points in the flight profile. The program stability, control, and handling qualities engineer, Mr. Robert Kempel, FRC, launched an in-depth investigation.

The team had generated the HL-10 simulation from wind tunnel results. Before the first flight, h e assumption that this model was a relatively accurate representation of the actual flight vehicle had been made. Now, a logical approach was taken to verify that assumption. If flight- recorded control inputs were entered into the computer- ized model, and this model was a reasonable representa- tion of the flight vehicle at the same flight conditions, then the model dynamics (calculated motions) should be simi- lar to those of the flight vehicle.

Twelve maneuvers from the first flight results were selected as candidates for computer matching. These maneuvers varied from 5 to 15 sec in duration. Seven of these twelve maneuvers were successfully matched; how- ever, seven matches were considered marginally accept- able. This finding called for a further examination of the flight data.

The team decided to play the entire flight-recorded data back through the ground station. These data were recorded on magnetic tape, and the results were reviewed. This time, however, the team reselected the parameters which would be grouped together. These data were now arranged in the best logical organizational manner which would facilitate logical postflight analysis, so the physical rela- tionships between certain sets of data provided increased insight into how the vehicle was behaving dynamically. The arrangement of the control room strip charts had been laid out in a somewhat random manner and did not lend itself to assessment of families of data. The three sets of data, presented in the revised postflight manner, indicated some very interesting features. However, during certain portions of the flight, some of the traces would become blurred and fuzzy, particularly the control surface strain gauge data when some high-frequency disturbance appeared. With these data lined up on a common time interval, many data traces displayed a similar phenome- non. The question became identifying why this phenome- non occurred. This particular problem was not specifically apparent to Mr. Peterson, and he made little or no com- ment regarding it during the flight or postflight discus- sions. Something did, however, disturb him relative to the vehicle response to control input which caused the team to investigate further. It was apparent that each time this situ- ation occurred, the AOA was above 1 l o to 13". As AOA decreased through these values, the ailerons suddenly became very effective by producing a roll angular rate of 30 to 45 degJsec.

The team began to think that a possible massive flow separation over the upper aft portion of the vehicle at these high AOA may have caused the control surfaces to lose a large percentage of effectiveness. Figure 17 presents a 55- sec recording of the inboard right and left tip fin strain gauge data traces from flight. The flight-measured AOA trace in this figure is indicated actual, not corrected, AOA. Significant postflight AOA corrections were required for

such things as the angular difference between the nose boom and the vehicle longitudinal reference axis, up- wash, boom bending because of normal acceleration and pitch angular rate.

Angle of attack is included in Fig. 16 as a qualitative indicator of the flow separation. An AOA of 0" on this scale corresponds to approximately 7' corrected and is shown for comparison. As the AOA is reduced through 5" at point A, the flow attaches abruptly. Between 5 and 10 sec, the AOA is increased to approximately 8". and the flow separates. The trace gets fuzzy. At approximately 34 sec, the AOA is reduced through 7S0, and the flow becomes dramatically attached once more at point B. As the AOA decreased, the airflow would suddenly reattach, and the controls would behave in their normal fashion.

Repeated analyses of these data caused the flow separa- tion theory to seem increasingly plausible. Still, the wind tunnel data did not indicate a problem to the degree that had been experienced in flight. These data also indicated a significant loss of L/D above Mach 0.5 at an AOA of 12". These data further convinced the team that the problem was caused by massive flow separation.

About this time, the FRC team decided to call the team at LaRC and give them a preliminary assessment of the findings. With LaRC's urging, the FRC team traveled to Virginia to present first-hand these data and the hypothesis that the problem was caused by massive flow separation. At one point during the ensuing meeting, Mr. Robert Tay- lor, LaRC, jumped up from the table and angrily slammed a mechanical pencil to the floor as he gave forth a stream of oaths! Needless to say, everyone was shocked by this outburst. When Mr. Taylor had calmed down, he resumed his place at the table and said "I knew that this would be a problem!" He had had a gut feeling that the flow separa- tion which the LaRC team had seen on the wind tunnel model would, in fact, be worse in flight. The LaRC team indicated that they would give the problem immediate attention and propose a remedy. The FRC team felt good about the meeting, departed for California, and agreed that the next move was up to LaRC.

Throughout the winter and spring of 1967, the LaRC team worked on the problem and came up with two possi- b:e fixes. These fixes were identified as modification I and modification I1 (mod I and mod 11). Both modifications concentrated on changes to the outboard vertical fins (Fig. 17). During the summer of 1967, Mr. Kempel plotted all the data for mod I and mod I1 as a function of AOA for constant Mach numbers. When complete, these data were lined up for comparison. Some of the little wiggles (nonlinearities) present in the original data were not present. The FRC team hypothesized that if these nonlin- earities indicated flow separation, then the lack of these nonlinearities would indicate no flow separation or scpara- tion to a lesser degree. Based on this premise, mod I1 was selected as the appropriate fix.

To accomplish mod 11, the final aerodynamic configura- tion change had to be made to the HL-10. The FRC con- tracted with Northrop Corporation, in early autumn 1967, to design and install the mod I1 configuration change. Northrop and NASA decided that a fiberglass glove, backed by metal structure, would accomplish all configu- ration objectives very nicely.

In the spring of 1968, the final stages of vehicle prepa- ration were nearing completion. Flight controls, aerodynamic configuration, and internal systems correc- tions were completed. With the injury of Mr. Peterson in the M2-F2 landing accident, Captain Jerauld "Jerry" Gen- try, USAF, was named as HL-10 program pilot. Figure 18 shows Captain Gentry, who worked as a true professional and gave the flight preparation his complete attention. After many hours of simulation time, he was finally ready to go fly.

The second flight occurred on March 15, 1%8, with Captain Gentry at the controls. This relatively typical lift- ing body flight was launched from an altitude of 45,000 ft at Mach 0.65. The flight plan called for pitch and roll maneuvering to allow the pilot to get the feel of the air- plane. Mild pitch and roll maneuvers were performed up to an AOA of 15" to evaluate the possibility of control degradation similar to that which occurred during the first flight. In general, the flow did not significantly separate, and no degradation of control occurred; however, some sensitivity to AOA was observed. Flight time from B-52 launch to touchdown of the HL-10 vehicle lasted approxi- mately 4.4 min.

Problems, there were none. The flight occurred exactly as planned and was a resounding success from every- body's point of view. The pilot found that the HL-10 per- formed as well as an F-104 airplane when making a similar approach. Throughout the life of the program, the HL-10 underwent minor adjustments to make it the best of the best. Before the program ended, the HL-10 went on to fly 35 additional successful flights. Five pilots participated in this program. Several technical "firsts" which occurred during this program are discussed in the following subsections. First Powered Flight

The first lifting body powered flight was attempted with the HL-I0 on October 23, 1968, with Captain Gentry at the controls. The rocket failed shortly after launch requir- ing propellant jettison and an emergency landing on Rosamond Dry Lake, 10 miles southwest of Rogers Dry Lake, but within the boundaries of Edwards AFB, Califor- nia. The first successful lifting body powered flight was subsequently made on November 13,1968, with Mr. John Manke, FRC, at the controls. Figure 19 shows Messrs. Weneth D. Painter, Herbert Anderson, Jack L. Kolf, Manke, and Joe Huxman, FRC.

First Supersonic Flight The first supersonic flight achieved by a lifting body

was completed on May 9, 1969, by Mr. Manke during flight 17 of the HL-10 program. On this date, the HL-10 reached a maximum altitude of 53,300 ft and Mach 1.13. This flight went according to plan. Mr. Manke later reported that during the flight there were no significant problems, and "everything went real well." Fastest and Highest Flight

The HL-10 was the fastest and the highest flying of any of the lifting bodies. Figure 20 shows Major Peter C. Hoag, USAF, who achieved Mach 1.86 on February 18, 1970, during flight 34. The duration of this maximum Mach number flight from B-52 launch to touchdown was 6.3 min. This speed was the fastest that any of the lifting bodies achieved.

Figure 21 shows test pilot William H. Dana. FRC, who reached an altitude of 90,303 ft 9 days later during flight 35. The flight to maximum altitude from B-52 launch to touchdown lasted 6.9 min. This flight was at the highest altitude any of the lifting bodies would achieve.

Significant Contributions and Lessons Learned

Design, development, and flight testing of the low- speed, air-launched, rocket-powered HL-10 lifting body was part of an unprecedented effort. NASA Langley Research Center conceived and developed the vehicle shape and conducted numerous theoretical, experimental, and wind tunnel studies. NASA Flight Research Center was responsible for the final, low-speed (Mach numbers less than 2.0) aerodynamic analyses, piloted simulations, control law development, and flight tests. The prime con- tractor, Northrop Corporation was responsible for hard- ware design, fabrication, and integration.

The HL-10 completed a successful 37-flight program, achieved the highest Mach number and altitude of this class vehicle, and contributed significantly to the technol- ogy base used to develop the space shuttle and future gen- erations of lifting bodies. This program

proved that changes in program structure which per- mit decisions to be made at the technical and engi- neering level eliminate unneeded layers of management and unnecessary paperwork and result in surprisingly low unit cost, for example the $1.2 million for the M2-F2 and HL-10 lifting bodies,

demonstrated that hybrid simulations could be cre- ated by interfacing analog computers with high- speed digital computers to generate complex, nonlin- ear, aerodynamic functions, and

assisted in demonstrating the importance and inher- ent reliability of speed brakes for unpowered reentry vehicles.

In the process of making such contributions, several lessons were learned. This program showed that

piloted entry vehicles could adequately complete rel- atively steep high-energy approaches, and such approaches were accurate and safe operational tech- niques, lifting bodies flying steep, high-energy approaches and equipped with speed brakes could spot land with an average miss distance of less than 250 ft, vehicles with very high dihedral effect, a characteris- tic of lifting bodies, could be flown safely, and powered landings using shallower approaches than the steep ones normally used with the unpowered HL-10 lifting body provided few benefits.

References 1. Faget, Maxime A,, Benjamin J. Garland, and James

J. Buglia, "Preliminary Studies of Manned Satel- lites-Wingless Configuration: Nonlifting," NACA Conference on High-speed Aerodynamics, NACA TM X-673 19, Ames Aeronautical Laboratory, Mof- fett Field, CA, Mar. 18-20, 1958, pp. 19-33.

2. McTigue, John G. and Bertha M. Ryan, Lifting- Body Research Vehicles in a Low Speed Flight-Test

Program, International Congress of Subsonic Aero- nautics, New York, Apr. 1967.

3. Flight Test Results Pertaining to the Space Shuttle- craft, NASA TM X-2101, 1970.

Acknowledgment

Mr. Milton 0. Thompson, NASA Dryden Flight Research Facility Chief Engineer, supported and encour- aged the generation of this report. In addition, he provided many outstanding suggestions and useful comments regarding its content. Mr. Thompson flew the first light- weight and heavyweight M2 vehicles. The first heavy- weight lifting body flight, the M2-F2, took place on July 12, 1966, with Mr. Thompson at the controls. Without his outstanding ability as a research test pilot and engineer, none of the work described in this report would have taken place. Without his vision, in all probability, the success of the space shuttle never would have been seen. He was truly a man of far reaching vision.

On August 3, 1993, Mr. Thompson completed his final editorial review of this report. On the evening of August 6, 1993, NASA planned a dinner celebration in honor of Mr. Thompson. This dinner was to be a tribute from his coworkers, friends, and colleagues. However on the morn- ing of August 6, he passed from this life. The tribute was held anyhow. The attendees remembered the 37 years of NACAINASA service he rendered as a true professional and as an aerospace pioneer, remembered him as a friend, and remembered him as a visionary. Milton 0. Thompson, we salute you.

Fig. 1. Orbital entry footprints.

Fig. 2. The M2-F1 lifting body in flight.

Fig. 3. Original, informal, 1962 sketch of the proposed HL-10 configuration.

461

3.42 m (1 1.22 ft)

(a) Basic.

6.45 m (21 -17 ft)

(b) Final. Fig. 4. The HL-10 lifting body configurations.

EON l Obi Fig. 5 . The HL- I0 lifting body as it appeared at rollout at Northrop Corporation.

E 20304 Fig. 6. Instrument panel arrangement.

Bottom view Top view

/ Elevon flap E1evOn 7aii

Inboard tip fin flap

(21.1 7 ft) Rudder and speed brakes

(1 1.22 ft)

Fig. 7. The HL-10 lifting body. three-view drawing.

E 21536 k2153 (a) Transonic. (b) Subzonic.

Fig. 8. The HL- 10 lifting body flap positions.

E-2 1087 E-16174 (a) In flight. (b) Ground view.

Fig. 9. The B-52 aircraft and HL-I0 lifting body mated configuration.

Distance, m

Ground track, powered .------- Ground track, glide -.-.- Low-energy track - -- - High-energy track } Powered

Configuration change (transonic to subsonic)

Distance, n. mi. 0 2 4 6 8 10 12 14 16 I 1

Glide flight ground track

--. -.

runwa! Point above runway

I intersection I I h = 10,670 m I I

I (35,000 ft) I I

: j ;., \Powered flight ;! . . , a a . . -._, launch point, h = 13,716 m ground track -.. ':.,:: . ,.

I - ., . ,, ., - - 5 8 , .,, (45,000 ft)

5 10 15 20 25 3 0 ~ lo3 Distance, m

930386

6 Distance, n. mi.

4

Fig. 10. Typical HL-10 powered and glide flight ground track in the terminal approach and landing pattern.

Altitude above 609.6

ground, (2000) m (ft) I Flare initiation

Postflare deceleration :

I I I I

0 0.8045 (5) 1 .SO9 (1) Distance, km (mi)

930387

Fig. 11. Typical lifting body unpowered final approach, flare, and landing segments.

(a) Terminal approach and landing pattern. Fig. 12. Final approach, deceleration, and landing of the HL- 10 lifting body.

Fig. 13. The first HL- 10 glide flight landing flare.

E-16199

Fig. 14. Mr. Bruce Peterson, NASA project pilot, after the first HL-10 glide flight.

Pitch angular rate,

deglsec

2

Elevon position, deg from V V " V V V V " V y v v v I SAS servo

-2 0 2 4 6 8 10

Time, sec 930472

Fig. 15. The HL-10 first flight control system limit cycle time histories.

I,/ 2" nose right sideslip Right

Flow attaches Flow attaches

V

Left

0 5 10 15 20 25 30 35 40 45 50 55 Time, sec

930486

Fig. 16. Inboard tip fin flap strain gauge and angle-of-attack response.

Model centerline Section 1

Section 1

Sectlon 1

reference plane

Model centerllne Section 2

'Section 2

Horizontal reference plane

Section 2 Fig. 17. Proposed tip fin modifications I and 11.

E 18875 Fig. 18. Captain Jerauld "Jerry" Gentry, USAF HL-10 project pilot.

E 20492 Fig. 19. Messrs. Weneth D. Painter, Herbert Anderson. Jack L. Kolf, John Manke, and Joe Huxman with USAF fire truck and crew (left to right).

E-20777

Fig. 20. Major Peter C. Hoag, USAF, in a pressure suit.

E-20288

Fig. 2 1. Mr. William H. Dana, NASA HL- 10 project pilot.