SP-3300 Flight Research at Ames, 1940-1997

 

Aerodynamics Research

 

[11] Early flight research that focused on aerodynamic issues was concerned with understanding drag, air loads, and compressibility phenomena that influenced both the performance and control of the aircraft at high speed. This work was motivated by problems uncovered in the design of these high-performance aircraft and in early operational experience with them. Furthermore, as Ames engineers and pilots gathered information on their own, additional ideas surfaced that suggested new approaches to solving these problems. 4 Many of the World War II and postwar aircraft involved are noted in table 3 and shown in figures 15-28. Members of the flight research section, as they appeared in 1946, are shown in figure 29. The new flight research hangar, under construction and soon to be occupied, can be seen in figure 30.

It was of particular interest to aerodynamicists that data be acquired from an aircraft in flight to use in validating wind tunnel measurements of drag. The first aircraft used in this effort was the P-51B Mustang (one of the first production aircraft to have laminar flow airfoils). To carry out this experiment without interference from the propeller slipstream, the propeller of the aircraft was removed and its oil and coolant ducts blocked so that it resembled the wind tunnel model. The aircraft, flown by James (Jim) Nissen, was towed aloft by a P-61 and released. Careful measurements of longitudinal deceleration were used to determine aircraft drag, and the pilot of the Mustang made a powerless, gliding landing. During one of these flights, the tow cables accidentally separated from the P-61 and wrapped themselves around the P-51, interfering with the pilot's control of the aircraft. Despite a crash landing, Nissen...

 

TABLE 3. AIRCRAFT USED FOR AERODYNAMICS RESEARCH

.

Aircraft Name

Arrival or First Flight Date

Departure Date

.

P-38F (AAF41-7632)

December 30, 1942

July 16, 1943

P-51B (AAF43-12111)

August 11, 1943

September 7, 1947

P-39N

September 13, 1943

August 29, 1944

P-63A (AAF42-68892)

February 17, 1944

June 18, 1946

XSB2D-1 (Bu. No. 03552)

June 12, 1944

January 10, 1946

YP-80A (AAF44-83023)

September 19, 1944

January 27,1947

P-51B (AAF43-12094)

November 16, 1944

September 9, 1947

P-63A-6 (AAF42-68941)

January 27, 1945

June 18, 1946

P-47D-25 (AAF42-26408)

April 27, 1945

September 7, 1947

P-80A-1 (AAF44-85299 NACA 131)

December 18, 1946

June 6, 1955

YP-84A-5 (AAF45-59488)

December 2, 1947

October 5, 1948

YF-84A (AAF)

February 14, 1949

December 20, 1950

F-86A (AAF48-291 NACA 116)

August 29, 1949

January 11, 1960

YF-93A (AF48-317 NACA 139, 48-318 NACA 151)

February 5, 1951, June 5, 1951

1953

YF-86D (AF 50-577 NACA 149)

June 26, 1952

February 15, 1960

F4D-1 (Bu. No. 134759)

April 4, 1956

October 16, 1959

F5D-1 (Bu. No. 139208a NASA 212)

August 20, 1957

January 16, 1961

Lear 23 (cn23-049 NASA 701)

September 17, 1965

January 11, 1980

 

Figure 15. Lockheed P-38F Lightning.

Figure 15. Lockheed P-38F Lightning.

Figure 16. P-51B.

Figure 16. P-51B.

Figure 17. P-51B towed by P-61A Black Widow

Figure 17. P-51B towed by P-61A Black Widow.

 

[12] ....survived without major injury. Comparison of flight results and data from the 16-foot wind tunnel are presented in reference 9; they show good agreement below the drag-rise Mach number. In flight, the drag rise occurred at lower Mach numbers than it did in the wind tunnel; it was speculated that aeroelastic deformation under flight loads was the cause. The NACA technical report is prefaced with an editorial note by the chairman of the NACA commending Nissen's "great skill and courage" in staying with the airplane to keep this crucial data from being lost. Nissen eventually left Ames, initially for North American, and then on to develop what was to become the San Jose airport. He ultimately managed the airport for the City of San Jose. In his later years he owned and flew a Curtiss Jenny and a Thomas-Morse Scout from the airfield at his home near Livermore, California. 5

During and after World War II, a number of military aircraft were used in general investigations of high-speed flight phenomena. Because of structural failures in the tails of several fighter airplanes during high-speed dives, Ames undertook the measurement of tail loads on the Bell P-39N under a range of flight and maneuver conditions. Larry Clousing conducted most of those tests and demonstrated considerable courage in pushing the aircraft to its limits to obtain measurements of critical loads. Results of that work, examples of which are presented in references 10-12, pointed to deficiencies in the prediction methods, which not only underestimated loads but in some cases showed them to be in the opposite direction. Buffet loads were a contributing factor to tail loads and may have led to partial failure of the tail on one flight. Photographs from the tests showed that the fabric covering on the elevator bulged out at higher airspeeds, a condition that was followed by a partial structural failure of the horizontal stabilizer during the pullout from a dive. Analyses of the data established the effects of critical Mach number on the wing center of pressure and showed that the lift curve behaved as predicted. Other tests had shown the contribution of shock-induced wing stall to tail angle of attack and that influence, in turn, on a strong nose-down pitching moment. Distortion of the elevator fabric also served to move the stick free neutral point aft, thereby increasing the stick force gradient (ref. 13). Analytical predictions of vertical tail loads during rolling pullout maneuvers compared favorably with flight measurements for the airplane based on dynamic analysis of sideslip excursions during the maneuver (ref. 14).

Drag measurements were also performed on the P-39N-1, in which minimum drag and drag-rise Mach numbers were documented in tests carried out to Mach 0.8 (ref.15). Engine thrust was estimated using propeller efficiency and engine horsepower predictions in order to extract drag from the data.

High-speed buffet was evaluated in dive tests with the P-51 by George Cooper. During these tests, Cooper observed sunlight refracting through the shock wave, identifying its presence on the wing and noting a correlation between its movement and the occurrence of buffeting. Cooper co-authored a report with George Rathert on the visual observation of these shock waves (ref. 16). At one time, one of the ....

 

Figure 18. Bell P-39N Airacobra.

Figure 18. Bell P-39N Airacobra.

Figure 19. Douglas XSB2D-1 crashed in prune orchard in Sunnyvale, Calif.

Figure 19. Douglas XSB2D-1 crashed in prune orchard in Sunnyvale, Calif.

 

Figure 20. Lockheed YP-80A Shooting Star.

Figure 20. Lockheed YP-80A Shooting Star.

 

[13] ....P-51H aircraft was fitted with a Schlieren system for visualization of shock waves on the wing (ref. 17).

Strong nose-down pitching moments occurring at high speed limited pitch control for these high-performance aircraft. Wind tunnel tests indicated that slight upward deflection of the flaps could reduce those moments and expand the controllable flight envelope. At the suggestion of John Spreiter and Jim Nissen, flaps on the P-51H and F8F-1 were modified and a program was conducted on the two aircraft to substantiate the tunnel results. Maurie White and Melvin (Mel) Sadoff ran the tests, and George Cooper and Larry Clousing carried out the flights after Nissen had left Ames. The degree of elevator required to trim was reduced substantially for the P-51H up to Mach number 0.795, the highest speed tested. However, results from the F8F-1 were not encouraging; the favorable contribution realized on the P-51H through reduction of changes in tail angle of attack was offset by an increased nose-down pitching moment contribution from the wing (ref. 18).

Different propellers were tested on the XSB2D-1 during maximum power ground runs and in-flight performance evaluations. The aircraft was also flown so flight data and data obtained in tests run in the 40- by 80-foot wind tunnel could be compared. Welko Gasich reported the results of these tests and used them to compare alternative methods for calculating takeoff ground run (ref. 19). During what turned out to be the final flight of the XSB2D-1, an engine fire occurred, and George Cooper made a successful emergency landing, short of the main runway at Moffett Field and between tree rows in a Sunnyvale prune orchard. In the process, 84 trees were mowed down and the airplane's wings were severely cropped, but Cooper and Gasich, who flew on board as the test engineer, escaped without injury. The local farmer, who was personally acquainted with Cooper, was astonished to see George climb from the cockpit. Cooper, unflappable as usual, exclaimed to his friend, "You keep asking me to drop in on you sometime, so here I am." 6

Cooper was also involved in flight tests of a reversible-pitch propeller on the P-47D Thunderbolt. These tests were performed to evaluate the handling characteristics of the aircraft with the propeller used as a brake during a dive. At the conclusion of one test, the propeller failed to return to its normal pitch setting. Cooper was almost forced to land with the aircraft in this condition, but fortunately the propeller snapped out of its reversed pitch setting while the aircraft was on approach. He added power, climbed away, and made a safe landing on the next approach. 7

With the coming of the jet aircraft, compressible flow phenomena raised issues with aircraft performance and handling, leading to new demands for flight testing. The YP-80A was the first jet aircraft at Ames. The aileron buzz problem on the P-80 was of particular concern to its designers and became the focus of an intensive investigation in the 16-foot transonic tunnel as well as in the P-80A-1. 8 The Ames team consisted of Harvey Brown, George Rathert of the engineering staff, and Larry ....

 

Figure 21. Bell P-63A-6 Kingcobra.

Figure 21. Bell P-63A-6 Kingcobra.

 

Figure 22. P-80A-1 showing vortex generator installation.

Figure 22. P-80A-1 showing vortex generator installation.

 

Figure 23. Republic YP-84A-5.

Figure 23. Republic YP-84A-5.

 

[14] ....Clousing, the principal test pilot. Clousing performed dive tests to the highest speeds yet achieved with the aircraft in pursuit of data to identify the source of the problem. In so doing, the aircraft expanded the transonic flight envelope to 0.866 Mach. Results reported in references 20 and 21 showed the effect of critical Mach number on the aileron oscillation and tied it to shock-induced separation on the upper surface that influenced aileron hinge moments. The data from the 16-foot wind tunnel gave a good indication of the onset conditions. The phenomena had been observed earlier in P-39N dives and the correlation with Mach number was noted at that time; the oscillating shock was identified as the cause (ref. 22). Flight tests of these aircraft also investigated boundary-layer characteristics and removal, which is important for maintaining the proper airflow to fuselage-mounted jet engines. The YP-80A was also used for tail pipe temperature measurements. Clousing's contribution to these tests, as well as to the earlier testing of the P-39, was due to his skill and courage as a test pilot and to his interpretation of the results and the test techniques involved. For the benefit of his fellow pilots and engineers, he published a report (ref. 23) that reflected on the hazards of high-speed testing with these aircraft.

John Spreiter led a series of flight tests using several aircraft to determine the effect of Mach number and Reynolds number on maximum lift for comparison with results from the 16-foot tunnel. These tests were made on the YP-80A and on five propeller-driven fighters, the P-38F, P-39N, F6F-3, P-51B, and the P-63A. The data correlation was encouraging except when buffeting limited the angle of attack that could be achieved in flight (ref. 24).

George Cooper and Rudy Van Dyke began flight tests of the Air Force's new F-86A Sabre in 1949. They made prolonged dives, starting from 46,000 feet, in which the F-86A reached very high speeds. These flights opened up the aircraft's supersonic envelope and preceded North American and Air Force tests of the aircraft at these speeds. At about the same time, people in the general area began to hear explosions that occurred without any apparent reason. Eventually, these "explosions" were correlated with the dive tests of the F-86 Sabre; they occurred when the aircraft reached supersonic speeds. This was the first time the "sonic boom" phenomenon had been associated with the supersonic flight of an aircraft. 9 It is also noteworthy that these two pilots were routinely breaking the sound barrier at a time when only a small number of others, based primarily at Muroc Dry Lake, had done the same thing.

The F-86A was known to have problems with pitch-up and lateral control during transonic flight, and a flight program was carried out to document the aircraft's longitudinal stability characteristics for comparison with wind tunnel tests (ref. 25). Two F-86s were then flight tested to assess the effects of the height of the horizontal stabilizer on pitch stability and to check earlier wind tunnel data (ref. 26). It was found that the two tail locations tested had no bearing on the pitch-up characteristics. A more significant factor was the use of servo-powered longitudinal controls on one airplane, which eliminated the stick-free instability that characterized the original reversible controls. Comparisons of wind tunnel and flight data were inconsistent; .....

 

Figure 24. North American F-86A Sabre.

Figure 24. North American F-86A Sabre.

 

Figure 25. North American YF-93A with NACA submerged inlet.

Figure 25. North American YF-93A with NACA submerged inlet.

 

Figure 26. North American YF-93A with NACA scoop inlet.

Figure 26. North American YF-93A with NACA scoop inlet.

 

[15] ....these inconsistencies were attributed to detailed differences between the tunnel models and the aircraft and, as well, to the differences in Reynolds number. In further tests, Seth Anderson and Frederick Matteson investigated alterations of the wing leading edge to determine the influence on pitch-up. Leading-edge camber was found to have no effect on longitudinal stability, just as wind tunnel testing had indicated. Modest increases in maximum lift were observed, although abrupt asymmetric stall was experienced until flow fences were installed (ref. 27). Partial-span leading-edge chord extensions did eliminate pitch-up below Mach 0.84, but had no comparable influence at higher Mach numbers (ref. 28). Mel Sadoff, John Stewart, and George Cooper performed studies to correlate pilot opinion with the pitch-up characteristics of several aircraft. Using data from the F-84F, F-86A, D, and F models, and F-100 fighters and the B-47 bomber, they found that pilot opinion could be related to angle of attack and normal acceleration overshoots. Pitch-up tendencies were noted to range from mild to severe and, in the worst cases, design load factors and tail loads were exceeded for some of these airplanes (ref. 29).

 

Figure 27. North American YF-86D showing vortex generator installation.

Figure 27. North American YF-86D showing vortex generator installation.

 

Figure 28. Douglas F4D-1 Skyray with Don Heinle, Stew Rolls, and Walter Liewar.

Figure 28. Douglas F4D-1 Skyray with Don Heinle, Stew Rolls, and Walter Liewar.

 

Figure 29. Flight Research Section circa 1946. Front row: Bob Bishop, Mary Anderson, Helen Brummer, Mary Thompson, Larry Clousing, Chan Cathcart. Second row: Ben Gadeberg, Carl Stough, Bill Turner, George Galster, Betty Adams, Bob Reynolds. Third row: Stew Rolls, eorge Rathert, John Spreiter, Tom Keller, Mel Sadoff, Kinsenger, Paul Steffen. Fourth row: Welko Gasich, Maurie White, Steve Belsley, Bill Kauffman, Seth Anderson, Carl Hanson, Harvey Brown. Not pictured: Howard Turner, Dick Skoog, Gavras, Don Christopherson, Bunnel.

Figure 29. Flight Research Section circa 1946. Front row: Bob Bishop, Mary Anderson, Helen Brummer, Mary Thompson, Larry Clousing, Chan Cathcart. Second row: Ben Gadeberg, Carl Stough, Bill Turner, George Galster, Betty Adams, Bob Reynolds. Third row: Stew Rolls, eorge Rathert, John Spreiter, Tom Keller, Mel Sadoff, Kinsenger, Paul Steffen. Fourth row: Welko Gasich, Maurie White, Steve Belsley, Bill Kauffman, Seth Anderson, Carl Hanson, Harvey Brown. Not pictured: Howard Turner, Dick Skoog, Gavras, Don Christopherson, Bunnel.

 

[16] Lateral stability and control tests on the F-86 showed that wing dropping was caused by a directional asymmetry, and an abrupt increase in dihedral effect accompanied by a decrease in roll-control effectiveness (ref. 30). Later, the F-86A and D were both flown with vortex generators in attempts to improve longitudinal and lateral control; the vortex generators proved successful in alleviating these problems (ref. 31). Flow fences were also installed and tested on the F-86A and they too were shown to reduce the pitch-up tendency at transonic speeds (ref. 32). Pitch-up tendencies of the YP-84A-5 were also evaluated.

The YF-93A aircraft was the first to use flush NACA engine inlets. The flush inlet design had undergone extensive development in the 7- by 10-foot, 40- by 80-foot,....

 

Figure 30. New flight research hangar under construction circa 1945.

Figure 30. New flight research hangar under construction circa 1945.

 

Figure 31. Ames Research Center wind tunnels.

Figure 31. Ames Research Center wind tunnels.

 

[17] ....and 16-foot wind tunnels (fig. 31) under the guidance of Emmett Mossman. North American developed two YF-93A prototype aircraft from the F-86 Sabre design under Air Force sponsorship. One aircraft (AF 48-317) was built with flush inlets; the other had conventional scoop inlets. Two interchangeable tail sections were provided as well. 10 The NACA acquired both aircraft, and Stewart (Stew) Rolls conducted flight tests of them to compare the two inlet designs and to check results against data from the wind tunnel tests used in their development. Measurements were made of inlet pressure recovery and overall airplane drag of the aircraft. Flight data showed that the submerged inlet had higher pressure recovery and higher drag than the scoop inlet below Mach 0.89, with overall performance essentially the same between the two. Sealing the boundary-layer bleeds improved the performance of both inlets (ref. 33).

Thrust measurements were made on the YF-93A by Stew Rolls using a movable pitot-static and temperature probe in the jet exhaust. Gross thrust and airflow at the jet exit were obtained with the thrust measurement accuracy determined to be within 5 percent at full power (ref. 34).

The general performance characteristics of the F4D-1 aircraft were examined at Ames. One report on these tests (ref. 35) presents a thorough analysis of minimum drag and drag due to lift for this tailless delta wing configuration and compares the results with data from the 14-foot transonic wind tunnel. Tunnel measurements for minimum drag were generally lower than those obtained in flight, a discrepancy attributed to lack of detail in the wind tunnel model.

Ames had been conducting research on leading-edge vortices for low aspect ratio swept wings and, in response to discussions with French officials concerning a new wing planform for a supersonic transport, a program was initiated in the 40- by 80-foot wind tunnel to investigate the application of the so called Ogee planform to the F5D-1 aircraft. 11 The sharp and highly swept inboard portion of the leading edge produced a strong vortex that was shown in the wind tunnel tests to stabilize the airflow over the outboard portion of the wing. Subsequently, the F5D-1 was modified by mounting wooden extensions to the wing leading edge to model the Ogee design (fig. 32). Stew Rolls was the lead engineer on the project, and Fred Drinkwater performed the flight tests. Even under aggressive maneuvering, a stable vortex configuration was observed, relieving concerns about abrupt disturbances to the aircraft from asymmetric bursting of the vortices. 12 As noted in reference 36, the pilots were able to decrease the approach speed of the aircraft by 10 knots, reflecting the improved flight characteristics of the Ogee over the more conventional planform of the F5D-1. Data were provided to the Anglo-French team, which was in the process of designing the Concorde, giving it assurance that the Ogee planform was suitable for the aircraft.

In a postlude to this area of research, in the 1970s flight tests were performed to determine the response of aircraft when encountering trailing vortices in the wake of wide-body transport aircraft. This activity was carried out as part of a joint program...

 

Figure 32. Douglas F5d-1 Skylancer with Ogee wing planform.

Figure 32. Douglas F5d-1 Skylancer with Ogee wing planform.

 

[18]

Figure 33. Lear 23 in formation with Dryden B-747 and T-37.

Figure 33. Lear 23 in formation with Dryden B-747 and T-37.

 

....with the FAA that involved Ames, Langley, and the Flight Research Center to help the FAA determine if it was feasible to reduce aircraft separation during the terminal-area approach as a means of increasing airport capacity. The objectives of the flight program were to document the magnitude of wake-vortex upsets for different pairs of generating and trailing aircraft, to investigate different ideas for reducing the vortex strength, and to develop methods to predict the magnitude of a trailing aircraft's upset in the wake of a lead aircraft. 13 Along with other aircraft from the Flight Research Center, the Ames Lear 23 (fig. 33) was flown into the wakes of a Boeing 747 and Lockheed C-5A to obtain these measurements. Robert (Bob) Jacobsen was the project engineer and Fred Drinkwater and Glen Stinnett carried out much of the flying (ref.37). Richard Kurkowski had performed earlier tests of the wake of a Boeing 727. The program included tests in the 40- by 80-foot wind tunnel to obtain measurements of wake-vortex size and circulation using a newly developed laser velocimeter in order to see how aircraft configuration changes affected vortex strength. Flight tests employed the velocimeter for the same purpose. Experiments were also carried out on the Six-Degree-of-Freedom simulator by Robert Sammonds to develop valid simulation techniques based on flight experience and to extend the flight results. Bruce Tinling (ref. 38) and Barbara Short and Bob Jacobsen (ref. 39) generalized the flight experience using methods developed from the wind tunnel tests to predict the bank-angle upset imposed on a variety of aircraft types when following a heavy transport at different distances. This work contributed to the definition of the FAA's separation criteria for landing behind large heavy aircraft.

Aerodynamic research in flight, with the exception noted in the preceding paragraph, was concluded with the F5D-1 tests in 1961. At that time, high-performance flight research was transferred by NASA headquarters directive to the Flight Research Center at Edwards Air Force Base. The results of Ames' work over two decades led to a better understanding of aerodynamic effects on performance and loads in subsonic, transonic, and supersonic flight. Out of flight exploration of the transonic regime came the realization that there was an aerodynamic continuum from subsonic to supersonic flight with no aerodynamic impediment or "barrier" in the flow. Aerodynamicists were thus encouraged to press for wind tunnel facilities that could explore this flow region more generally. 14 George Cooper received the Octave Chanute and Arthur S. Flemming Awards in 1954 for his numerous contributions as a test pilot in several of these programs.


4. Harry Goett 1998: personal communication.
5. George Cooper 1998: personal communication.
6. George Cooper 1998: personal communication.
7. George Cooper 1998: personal communication.
8. Seth Anderson 1998: personal communication.
9. George Cooper 1998: personal communication.
10. Stew Rolls 1998: personal communication.
11. Bill Harper 1998: personal communication.
12. Fred Drinkwater 1998: personal communication.
13. Bob Jacobsen 1998: personal communication.
14. Bill Harper 1998: personal communication.


previousindexnext