Flightline: 73

"user314" (user314)
11/09/2020 at 13:30 • Filed to: flightline, Planelopnik, planelopnik history, nasa

The modified NF-15B STOL/MTD on display at the Edwards AFB Open House in 1989

In 1975, NASA’s !!!error: Indecipherable SUB-paragraph formatting!!! began to sponsor programs investigating the effect of two-dimensional thrust vectoring nozzles (TVNs) on aircraft takeoff, landing and maneuvering. Two years, Langley started a study with McDonnell Douglas to integrate TVNs onto an F-15, with a contract being awarded in 1984. The aircraft selected was the pre-production TF-15A (F-15B, USAF !!!error: Indecipherable SUB-paragraph formatting!!! ), the first two-seat Eagle built. Canard foreplanes were fabricated from F/A-18 horizontal stabilizers and set above the intakes at a 20° dihedral, and could be moved symmetrically for pitch or differentially for roll control.

NASA’s NF-15B showing the massive canards

After initial tests with only the canards, new thrust-reversing, thrust-vectoring nozzles built by Pratt & Whitney out of a titanium honeycomb, which could divert the thrust by up to 20° up or down, in unison or in opposition. The vector flaps could also close completely and act as a clamshell-style thrust reverser.

Closeup of the F-15 STOL/MTD’s thrust vectoring nozzles

Other changes included a beefed-up undercarriage capable of handling rough field landings at high rate of descent, upgrades to the APG-70 radar that allowed the MTD to locate the airfield with great precision and from a distance via a high-resolution ground-mapping mode, and improved cockpit controls and displays. The NF-15B was also equipped with a modified LANTIRN system to provide visuals to the pilot’s HUD. McDonnell Douglas and GE developed an integrated flight/propulsion control (IFPC) system to relieve the pilots of some of their tasks, with specific modes for conventional takeoff/approach, short takeoff/approach, short landing, cruise and combat. The IFPC was also tied into the quadruple-redundant fly-by-wire system.

The TVN’s and canards are apparent in this flyover of the NF-15B

The first flight with the thrust-vectoring nozzles took place on 16 May , 1989, after which the plane was transferred to Edwards AFB for joint flight tests by the Air Force and McDonnell Douglas on 25 May, 1989. The two-dimensional nozzles were first tested in flight on 23 March, 1990. Test flights demonstrated that the thrust-vectoring features of the new nozzles worked as anticipated and validated the changes that had been made to the IFPC system software in order to accommodate the new thrust-vectoring nozzles. It turned out that thrust-vectoring resulted in a 25% reduction in takeoff roll. The thrust-reversing feature made it possible for the F-15 to land on just 1650 feet of runway. In addition, it was found possible to use thrust reversal during actual flight to produce rapid deceleration, a useful feature to have during close-in air-to-air combat. During its flying life, the F-15 STOL/MTD made numerous vectored takeoffs with rotation demonstrated at speeds as low as 42 mph. The shortest landing made by the plane took only 1366 feet, with a basic F-15 needing 7500 feet to land. The STOL/MTD program ended successfully on 15 August 1991, with the TVNs being returned to P&W and the NF-15B placed in storage at McDonnell Douglas’ factory in St Louis.

Plan view of McDonnell-Douglas NF-15B STOL Demonstrator NASA 837 at the Edwards Air Force Base open house on October 10, 1999.

On 15 June 1993 the NF-15B was pulled from storage for a new NASA program, Advanced Control Technology for Integrated VEhicles (ACTIVE), under which it was given the tail number NASA 837, and fitted with new P&W !!!error: Indecipherable SUB-paragraph formatting!!! 100-PW-229 engines mated to TVNs which could vector 20° in any direction. The goal of the program was to test systems and techniques for thrust vectoring in the supersonic regime, as well as to explore the possibility of using thrust vectoring to replace vertical tailfins.

The NF-15B, marked with NASA logos and the ACTIVE mission patch, undergoing maintenance at Edwards AFB in 1995.

Initial flight trials began in February of 1996, with the first thrust vectoring flights taking place shortly thereafter. The first supersonic thrust vectoring took place on 24 April, 1996, and on 31 October, 1996, the aircraft demonstrated thrust vectoring at Mach 1.95.

McDonnell Douglas was interested in proceeding further with the ACTIVE program, with the aim of developing a stealthy variant of the F-15, tentatively identified as !!!error: Indecipherable SUB-paragraph formatting!!! , which would delete the vertical fin, rudders, and canards, and rely solely on the TVNs for control. The MANX, would it have been developed, would have been proposed to the South Korean Air Force for its F-X program, as well to the USAF as a possible replacement for the F-117 and F-15E. Neither party showed any interest however, and the MANX did not progress beyond renderings.

Artist’s conception of one of the possible configurations of the F-15 MANX

After the end of the ACTIVE program, the NF-15B was retained by NASA, and was next used in the Intelligent Flight Control System (IFCS) between 1999 and 2008. The objectives of the !!!error: Indecipherable SUB-paragraph formatting!!! program were to investigate next-generation flight control systems incorporating neural net learning computers to analyze adapt to changing flight conditions, providing improved flight characteristics, even if an aircraft is damaged. The project was carried out in two phases:

Generation 1

Generation 1 IFCS flight tests were conducted in 2003 to test the outputs of the neural network. In this phase, the neural network was pre-trained using flight characteristics obtained for the McDonnell Douglas F-15 STOL/MTD in a wind tunnel test and did not actually provide any control adjustments in flight. The outputs of the neural network were run directly to instrumentation for data collection purposes only.

Generation 2

Generation 2 IFCS tests were conducted in 2005 and used a fully integrated neural network. It is a direct adaptive system that continuously provides error corrections and then measures the effects of these corrections in order to learn new flight models or adjust existing ones. To measure the aircraft state, the neural network takes 31 inputs from the roll, pitch, and yaw axes and the control surfaces. If there is a difference between the aircraft state and model, the neural network adjusts the outputs of the primary flight computer through a dynamic inversion controller to bring the difference to zero before they are sent to the actuator control electronics which move the control surfaces.

The NF-15B is joined by NASA’s F-16XL during a flyover at Edwards AFB in 1999

NASA 837 showing the IFCS mission patch during an Edwards AFB Open House in 2005

The NF-15B was retired in 2009 after a 14 year career with NASA, and the plane was placed on display outside Armstrong Flight Research Center, though being on Edwards AFB limits access.