SP-4305 ENGINEER IN CHARGE

 

9

The Transonic Problem

 

[249] Throughout World War II, Langley and the other NACA laboratories faced a perplexing research policy problem: the need to balance refinement of conventional aircraft technology with a strong enough dose of basic research into the high-speed frontier of future flight. In 1941, American combat planes 'encountered the dangers of compressibility for the first time. When diving vertically to terminal velocity, they penetrated far into the transonic region, where the effects of such aerodynamic phenomena as shock waves on moving bodies were largely unknown.* Lockheed test pilot Ralph Virden's fatal loss of control in November of that year during a dive test of the P-38 Lightning, an extremely high-powered and aerodynamically clean fighter plane, dramatized the need for a more complete understanding of the essential characteristics of transonic flight.1 The military necessity of solving the many difficult problems associated with the development of jet engines and guided missiles and rockets underscored this need. This combination of factors-the waging of world war and the emergence of compressibility and other problems of high-speed aerodynamics-challenged...

 

 


[
250]

Test of Lockheed P-38 Lightning in Full-Scale Tunnel, December 1944.

Test of Lockheed P-38 Lightning in Full-Scale Tunnel, December 1944.

 

....Langley not only to create a whole new wind tunnel and flight research capability but also to move away as quickly as possible from the first duty of its wartime assignment, which was essentially to help industry catch up with and consolidate the many diverse gains of pre-1939 NACA research.

In spite of their parent organization's heavy commitment to cleaning up specific military configurations, many of Langley's more experienced and restless aeronautical researchers somehow managed to find the time and resources during the war to respond to this challenge. In the summer of 1942, for example, at the same time that Eastman Jacobs was attempting to develop the ducted-fan jet propulsion system (see the previous chapter), he and Arthur Kantrowitz were also busy designing and constructing the NACA's first supersonic wind tunnel. This tunnel, which had a small 9 x 9-inch test section, was built with the approval of NACA headquarters to serve as a partial model for a larger supersonic tunnel authorized for construction at the new Ames laboratory in California. Though it had a water condensation problem, the Jacobs-Kantrowitz model supersonic tunnel provided Langley aerodynamicist's timely education and experience in the fundamental phenomena of supersonic flow. In 1944 and 1945 Langley engineers changed this pilot facility to closed circuit, dry-air operation. This conversion immediately preceded the NACA's major drive for large supersonic tunnels, which began in 1945.2

But the deepest probe into high-speed aerodynamics at Langley during the early part of the war was made by John Stack's wind tunnel groups. [251] In December 1941, a few weeks after test pilot Virden lost his life test-diving the P-38, Stack's 8-Foot High-Speed Tunnel (HST) group began an investigation of the stability and control problems of Lockheed's new airplane using one-sixth-scale models.3 At about 450 miles per hour, shock waves formed on the upper surface of the P-38's wings. This formation of disturbed airflow-which was not unique to the P-38-made it very difficult, and sometimes impossible, for a pilot to recover the plane from a steep dive. Either controls stiffened up so much from the resulting loss of both lift and downwash on the tail that he could not pull out, or, as had happened in Virden's fatal case, violent buffeting and strong downward pitching motion tore the plane's tail off.4

In March 1942, after less than four months of tests in Langley's 8-Foot HST, Stack's engineers reported that they had an answer to the P-38's dive-recovery problem: a wedge-shaped flap installed on the lower surface of the aircraft's wings. They said that their tunnel tests showed that wings having this flap would retain enough lift at high speeds to enable a pilot to pull the plane out of steep dives.5 Langley then turned the dive-recovery program over to its sister facility in California-Ames Aeronautical Laboratory at Moffett Field-where the flap idea could be proved sound to nearby Lockheed more expeditiously than at faraway Langley. Further tests in Ames's new 16-Foot HST did prove the idea sound: NACA-style dive-recovery flaps eventually saw service not only on the P-38 but also on the P-47 Thunderbolt, the A-26 Invader, the P-59 Airacomet (America's first jet), and the P-80, the first U.S. airplane designed (by Lockheed) from the beginning for turbojet propulsion.6

Stack's 8-Foot HST group carried out the dive-recovery program to prevent failures of precious combat aircraft in dangerous high-speed maneuvers. But there was another purpose as well: the Stack group did this work as part of a more basic program to develop a new family of high-speed airfoils. At the time that the P-38 was experiencing its most severe troubles, Stack and his closest associates were discussing the validity of calculations showing conventional airfoils to have improved lift and moment performance when operated inverted with negative camber-meaning with a curved-in or caved-in camber. (This was contrary to the conventional design, which added different degrees of outward curvature from the chord line.) However, all the Langley engineers involved in these conversations dismissed this as an unthinkable approach to solving the P-38's problems. Their educations and working experiences as aerodynamicists told them that positive camber was inherently beneficial to the lifting power of airfoils because such curvature caused the comparatively low pressures on the top side of the airfoil necessary for great lifting force. This lesson of conventional subsonic....

 


[
252]

Republic P-47 Thunderbolt with dive recovery flaps, 1946.

Republic P-47 Thunderbolt, 1946. The dive recovery flaps are barely visible underneath the wings.

 

....aerodynamics was so deeply ingrained in the engineers that they dismissed the new high-speed airfoil data as an impractical aerodynamic curiosity. One engineer remarked that pilots would reject the inverted wing because it would make them think that they had to fly their airplanes upside down.7 The imaginative critic was being facetious, of course, because he realized that pilots would be willing to fly the new "upside-down" wing right-side up-especially if the wing helped them to pull out of supercritical dives.

Beyond the sniping at pilot mentality, though, Langley's work on the P-38 embodied an early attempt to obtain an airfoil with truly supercritical performance. The broader goal of the research program was to discover the new airfoil shapes that would make propellers and wings capable of flight at speeds of 500 miles per hour and beyond. At such velocities, which seemed so fantastic at the time, a plane would be flying into the mysterious transonic region. Tests in the lab's 12-inch and 24-inch induction-jet high-speed tunnels had already shown that even on approaching such speeds, air "bunched up" unpredictably on the upper surface of an airfoil. This bunching-up or burble caused an airfoil to lose its lifting power and controllability.8

In July 1943, the NACA directed Langley to investigate routinely this and other major impediments to safe and efficient flight at high speeds by creating a new Compressibility. Research Division at the lab. This [253] division incorporated all the ground-based high-speed research sections then at Langley, including the 8-Foot HST section, the 16-Foot HST section, the model supersonic tunnel section, and a small section under Arthur Kantrowitz involved with the study of fundamental gas dynamics. Five months later, George Lewis formed a special four-man panel (consisting of Stack, H. Julian Allen of Ames, Abe Silverstein of the Aircraft Engine Research Laboratory in Cleveland, and, as chairman, Russell G. Robinson of NACA headquarters) to coordinate NACA high-speed research.

 

The "Sound Barrier"

 

In 1935, a newsman asked British aerodynamicist W. F. Hilton what problem he was working on in the National Physical Laboratory's newest high-speed wind tunnel. Pointing to an airfoil drag plot, Hilton replied, "See how the resistance of a wing shoots up like a barrier against higher speed as we approach the speed of sound." The next morning, all the leading English dailies misrepresented Hilton's response by coining the phrase, "the sound barrier." 9

The men who specialized in high-speed aerodynamics at Langley and elsewhere during this era knew that no actual physical barrier existed. But they did realize that flying at sonic velocity required not only finding a set of yet unimagined practical solutions to a number of tremendously adverse and perhaps ineradicable aerodynamic and power plant problems, but also overcoming some major inhibitions against assaulting what was commonly held to be an impenetrable barrier. Realization of these problems did not stop aeronautical engineers from trying to add a hundred miles per hour or more to the maximum practical speed of contemporary aircraft by refining design techniques, carefully streamlining aerodynamic shapes, and improving the power and efficiency of aircraft engines. It did prevent all but a few farsighted individuals, however, from considering flight at speeds approaching and going beyond that of sound.

Indeed Stack and Jacobs, his section head, had worked on high-speed aerodynamic problems from very early in their NACA careers. In 1927 they constructed an 11-inch induction-drive high-speed wind tunnel whose airstream was provided by a rapid blowdown from the VDT. Though certainly not meant to explore transonic flight, tests in the 11-inch tunnel provided the Langley engineers with important formative experience in highspeed aerodynamics. Compressibility phenomena and their ill effects on the performance of airfoils became a major new concern.

In 1933 Jacobs, who as an amateur astronomer was familiar with various optical systems, suggested that Stack try to make compressibility....

 


[
254]

Amelia Earhart visiting Langley, 128.

During her tour of Langley in November 1928, Amelia Earhart had part of her raccoon fur coat sucked into the 11-Inch High-Speed Tunnel. To her left are Henry Reid and Col. Jacob Wuest, Langley base commander.

 

....phenomena visible by using schlieren photography, a method first used by the Austrian scientist Ernst Mach (1838-1916) for visual observation of supersonic flow. Using a schematic drawing of the method from Robert W. Wood's Physical Optics, found in the Langley library, Stack constructed a crude schlieren device. Though the quality of the first photographs was poor, the results were nonetheless sensational:

 

Shock waves and attendant flow separations were seen for the first time Visitors from all over the Laboratory, from Engineer-in-Charge Reid on down, came to view the phenomena.**

 

In 1934 Langley equipped its new 24-Inch High-Speed Tunnel (built with $10,000 provided by the Public Works Administration) with an improved version of the schlieren photographic system.

 

 


[
255]

Schlieren photographic system in 9-Inch Supersonic Tunnel.

Schlieren photographic system in 9-Inch Supersonic Tunnel.

Wind tunnel researchers used schlieren photographic systems to visualize high-speed aerodynamic phenomena affecting airfoil performance. In the photograph, a general view of Langley's 9-Inch Supersonic Tunnel, showing the control desk, the test section, and, just to the right of center, the schlieren apparatus. (Schlieren systems continue to be used today.)

 

Jacobs used the data from the first tests in this new tunnel as the basis of a paper he was preparing for the Volta Congress on High-Speed Aeronautics (in Italy). In this important paper, Jacobs tried to describe what was actually happening in the compressibility burble (i.e., the region....

 


[
256]

Schlieren photographs showing aerodynamic phenomena at various speeds.

Schlieren photographs showing shock waves, wakes, and other aerodynamic phenomena affecting a low-speed and a high-speed airfoil at subsonic, transonic, and supersonic speeds.

 

....of disturbed flow generated behind a shock wave). He also attempted to determine whether the critical Mach number could be increased by changing the shape of the airfoil in minor ways. What Jacobs actually did in this paper was derive the relationship between the local Mach number on an airfoil at high speed and the suction pressure on the airfoil at low speed. This derivation enabled aerodynamicists to estimate the critical Mach number from the low-speed pressure distribution on the airfoil. They could then improve the high-speed performance by making shape changes determined on the basis of simple incompressible theory or low-speed tests.10

In 1933 and 1934 Stack conceptualized on his own initiative an experimental aircraft for compressibility research and published a drawing of it in the Journal of the Aeronautical Sciences. He concluded from his paper study that a small propeller-driven monoplane, powered by a 2300-horsepower Rolls-Royce engine and equipped with wings designed in accordance with the NACA's new high-speed airfoil sections, might fly to a maximum 566 miles per hour. Though the NACA apparently never considered helping Stack to find a developer for the airplane, the optimistic results of his paper study convinced many people at Langley that the potential for flying at speeds far in excess of 500 miles per hour was there. To realize that....

 


[
257]

John Stack

John Stack

.

.

John Stack as an MIT student in 1927, and (larger photograph) as head of Langley's Compressibility Research Division in 1944.


 

.....potential, however, the first thing Stack felt that aircraft designers needed to have from the NACA was a much more complete understanding of the basic aerodynamic phenomena in the transonic region.11

Leaders of the aerodynamics staff at Langley like John Stack and Eastman Jacobs understood the physics of fluid flow well enough to know that though "barrier" was the wrong word, there did exist a definite set of transonic flight problems. There were flight and propulsion ingredients to this problem-set, and there was a major wind tunnel ingredient. (In demonstrating how the resistance of a wing shot up "like a barrier" as it approached sonic speeds, had not Hilton pointed the English newspaper reporter to an airfoil drag plot drawn from a series of tests from a wind tunnel?) A limbo in the state of tunnel technology existed just below and just above the speed of sound, preventing fruitful research until there were practical solutions to difficult questions: Why did strange things happen in Langley's own tunnels as airflows approached Mach 1, the velocity of sound? Why could one get Mach 1 in an empty high-speed tunnel but not [258] in one with a model installed in the test section? Why did the airflow in the 8-Foot HST always tend to choke up in the tunnel's throat at some Mach number, generally above 0.7, that was lower the larger the model size and its blockage effect? Why did this still happen no matter how fast NACA mechanics made the driving fans turn? Why did shock waves form off the test model, reflect off the tunnel wall, and thereby inhibit accurate measurement of flow characteristics and behavior around the model? Since the model supersonic tunnel was capable of producing airstreams speeds in excess of Mach 2, why did it also choke when slowed down to produce transonic airspeeds?

Though these questions plagued aerodynamicists in the late 1930s, physicists had actually known the fundamentals of the choking problem and the identity of what was now called the "sound barrier" long before that time; in fact they had known them even before history's first wind tunnel had been built in the 1870s. In the 1830s, French scientists Wantzel and Saint-Venant had derived mathematically that a gas flowing through the narrowest part of a constricted duct could not exceed sonic velocity no matter how much additional driving pressure was exerted. To later scientists, this finding did not mean that supersonic flow was unattainable in channels; it meant only that the channel area had to be expanded or diverged to accommodate the increased volume required by the flow as it accelerated above Mach 1 downstream of the throat. In the late 1880s, Swedish physicist-mathematician De Laval applied this principle to achieve supersonic velocities in the convergent-divergent nozzles of his steam turbines.12

In the 1920s, Americans L. J. Briggs and Hugh L. Dryden obtained supersonic flow during experiments at the Edgewood (New Jersey) Arsenal with a small free-jet apparatus having a convergent-divergent nozzle. Results indicated that radical changes in the behavior of air happened as the speed of its stream approached that of sound.13 Later in the decade, Eastman Jacobs and John Stack found during test runs of Langley's original high-speed induction tunnel that higher Mach numbers could be reached with an open throat than with a closed throat. Throughout the 1930s Stack and colleagues explored the nature of the tunnel-choking problem. Among other things, they made a detailed study of the blockage effect caused by the presence of a model in the test section. Until aircraft reached much higher-flying speeds, the choking problem did not demand a practical solution, however; at low airspeeds, the choking effect was small and accurately correctable.

In the spring of 1940 William J. Orlin, an independent-minded engineer in Stack's 8-Foot HST section, developed a small water channel to visualize [259] the process of the tunnel choking problem by hydraulic analogy. Though this little facility provided some valuable insight into the dynamics of a choking fluid, its operation suggested no solution to the problem of choking.14 By December 1941 it was clear to Stack and his wind tunnel engineers that no one was likely to solve this problem for some time if ever. No one was making any significant progress in the theory of transonic flows. Thus they envisioned only one alternative to boarding up the now vitally important transonic region and closing it off from study: a specially instrumented full-scale transonic research airplane.

 

Stack Gets Go-Ahead

 

Stack sold Langley management on the technical merits of the transonic research airplane idea in the spring of 1942. Then, though he and everyone else at the lab knew very well that the NACA charter did not allow construction of a complete airplane-and that the NACA budget could not finance such an enterprise even if it did-he brought the idea before George Lewis, director of NACA research. Lewis did not care for the research airplane idea on principle, but characteristically, he tried not to react too negatively. He liked Stack: not only was Stack a rugged individualist, a "man's man," with self-assured ways and ambitious ideas, but he had also proved over the last 15 years to be one of the NACA's best researchers. One question Lewis asked Stack playfully, before talking with him more seriously about the present strain of military research and the inevitable problem of getting such a project off the ground, was whether people might interpret the NACA's unprecedented desire for a research airplane as an admission of some basic failure on the part of all those expensive wind tunnels and their champions at Langley. He left Stack with the idea, however, that some low-priority, back-of-the-envelope estimates to identify the most desirable design features of a transonic airplane could not hurt anyone, providing they did not distract from more pressing business.

Though he knew that this go-ahead by no means implied Lewis's general approval to develop and procure an airplane, Stack immediately assembled a special team of NACA researchers to work out the design requirements of a transonic research aircraft. Stack selected engineer Milton Davidson and junior engineering aide Harold Turner, Jr., to make preliminary layouts and performance estimates. By the early summer of 1943 Davidson and Turner finished a preliminary design of a small turbojet-powered plane capable of flying safely to high subsonic speeds, from Mach 0.8 to 1.0. 15 had also helped Eastman Jacobs design his proposed research airplane, the one described in the previous chapter, but this new design was far different.) [260] Stack wasted no time in sending news of his pet project to friends and selected acquaintances inside the aeronautical branches of the army and navy.

Military personnel who learned about Stack's transonic research airplane idea had solid reasons to be cautiously interested in it. Ezra Kotcher, the chief of aeronautical research of the Air Service Materiel Command at Wright Field, felt, too, that the speed of sound was only a wind tunnel and mental barrier. In 1939, Kotcher (a 1928 graduate in mechanical engineering from the University of California) had himself suggested the construction of a transonic research airplane to be powered by either a gas turbine or rocket propulsion system.16 Because it came before news of the British and German successes with turbojets, General Arnold had rejected Kotcher's suggestion. Several things had happened since 1939, however, to change Arnold's attitude: the army had found out about the Whittle engine; at Arnold's instigation, secret development of an American turbojet had begun; the Guggenheim Aeronautical Laboratory at the California Institute of Technology (GALCIT) had completed a series of experiments with JATO (jet-assisted takeoff) rockets for army ordnance;17 and, perhaps most importantly, two of the army's newest high-speed combat aircraft-the Republic P-47 Thunderbolt and the Bell P-39 Airacobra- had experienced fatal tail failures during high-speed dives. This frightening epidemic was hitting navy airplanes just as hard: for example, the horizontal tail of its Curtiss SB2C Helldiver was fluttering so badly in high-speed maneuvers that the tail was breaking off.18 Together, these developments were making answers to compressibility problems an urgent necessity.

Although the loss of life and aircraft heightened both army and navy interest in a possible NACA high-speed research airplane, few officers in either service responded to Stack's proposal immediately. Kotcher, for one, was too busy coping with buzz bomb problems (the army had just assigned him the formidable job of copying the German pulsejet-powered V-1 robot missile). One military man who did actively advocate support for Langley's concept was Capt. Walter Diehl, USN. A longtime friend and close working associate of the NACA, Diehl argued in late 1943 during meetings with the chief of BuAer's structures branch that a transonic research airplane was the only way to convince people that the "sound barrier" was "just a steep hill." 19

The first time that the military actually recognized and formally discussed Stack's idea, however, was at a conference between service and NACA personnel held at Langley laboratory on 15 March 1944. On that day, two meetings-one chaired by Captain Diehl and the other by Col. Carl Greene, the permanent liaison officer at Langley from the Materiel [261] Command-dealt with the development of a possible transonic research aircraft. During these meetings the NACA put its weight behind Stack's proposal of a joint NACA-military program to develop and construct a transonic research airplane. According to NACA spokesmen, the purpose of this airplane would be to collect the aerodynamic data needed for transonic flight that could not be obtained in any wind tunnel.20

Langley's visitors could not reach agreement with NACA representatives that day, however, over the goals of a transonic research airplane, let alone over its basic design features. Army spokesmen conceived of the airplane more "as a major developmental step toward higher operating speeds extending upward through Mach 1," and navy representatives were inclined to view it "as a means of dispelling the myth of an impenetrable barrier and providing needed high-speed data." 21 Some of the differences of opinion were quite outspoken. For example, Maj. Gen. Oliver P. Echols, assistant chief of staff at Air Corps headquarters, questioned the wisdom of procuring in wartime a nonmilitary research airplane, especially if the army was going to pay for most of it. By the end of the meeting, two things were frustratingly clear to Stack and his colleagues at the NACA: the competitiveness of the services was going to make it very difficult to win joint army-navy cooperation in the development of a single experimental vehicle; and the NACA had better find some other reliable method to collect transonic data.

 

Stopgap Methods

 

Although the Langley staff slowly came to believe that Stack's transonic airplane should perform the ultimate experiments, it supposed at first that an area as mysterious as the transonic speed region had to be attacked from every possible theoretical and experimental approach. After attempting some theoretical studies, however, the staff chose to treat the problem of transonic flows as essentially an experimental problem. In its opinion, transonic flows involved too many unknown physical principles and complex mathematical relations to recommend the theoretical approach.

In 1942 John W. "Gus" Crowley and Floyd L. Thompson of the lab's flight research section suggested that the NACA improve understanding of the aerodynamic phenomena occurring at high speed by mounting a wing model on a bomb-shaped missile and dropping this body from an airplane flying at great altitude. (The Germans had already tried this technique in 1941, but it is unclear whether Crowley and Thompson knew about this previous experience.) After some preliminary tests, Langley temporarily abandoned this drop-body technique. Not only did the lab lack adequate radar and radio telemetering equipment to measure the high-speed flow [262] around the test model accurately, but it was also just too much of a chore to build the big test model (made from a piece of pipe 12 feet long and 10 inches in diameter, filled partly with concrete), carry it up to 40,000 feet, and then find and salvage the model after it had sunk several feet into the muddy bombing range at Plum Tree Island near Langley Field.22

In December 1943 the NACA revived the falling-body idea in response to a proposal for a joint American-British effort on the transonic research problem made by William S. Farren, director of the British Royal Aircraft Establishment. During his visit to Washington that month to give the annual Wright Brothers Lecture, Farren described how the RAE, too, had experimented with dropping weighted bodies from high altitudes. He indicated that such investigations could now be carried out very effectively thanks to tremendous advances in the development of radar and other electronic telemetering devices.23 Farren's message convinced the NACA Executive Committee that this type of research should be tried again at Langley. On 25 March 1944-ten days after the conference dealing with Stack's transonic airplane concept-the committee approved research authorization 1224, a confidential "Investigation of Aerodynamic Characteristics of Free Bodies at High Mach Numbers."

Langley began new falling-body tests with a Boeing B-29 Superfortress borrowed from the army, which it equipped with the navy's most advanced SCR-584 radar-tracking unit. The large airplane carried the test missile to 30,000 feet and then released it. Ground observers received radio signals relayed from instruments inside the body. These instruments, which were developed just for this test program by Edmund C. Buckley of Langley's Instrument Research Division, measured the forces on the body as it dropped at velocities sometimes equaling or exceeding that of sound.24 Though the speed range and therefore the data were limited by the maximum operational altitude of the B-29, and though Theodore von Karman of Caltech raised the possibility of errors due to "acceleration effects," NACA engineers considered these data reliable enough to estimate the drag and power requirements of a transonic airplane.25 (By dropping identical models of varying weight, the NACA later proved that the acceleration, or virtual mass, effects were negligible.)

In 1944, an NACA engineer devised a second alternative method of transonic research whose value was perhaps even greater. Robert R. Gilruth, the young engineer from the University of Minnesota in charge of Langley's flight research section, had noted as early as 1940, during dive tests of the Brewster XF2A-2 airplane, that the transonic flow fields developing on wings in actual flight were 10 to 20 times larger than those predicted by wind tunnel tests on models. More recently, he had heard pilots who had put....

 


[
263]

Drop-body model with instrumentation for transonic drag investigation, 1948.

Drop-body model with instrumentation for transonic drag investigation, 1948.

The high speeds of diving aircraft and missiles during World War II took the NACA by surprise, for there were no wind tunnels in its laboratories capable of exploring transonic aerodynamics. One of the stopgap methods developed by Langley after 1944 to acquire meaningful data near Mach 1 involved dropping instrumented bodies (top) for free falls from high altitude. The first aircraft used by the NACA to drop its instrumented bodies was a Boeing B-29 Super fortress (bottom), the type of aircraft that dropped the first atomic bomb. In the early 1950s, the NACA used a North American F-82 Double Mustang (just below) for its drop-body tests. A test body is shown mounted beneath the Double Mustang's center wing section.

Aircraft used for drop-body tests, the North American F-82 Double Mustang, early 1950s.

Aircraft used for drop-body tests, the Boeing B-29 Superfortress, 1945


 


[
264]

From the Langley Air Scoop, 5 October 1946.

From the Langley Air Scoop, 5 October 1946.

 

....the new North American P-51 Mustang into dives report that they could actually see, if the sunlight was just right, the shadowy edges of shock waves cutting across the streamlines of their airplane's wings. Though this sleek little fighter plane was diving at speeds only to about Mach 0.75, naked-eye observation of this phenomenon by pilots suggested to Gilruth that the flow of air over a portion of the P-51's wing was moving quite smoothly through the speed of sound to low supersonic speeds! Although admitting that a number of Langley's wind tunnel researchers were far more expert than he was in the physics of airflow, the head of the flight research section believed that the supersonic flow region that existed on wings at supercritical (but still subsonic) flight speeds could be used as a test environment. This "flying wind tunnel" would have one great advantage over tunnels on the ground: it would not have walls to constrict and distort the airstream.26

In his first application of the wing-flow technique, Gilruth mounted a small airfoil model perpendicular to the upper surface of a P-51D's wing. He placed the model vertically just above the Mustang's wing, making sure, in order to generate uniform flow for valid testing, that it rested in that part of the supersonic flow region where the induced velocity was most constant. An NACA test pilot then flew the Mustang to the desired....

 


[
265]

Robert R. Gilruth

Melvin N. Gough (right, 1956), Langley's chief test pilot

After extensive flight experience as an engineering observer in the late 1930s, Robert R. Gilruth (left) had a keen appreciation for the pilot's side of the man-machine relationship. In particular, he had learned a great deal from Melvin N. Gough (right, 1956), Langley's chief test pilot, who took great pains to educate Gilruth about airplane handling characteristics.

 

...altitude and dived the plane as rapidly as he safely could, which was to about Mach 0.81. As the speed of the plane in its dive increased, airflow around the small wing-mounted model passed from subsonic through the transonic region to supersonic velocities on the order of Mach 1.4. A small balance mechanism fitted within the P-51's gun compartment and tiny instruments built into the mount of the model recorded the resulting forces and airflow angles. (Because diving the airplane to high speeds was a dangerous maneuver, Langley developed an advanced system of rapid-response instrument surveillance to indicate whether the pilot was headed for trouble in handling the airplane's controls. Specialists in stability and control cleared each test flight that was to go to a previously unexplored high speed, and specialists in reading and interpreting the instruments indicating control handling qualities constantly advised the pilot whether he should proceed with the test once it had started.)27

The first reaction of the majority of Langley's, wind tunnel groups to Gilruth's wing-flow technique was negative: "There was great consternation amongst the wind-tunnel people in why a young upstart could come along [with a solution] when all their wind tunnels had" failed.28 Some of Gilruth's best friends were completely against using the method. Engineers in the....

 


[
266]

Wing-flow method of transonic research using P-51 Mustang, 1948.

Gilruth based his small-model wing-flow technique on the physical fact that air above the wing of a high-speed airplane, like the P-51 Mustang, went quite smoothly and uniformly through the speed of sound.

 

....16-Foot HST group, for example, pointed skeptically to its "many obvious problems" and "impurities"-the nonuniformities of the flow field, the unknown effects of the wing boundary layer, the problem of wing shock passage over the test model, and the very low test Reynolds numbers.29

Gilruth agreed with but discounted the technical content of most of these criticisms by confronting the method's critics with two rhetorical questions: Was not the collection of any usable transonic data preferable to the collection of none? Had technical problems and impurities comparable to those handicapping the wing-flow method ever stopped wind tunnel researchers from experimenting with new ways of doing things? Of course not, Gilruth knew. The young engineer tried to reassure the wind tunnel specialists by telling them that in applying the wing-flow method his flight researchers were not using just any airplane wing. They were modifying the wing surface experimentally to meet all the aerodynamic conditions essential for valid testing.***

[267] Gilruth's first wing-flow test results-which the NACA kept secret-impressed Langley's wind tunnel groups.30 Besides demonstrating airflow trends that conformed to expectations, they gave the most systematic and continuous plots of transonic data assembled by the NACA up to that time. This success not only prompted Langley to put an entire series of wings of various thicknesses through the wing- examination, but also eventually helped to confirm its opinion that supersonic flight required a thin wing.

About one year after starting to use the drop-body and wing-flow techniques extensively, NACA Langley began to develop a third stopgap method of acquiring transonic data: rocket-model testing. Conducting research on "pilotless aircraft" (the military's name for all types of guided missiles) was not at that time new to the laboratory, however. From the Executive Committee's authorization of a test of the General Motors "flying bomb" in the Full-Scale Tunnel in June 1941 to the time it approved rocket-model testing in early 1945, Langley had worked, in one way or another, on practically every guided missile project started by either service, including the testing and development of glide, shrouded, and buzz bombs, gliding torpedoes, and various types of interceptor missiles.31 In December 1944, acting engineering-charge John Crowley organized a "Special Flying Weapons Team" to oversee all missile research at Langley.32 (Henry Reid was at the time in France with the Alsos Mission, a secret group sent by the secretary of war to the European theater to identify and collect valuable scientific research information abandoned by the retreating German army.)

Along with its support of the military's top-secret guided missile projects, Langley began an ingenious program of more basic aerodynamic tests. From the remote Atlantic coast beaches at Wallops Island, some distance from Hampton off the Eastern Shore of Virginia, a small team of researchers launched rocket models weighing about 40 pounds, of which about 50 percent was the weight of the rocket motor and about 20 percent the rocket fuel. These rocket models shot up into the air to a maximum velocity (at an altitude of only 2000 feet) of about Mach 1.4, continued upward, and then splashed into the Atlantic Ocean. The useful portion of the rocket's flight terminated at about 15,000 feet, meaning that the data were obtained in relatively dense air where the Reynolds number was high. Originally the researchers tracked the models and determined...

 


[
268]

Rocket-model launching site at Wallops Island, Virginia, 1945.

The NACA 's first year of operations at remote Wallops Island, Virginia, resembled the activities of advanced military bases in the Pacific during World War II.

 

...their speed, drag, and control effectiveness by using the immediately available telemetering instrumentation previously developed for the falling-body tests; later, they adapted a Doppler radar system which, for certain research purposes, made it unnecessary to place complex instruments inside the test body. In the spring of 1945, Congress approved the NACA's request for a supplemental authorization for a permanent rocket launch facility at Wallops. The purpose the NACA had in mind for this facility was not only to support the military's ballistics projects but also to help define the basic airplane wing and fuselage configuration best able to fly in and through the transonic range.33

The rocket-model test was a challenging technique for NACA flight researchers to perfect. It required them to acquire and apply new knowledge about how to measure, transmit, and record accurate test data during the few fleeting seconds of a flight, which changed speed, altitude, and model attitude rapidly. By comparison, the aerodynamic goals of the initial test flights were relatively simple, such as to trace the minimum drag curve throughout the transonic region for a variety of representative test objects. Beyond the exploratory flight-testing for which the rocket models were suited originally, however, the new technique made it possible in some later cases to employ systematic parameter variation. Beginning in the summer [269] of 1945, a succession of rocket models were launched at Wallops, each model identical to the next except for one geometrical feature. Though wind tunnel groups at Langley knew that the rocket-model technique was not well suited for advanced aerodynamical research involving extensive pressure distributions, flow surveys, boundary layer measurements, and flow visualization, they did not criticize the rocket model technique-as they had the wing flow technique-on scientific and technical grounds. The wind tunnel groups realized that the new flight technique was largely free of the impurities of the other stopgap transonic techniques and thus constituted exactly what was needed by the NACA at the time. They credited Robert Gilruth and his principal assistants for coping energetically and ingeniously with the inherent problems of the technique, and credited Edmund C. Buckley and his group for devising the indispensable tiny flight instruments.

But wind tunnel groups did eventually criticize rocket-model testing for interfering with tunnel programs. Each firing required the sacrifice of a precious test model, many of them having expensive instruments inside. Though Langley employed its own shop staffs to build these models and incorporate the instruments, wind tunnel proponents complained, especially after the June 1946 conversion of the "Auxiliary Flight Research Station" at Wallops from a subordinate unit of Langley's flight research department into a separate "Pilot less Aircraft Research Division" (PARD), that the "voracious appetite" of the rocket-model specialists was resulting in "a major slowdown" in the production of their own....

 

 


A rocket model is fired from Wallops in 1949

A rocket model is fired from Wallops in 1949.

 

 

[270] .....necessary test models and instruments. In the years 1947, 1948, and 1949, PARD expended no fewer than 386 models. Wind tunnel personnel told themselves that this expenditure was "roughly equivalent to the requirements of perhaps 10 major wind tunnels." Privately, they also said that the practitioners of the rocket-model testing technique tended, partly out of necessity, to be "as much interested in making the rocket models to do more things accurately as they were in the research problems." This tendency was apparent to them in that a majority of PARD reports discussed and analyzed the performance of specific model configurations without shedding much new light on the underlying flow processes-which were, after all, the main object of rocket-model studies in the first place.34

Though they could not totally dispute the charges, PARD employees objected strongly to implied criticism of the value of rocket-model testing in comparison with the value of wind tunnel testing. Though rocket-model testing appeared expensive because of the loss of complex and costly models to the depths of the Atlantic Ocean, a single test provided enough important data, they said, to establish the key flight parameters. Thus, the dollar-for dollar return on the NACA's investment in rocket-model research at Wallops at least matched that provided by the wind tunnels.35

However stridently the individual research groups may have debated the scientific, technical, and budgetary validity of the new drop-body, wing-flow, and rocket-model techniques, the internal debate never overshadowed the commitment of Langley's research staff as a whole to exploring every avenue of transonic research. As NACA engineers and scientists, they knew that there existed, between the study of fundamental fluid mechanics and the large-scale testing of specific ideas, a range of problems for which either wind tunnel or free-air methods of research were most suitable. They knew also that every particular method had advantages and disadvantages. Thus they concluded that the peculiarities of the individual problem should dictate the choice of method. History bears out the truth of this conclusion: the early years of the rocket-model program at Wallops (1945-1951) showed that Langley was able to tackle an enormously difficult new field of research with innovation and imagination.

Back in March 1944, before these alternative, free-flight methods of transonic research had been established, no one understood the need for flexibility in research method better than John Stack. If the frustrating interservice rivalries and differences of opinion that surfaced at the Langley conference that month made it seem impossible for the army and navy to cooperate in a high-speed research airplane program, then the NACA should try a new approach: it should try to persuade one of the services, or each of them individually, to procure its own transonic research airplane.

 


* The transonic range begins for a particular aircraft when the flow over any part of the aircraft's wing exceeds the speed of sound, or Mach 1. At the point when this speed is reached, at what is known as the critical Mach number, there are no adverse aerodynamic forces. But as the critical Mach number is exceeded, a shock wave, or major pressure disturbance, forms on the top of the wing (at 90 degrees to the airstreams) and propagates through the surrounding air at sonic speed. Because it forces the wing to encounter a mixture of subsonic and supersonic conditions, this shock presents serious problems for the aircraft (see further in text).

Another term that will appear in this chapter is supercritical. This refers simply, as it did in the 1940s, to any speed beyond the critical Mach number. In the 1960s, Richard T. Whitcomb of NASA Langley introduced a more restrictive meaning of the term with his invention of a "supercritical" airfoil to delay the drag rise that accompanies transonic airflows. Thus supercritical has also come to mean airfoil operations in the speed region between the critical Mach number and drag-rise Mach number. For an introduction to the principles of high-speed flight, see Van Deventer, An Introduction to General Aeronautics, pp. 108-128.

** John V. Becker, The High-Speed Frontier, p. 16. According to Becker, Theodore Theodorsen, Langley's ranking theoretician, viewed the schlieren photos skeptically; proclaiming that what appeared to be shock waves was really an optical illusion. At a banquet of the LMAL staff in 1936, Stack played the role of Theodorsen in a skit making his proclamation of the illusion.

*** There were four surface conditions necessary for the wing-flow method to be valid: (1) the chord-wise velocity gradient had to be sufficiently small; (2) the velocity gradient normal to the airplane wing had to be sufficiently small for a distance somewhat exceeding the span of the test model; (3) the boundary layer on the airplane wing had to be sufficiently thin so as not to affect the flow at the juncture of the model and the main wing; and (4) the normal shock that occurs on the main wing had to be sufficiently far back so that the pressure rise through the shock acting back through the boundary layer could not affect the flow at this same juncture. See Robert R. Gilruth, "Résumé and Analysis of NACA Wing Flow Tests," unpublished paper presented at the Anglo-American Aeronautical Conference, Sept. 1947, copy in Langley Central Files (LCF), A184-9, "High-Speed Research."


previous pageindexnext page