Rockwell MBB X-31

from Wikipedia, the free encyclopedia
Rockwell MBB X-31
Rockwell MBB X31.jpg
Type: Experimental airplane
Design country:
First flight:

October 11, 1990


Flight tests ended in 2003

Number of pieces:


The X-31 was a single beam experimental aircraft from US-American - German co-production. The aircraft based on the draft Tactical Combat Airplane 90 (TKF-90) was used for practical testing of the thrust vector control for flights beyond the maximum dynamic angle of attack . The concept of being able to continue flying in a controlled manner even after reaching the maximum angle of attack (English post- stall technology (PST) ) was invented by Messerschmitt-Bölkow-Blohm at the end of the 1970s and was intended to increase the maneuverability of future combat aircraft. MBB and above all the project manager and "father of the X-31", Wolfgang Herbst, saw what they called the supermaneuverability as an answer to the new infrared-guided short-range air-to-air missiles that can turn off targets from any angle, and no longer just from behind. The program manager Robinson pointed out that the X-31 project is one of the few with reverse technology flow for the United States , that is, that the United States would benefit from the knowledge of others, which so far only to a greater extent through systematic evaluation The high technology of the Third Reich (in the Paperclip project ) had succeeded in transferring considerable know-how from Germany to the USA.

The X-31 was the United States ' first X-aircraft to be developed through international cooperation, the first fighter aircraft whose thrust vector control (SVS) enabled control of movements around both the pitch and yaw axes, and the first to be made exclusively with the control stick could be flown. With the resumption of a new test section after the reactivation of the X-31 after a four-year break (stored in Palmdale , California), the test flights, as it were, without vertical stabilizers, were also the first in which an aircraft reached supersonic speed without the stabilizing effect of a vertical stabilizer. Furthermore, a 3D audio system and a virtual target display using augmented reality were tested. At the same time were from the NATO RTO Working Group 27 opinion polls among experienced pilots on the topics of 3D thrust vectoring , 12 g - flight envelope , helmet visor and negative G-loads carried out tests to assess the usefulness of these concepts.

The rollout took place on March 1, 1990, the first flight on October 11 of the same year. Two planes were created, one of which crashed during the tests. The Americans only used the X-31 for general experiments, such as air-to-ground attacks in the JAST program, and the US Navy was interested in the benefits of thrust vector control (SVS) when landing on aircraft carriers. Germany and other countries that wanted to participate in the X-31 program, on the other hand, wanted to implement the thrust vector control (SVS) in the Eurofighter Typhoon and Saab 39 Gripen and use the X-31 as a test aircraft for the nozzle and engine. For example, the Eurojet EJ200 engine with a thrust vector nozzle was to be installed and flown in the X-31 as part of a German-Spanish-American agreement; however, for various reasons this did not materialize.


The concept

MBB and Rockwell had been working on a joint, highly agile fighter aircraft design since 1981. After MBB left the EAP project , an agreement was reached with Rockwell in 1982 to pursue the concept of super maneuverability. MBB had already modified a Saab 37 with Saab to demonstrate the decoupling of flight path and fuselage alignment. MBB presented the concept to the Luftwaffe in 1983, but the latter decided not to incorporate it into the European Fighter Aircraft (EFA) due to technical immaturity . MBB then encouraged Rockwell to seek funding from the Defense Advanced Research Projects Agency (DARPA). Rockwell then submitted unsolicited plans for a Supernormal Kinetic Enhancement (Snake) test aircraft in 1983. From November 1984 studies were carried out to manufacture an inexpensive experimental aircraft on the basis of these findings. The aircraft was supposed to demonstrate the flight control after a stall, which MBB had been working on for years, in order to later integrate it into the EFA. In November 1984, the DARPA awarded a feasibility study to Rockwell, which in turn involved MBB as a subcontractor. In 1985, the Senators initiated Sam Nunn and Dan Quayle , the Nunn-Quayle NATO co-operative Research and Development Initiative , which was approved in 1986 by the US Congress.

Germany and the United States then signed a Memorandum of Understanding in May 1986 , which initiated the design phase that lasted until September 1987. In September 1986 DARPA awarded a preliminary contract for one year of development work. Relationships now leveled out as negotiations were conducted directly from government to government. DARPA was responsible for the order management, the head came from the US Navy, his deputy from the BMVg . MBB and Rockwell concluded an agreement that regulated the work shares and responsibilities: MBB was responsible for the flight control and control laws, thrust vector nozzle, air inlet and the CFRP double delta wings, Rockwell for the rest. The US Navy was interested in how the landing speed was based Aircraft carriers. In February 1987 the official designation X-31 was given to the project. It was the United States ' first X aircraft to be developed through international cooperation. The cost of the construction of two X-planes and their tests were performed with 75 million US dollars estimated. The MBB share of 20% was borne by the German state, Rockwell got its costs financed through the Nunn Quale Initiative. In 1987, tests began with the engine outlets in the United States. In September 1987 the design was finalized and manufacturing began.

A year later, in 1988, the project had to overcome its first hurdle: the US Senate was bothered by the 80/20 division of the work shares and refused the funds. The Pentagon then wrote a letter to the Senate stating that the United States had invited Germany to participate in the project and that withdrawal would embarrass the United States. MBB had already made a composite wing, the other was in the process of being made. The Senate eventually released $ 47.3 million for the next 22 months and the first 12 hours of test flight shortly before the end.

At the beginning of 1989 the date of the first flight was set to November / December 1989, the second aircraft was to take off three months later. By January 1991 there were to be 300 flight hours in 400 flights. Half of the flights were intended to extend the conventional flight envelope , the other half after a stall. The first series of tests was to take place at Rockwell in Palmdale, California (United States Air Force Plant 42). The tactical application was to be demonstrated in Naval Air Station Patuxent River , as the tracking capabilities of Nellis Air Force Base were available nearby. The American and German governments demanded that the project proceed quickly in order to demonstrate the tactical benefits of autumn maneuvers and others. The tactical evaluation should be done first against computer generated targets, then against the other X-31 without active thrust vector control, then against aggressor aircraft.

Maneuver tests

At the beginning of 1989 the final assembly of the first X-31 was in progress. On March 1, 1990, the first aircraft was rolled out. The first flight should be in late April, and the second X-31 should be ready in June. The flight control software was nearing completion and was tested in a flight simulator in California by MBB and Rockwell test pilots at an angle of attack (AOA) of up to 70 ° . The development took place in parallel in West Germany and the United States, the data was exchanged via modems . The excessive use of tried and tested components made it possible to reduce the cost of building the two X-31s to the level of two F-16s. In order to be able to dispense with breaking tests on an Iron Bird , all loads were oversized by 110%, which also enabled quick certification - Germany the wings and the thrust vector control, the rest of the United States.

For unspecified safety considerations, the first flight was postponed to mid-June. Ultimately, the maiden flight with Rockwell test pilot Ken Dyson took place on October 11, 1990, five months behind schedule due to fine-tuning with the fly-by-wire system. The test program should now be completed in 1992. On April 23, 1992, after a long maintenance interval, the first flight took place at NASA's Dryden Flight Research Center , after the two X-31s were relocated there. Here the flight envelope was to be expanded to be relocated to the NAS Patuxent River in early 1993 . The combat exercises should begin there. In May 1993, the stall test flights were completed. The first test battles against conventional combat aircraft were planned for December.

In October 1993 it was announced that the vertical stabilizer should be removed during test flights in 1994 in order to reduce air resistance and the radar signature. Another reason was that on flights with an angle of attack of more than 40 ° the rudder was useless, and from 45 ° the rudder was also useless. Before removing the rudder unit, the effect of a smaller rudder was first investigated by programming the rudder to be destabilizing and the effect had to be compensated for by means of thrust vector control (SVS). The goal was a quasi-vertical stabilizer-less variant, in which only the root of the stabilizer should remain. The project aimed to investigate the usefulness of thrust vector control (SVS) in supersonic flight. To this end, the sound barrier was to be broken for the first time in January 1994. The combat tests against F-18s were promising up to this point, maneuvers such as Pitch Reverse, J-Turn and Helicopter Gun Attack were evaluated. On March 17, 1994, the first flight with the neutralized vertical stabilizer was carried out and Mach 1.2 was reached. It was the first time that an aircraft without the stabilizing effect of a vertical tail reached supersonic speed. Radar signature, air resistance, fuel consumption and weight were named as advantages.

In early 1994 DASA, which was responsible for the control laws of the Eurofighter flight control system, worked on incorporating parts of the control laws of the X-31 into the Eurofighter project. Germany advocated equipping the Eurofighter with thrust vector technology later to increase combat value. Eurojet Turbo emphasized that this was not an official requirement, but that MTU had advertised it. At the same time, further flights without vertical stabilizers were announced. In early 1995 the Eurofighter consortium rejected the development of an alternative flight control system (FCS) based on the X-31. Although technically feasible, the current FCS met the aircraft's needs and requirements. It was feared that a radical change at this point would cause time and cost overruns.

At the turn of the year 1994/1995 the X-31 program was in financial difficulties: Germany agreed to bear 45 million US dollars or 50% of the costs of a follow-up test program, which the United States rejected. In addition, test pilot Karl Lang, who was able to save himself with the ejection seat, crashed with his X-31 on January 19, 1995. The preliminary accident investigation by Rockwell and DASA suspected icing of the pitot tube as the cause. The speed was displayed incorrectly, then oscillations around the pitch axis occurred. The briefing was also criticized because Lang was not informed before the flight that the pitot heating system had been removed for adjustments.

At the 1995 Paris Air Show

Meanwhile, the second machine was relocated to Manching on May 23, 1995 in a Lockheed C-5 Galaxy to take part in the Paris Air Show . Germany and the United States shared the cost of the demo flight. The X-31 team solicited more money from the US Navy, as the thrust vector control would allow the landing speed to be reduced to 80-90 kts without loss of control. The simulated approaches were flown in the quasi-vertical stabilizer-less mode, and simulations indicated 65–70 kts as a possible lower limit. The simulated carrier landings were flown up to 30 m above the ground and were financed with funds from the JAST program, which is why the usefulness of SVS in air-to-ground attacks was also examined. The United States now tried to go its own way with the F-16 MATV and F-15 ACTIVE.

ESTOL and Eurofighter

Since the X-31 had only consumed half of its useful life, financiers were sought in order to be able to carry out further test flights. In mid-1996 it became clear that Germany and Sweden wanted to evaluate the thrust vector technology in order to develop variants of their aircraft without vertical stabilizers. Germany wanted to use the test flights to get the thrust vector technology into the Eurofighter and Sweden into the JAS 39 . The US Navy had its eye on the F-18, and there was speculation that it could be used by the Air Force (F-15/16). The three-nation talks revolved around the implementation of thrust vector technology in Typhoon and Gripen, extreme short take-off properties for carrier landings without catch hooks and for landings on damaged runways with significant payloads, and the development of an advanced air data system without the boom on the nose as in the Having to use X-31.

In October 1997, Volvo confirmed that Saab, General Electric, Daimler-Benz Aerospace (DASA) and Boeing were in talks about the X-31-VECTOR program. The AVEN thrust vector nozzle from GE should also be tested. Volvo advocated installing an RM-12 engine with 80.5 kN thrust in the X-31, which had been developed by GE on the basis of the F404 for the Gripen. The talks stalled because of the financing; Germany in particular was once again struggling with a cut in its defense budget, but wanted to share in the costs. In early 1998, the three states were expected to sign a Memorandum of Understanding on the X-31-VECTOR program in March . The nozzle and air data system should be revised. Aloysius Rauen, head of DASA's military division, wanted to see the results of the VECTOR program implemented in the Eurofighter EF2000 or JAS 39 Gripen in order to catch up with Russia in terms of thrust vector control.

In June 1999 Germany and the United States signed the MoU, and Sweden is expected to sign it shortly. For the Thrust Vectoring Extremely Short Take-off and Landing, Tailless Operations Research (VECTOR) project, the X-31, which had been stored for years in NASA's Dryden Flight Test Center, was transferred to Boeing's factory in Palmdale. The US Navy took over the US role in the project again, Boeing had bought Rockwell in the meantime. Saab Aircraft and Volvo Aero were now also involved. In September, the United States and Germany grew impatient as Sweden could not bring itself to a decision due to cuts in defense budgets. The Spanish company ITP suggested that the thrust vector nozzle of the EJ200, which had recently been tested on the ground, be tested in the X-31 instead. The United States and Germany decided to run the VECTOR program without Sweden if necessary, with the existing petals on the stern. Spain could get in later and combine the nozzle with the F404. The test flights should start in 2000.

At the beginning of 2000 the plans became more specific: The US Navy held talks with ITP about the integration of the thrust vector nozzle into the X-31. It was also contemplated that Spain could rent test time on the plane or that the nozzle tests could be done according to the VECTOR program. Germany and the United States agreed to finance the 25 test months of the VECTOR program. Sweden withdrew from the program due to a lack of funding. At the end of 2000, an agreement between the United States, Spain and Germany was imminent. It was to be agreed to test-fly the thrust vector nozzle of the EJ200 in the X-31 from the end of 2002. This was intended to encourage the Eurofighter partner countries to order tranche 3 with thrust vector engines. ITP had already extensively tested the nozzle, but no Eurofighter was available for test flights. The costs of around 60 million US dollars to install an EJ200 with SVS in the X-31 were to be borne mainly by the Spanish government, the rest by Eurojet Turbo GmbH. The NATO EF 2000 and Tornado Development, Production & Logistics Management Agency approved the delivery of the engines, whereby these should be diverted from the Spanish Quantum. The US Navy would only be responsible for managing flight tests, but the Europeans were concerned about the EJ200's technology transfer to the United States. At this point in time, the industry had not yet agreed on the exact integration of the thrust vector technology into the Eurofighter. The ITP project manager, Daniel Ikaza, suggested initially only enabling pitch control (2D) in the aerodynamic flight range and later enabling 3D vector control even after a stall through software updates to the flight control software. The director of NETMA was convinced of the benefits of the thrust vector technology and saw an integration into the Eurofighter for tranche 3 and beyond. Because the oars have to be moved less, the permanent turn rates could be increased and the take-off distance reduced by 25%.

Landing at a 24 ° angle of attack

Regardless of this, the X-31 was ready for the VECTOR test series in November 2000. On February 24, 2001, the X-31 took off again after a six-year break. The 40-minute test flight from the Patuxent River NAS marked the beginning of the "reactivation tests," which were to last two to three months. From November the ESTOL flights were to begin on a virtual runway in the sky until real ESTOL landings in Patuxent River were planned in November 2002. Problems in getting the plane back into the air properly after six years resulted in delays and cost overruns. Since the 60 million US dollars had already been used up, there was not enough money for the third phase of the VECTOR program, so in August 2001 the Navy was on the lookout for additional funds. Meanwhile, the "landings" with a 40 ° angle of attack (AOA) were prepared on the virtual runway and the new air data system was installed. In the meantime, further discussions were held with the Spanish Ministry of Defense about installing the EJ200 with SVS.

On April 15, 2003, the first approach was flown on the virtual runway 1500 m above the Patuxent River NAS. On April 29, the fully automated ESTOL approach to the naval base was carried out. With a 24 ° angle of attack and 121 kts (224 km / h) landing speed, the landing distance could be reduced to 31%. Normally the flight had to be approached with an angle of attack of 12 ° and 175 kt, which required a runway of 2400 m. It only took 520 m for the aircraft to be slow enough to be able to turn a circle on the runway. The differential GPS guided the aircraft to the target with an accuracy of 2 cm. This was also the last flight of the remaining X-31.


X-31 EFM in the Schleissheim aircraft yard

The X-31 was the only international program in the series of legendary X-aircraft of the United States to be seen in 2004 at the International Air Show (ILA) in Berlin. For this purpose, she was transferred to Munich on June 22, 2003 on board an American transport plane. An exhibition space in the Schleissheim aircraft yard of the Deutsches Museum was then allocated. After five years it was supposed to be shown in an American museum, but for unknown reasons this never came about.

DASA then dealt with the integration of the thrust vector technology into the Eurofighter. The advantages mentioned are: post-stall maneuvers to increase maneuverability, supersonic trim and independent A 8 / A 9 control for higher supercruise speeds , shorter take-off and landing distances, more control surfaces and better close combat capabilities for lower loss rates. To do this, the advanced air data system and the thrust vector nozzle must be integrated and the tax laws rewritten; the failure of an engine in the post- stall maneuver (PST) should be tolerated. In addition, the turning rate of the Eurofighter also increases in the angle of attack range of less than 30 °. Since the loads on the aircraft fuselage and wing were already measured on the X-31 flights, the load changes for the Eurofighter can be estimated with PST and were classified as low. The structural reinforcements required are also low. It was also found that the effects of non-linear aerodynamics no longer play a role in the PST. Since the SVS increases the weight of the tail, a head ballast was proposed to maintain the balance of the unstable aircraft. The additional weight could be compensated for by smaller tanks or a reduction in the size of the vertical stabilizer by about 1/3. In the latter case, air resistance, flutter and radar signature would also benefit. Since the E-Scan-Radar is around 100 kg heavier than the CAPTOR-M, the head ballast can probably be dispensed with.



Both the EFA and the ATF were designed to destroy opponents on the approach before they were supposed to locate their own aircraft. Should the fight not be decided in the BVR battle , both sides would approach at a speed of about Mach 2 and try to fire at the opponent head-on with heat-seeking LFKs. Should this also fail, the on-board cannon would be on the train. A ranged battle would end in a dogfight in a few seconds . The latest LFKs with heat seekers made it possible for the first time to shoot an enemy head-on. When Rockwell and MBB began conducting studies on future combat aircraft technologies from 1981 onwards, they found that a significant number of skirmishes that begin at a distance end in hand-to-hand combat. The ability to fire WVR missiles at the enemy from any position radically changed combat tactics, and thus also the demands on a combat aircraft. Simulations showed that mutual destruction could now occur if both sides fire at each other. Analyzes showed that the aircraft that can fly a closer turn and shoot at the opponent first from the curve with a helmet visor achieves an advantage. Since the turning radius depends on the G-load and speed, the slowest possible speed would be necessary, which can be flown in a controlled manner. MBB came to the conclusion that a brief, controlled flight after a stall was necessary, followed by a further flight within aerodynamic limits. Over 25,000 manned and unmanned simulations were carried out by MBB in order to develop control forces, combat maneuvers and a suitable aerodynamic configuration.

In order to have an advantage over existing machines, the maneuverability would have to be increased significantly. The X-31A Enhanced Fighter Maneuverability (EFM) should expand the maneuver envelope beyond the previous limits of aerodynamics, engine technology, structural and pilot loads. According to program manager Michael Robinson, the X-31 should combine the BVR capabilities of EFA and ATF with the dogfight capabilities of an F-86 Saber . Specifically, Robinson named a shooting ratio of 10: 1. To do this, the X-31 was supposed to advance into areas of the flight envelope that no other aircraft could fly before, and thus achieve a tactical advantage. Flying at a higher angle of attack (angle of attack) below the Cornerspeed should achieve very high turning rates. Furthermore, the pilot should be able to align the aircraft like a gun turret thanks to thrust vector control. The X-31 should also have the potential for superior maneuvering performance in supersonic up to Mach 1.3.

"Spar-Eurofighter" X-31 taxiing

The aerodynamic design of the X-31 was based on the TKF-90 and, like the aerodynamics of the Eurofighter Typhoon, is a compromise of low supersonic drag, maximum lift, minimum induced drag, and a balance between instability at high angles of attack and the required pitch recovery -Torque at high angles of attack. The wing profile developed by MBB and optimized by Rockwell, with a thickness of 5% and a large leading edge radius, was intended to improve performance at high angles of attack (angle of attack). Split leading and trailing edge flaps were installed as in the Eurofighter. Garrett's actuators for the trailing edge flaps were also located in "bathtubs" under the wings, while those of the leading edges were hidden in the wing. The outer flaps also worked as ailerons , as the SVS could not generate any roll moments. The duck wings had a symmetrical profile. In the short term, it was considered to use the duck wings of the B-1B bomber for cost reasons , but this was not realized because of the weight. The wings were dry and made of aluminum with CFRP planking. The fuel tank and avionics were carried in the center of gravity of the aircraft, the measurement technology in the nose and a spin screen in the rear. The fuselage between the cockpit canopy and the wing trailing edge used a constant cross-section so that the eleven aluminum frames, which were connected to the four longitudinal frames, could be mass-produced cheaply with a hydraulic press. The stern tapered to reduce transonic drag. The front paneling consisted of CFRP panels from the B-1B program, as well as the vertical stabilizer, but made of sheet metal on the fuselage. The rudder was made of CFRP with a honeycomb core. The central fuselage was made of titanium frames and planked with titanium sheets, as nobody knew what the heat loads for the fuselage would be when using the thrust vector nozzle. The air inlet came from MBB experiments and, thanks to its movable lower lip, was able to deflect the air flow with little turbulence at high angles of attack and reduce the overflow resistance at high speeds. The air above the boundary layer separator was passed through heat exchangers to cool the oil and fuel. The strakes between the wing trailing edge and the nozzle should help get the nose back down from high angles of attack.

X-31 seen from the stern. The petals, which are used to deflect the jet of the engine, are clearly visible.

To reduce costs, the X-31 was supersonic, but could only achieve transonic speeds. The F404 engine was equipped with three petals that could deflect the thrust jet by 10 ° in the vertical and / or horizontal. According to the contract, the planes should be as cheap as possible. 43% of the curb weight of an X-31 came from F-16 and F-18 combat aircraft. The thrust vector petals were originally developed for a test with the F-14 Tomcat and were made of carbon fiber reinforced carbon . Metallic petals were also tested with the F-18, but these were too heavy and would have shifted the center of gravity too far back. According to studies, a thrust angle of 10 to 15 ° with a full afterburner was necessary to meet the requirements. The petals reached 10 °, which was about 17% of the engine thrust. In aerodynamic flight, the spreading petals could also be used as an air brake. Series machines would use a "clean" nozzle, but that would have been too expensive and time-consuming for the X-31. Menasco took over the landing gear from a Cessna Citation III , the main wheels from an A-7 and the front wheel from an F-16. The cockpit, ejection seat, actuators and displays were carried over from the F-18, fuel pumps, leading edge actuators and rudder pedals from the F-16. Other parts have been adopted from the F-5, F-20, V-22 and T-2C. The wings had been made to metric measurements by MBB, while Rockwell had used the imperial system.

Flight control computer

The flight control laws were written by MBB and, according to program manager Michael Robinson, were the most critical component of the entire aircraft. The flight control computers (FCC) were supplied by Honeywell. During development, MBB relied on experience with the F-104 CCV.

High angle of attack flight during the fall maneuver

The X-31 was the first combat aircraft whose thrust vector control (SVS) enabled control around the pitch and yaw axes and the first that could only be flown with the control stick. The rudder pedals were only needed for deliberate side gliding and cross wind landings. The pitch control was mainly based on the trailing edge flaps and was supplemented by the canards. The leading edge flaps were adjusted according to the angle of attack and the Mach number. In order to reduce the trim resistance in the supersonic, the X-31 was designed to be moderately unstable with 5% of the mean aerodynamic wing depth (MAC) at Mach 0.2. The control laws had been developed by MBB in 25,000 simulations when the project was still being carried out at MBB. They were implemented in a triple redundant computer system, with a fourth computer being used as the (original) " tie-breaker " and providing additional redundancy. The flight control also had a mode in which the nose of the aircraft could be decoupled from the flight direction: The pilot could pull the nose of the machine in the desired direction, while the change in the pitch and yaw axes was only unfavorably implemented on the trajectory.

The duck wings rotated downwards at high angles of attack in order to continue to steer the aircraft in a controlled manner. They were designed for pitch recovery rather than as a buoyancy aid. Maneuvering after an air flow stall (Post-Stall Technology, PST) was permitted up to an angle of attack of 70 °, in a height range of 10 to 30 kft, if a maximum of 4  g and 225 kcas initial speed were not exceeded. For the control laws, a linearized model for the movement around the three axes with a feedback loop for stability and handling was implemented. The flight control could be adjusted using weighting matrices. The feedforward control was independent of the feedback path using equations for static movements, i.e. H. assuming constant pilot inputs. Due to the complexity, lack of computing power, and model weaknesses, these equations were simplified. The position of the canard and trailing edge flaps was read from trim tables by the FCC during level flight. Using additional calculation methods, the air resistance was reduced at low angles of attack and the stability was improved at high angles of attack. Since this "cruise trim planning" would have led to high landing speeds, another trim table for maximum lift for landing was implemented, which the pilot could activate using a switch. The required pitch rate was calculated in real time for canard, trailing edge and the vertical thrust vector.

The angle of attack and G-load were commanded by means of a control stick: at low static pressures according to the angle of attack, but non-linear, since 2/3 of the play meant an angle of attack of 30 ° and a full deflection of 70 °. When the PST was switched off, a maximum angle of 30 ° could be achieved. At 2/3 the travel of the control stick there was a force feedback limit to show the pilot whether he was stalling. At higher dynamic pressures, G-load was commanded; the load limit of 7.2 g was applied for 2/3 of the distance  and did not change after the force feedback limit. The change between the angle of attack and G commands took place at around 2620 kPa dynamic pressure. The aircraft could only stall when steered in angle of attack mode.

The maximum roll rate was 240 ° / s. The rudder pedals allowed a maximum of 12 ° slip angle to be required. The roll around the longitudinal axis was automatically stabilized against yaw by the flight control computer (FCC), and the yaw commands were set to zero at the maximum roll rate. At high angles of attack, yaw control was perceived by the SVS. Only one SV-Petal could be placed in the thrust jet with a maximum of 26 ° (max. 16 ° deflection of the thrust jet), the FCC also read table values ​​here depending on the thrust and control commands. Should the SVS fail, the FCC was able to bring the machine back into the aerodynamically controllable range in a transition mode using aerodynamic control surfaces, even in the pitch axis. Only around the yaw axis there was no longer sufficient control, which is why rudder and asymmetrical flap commands were deactivated in this case. Carefree handling was demonstrated in the subsonic, but no tail slides were flown. No problems arose during the flight tests, only at angles of attack over 30 ° (especially around 50 °) the behavior in terms of roll rate and angle of slip deviated greatly from the prediction, which is why a feedback loop was integrated. Two updates - shortly before the first test flight and after - showed the flexibility of the control laws (or software): The feedback loops were redesigned in less than a month.

Cockpit and helmet visor

Air Force test pilot Quirin Kim with the GEC Marconi helmet visor

The cockpit was taken over from the F / A-18, including the SJU-5 ejector seat. The avionics software was written in JOVIAL , not in Ada , as is common in defense projects . In order to avoid a disorientation of the pilot when flying at high angles of attack and to improve his situational awareness, various aids were tested. For this purpose, the head-mounted display was supplemented by a 3D audio system and the end product was called the Helmet Mounted Visual & Audio Display (HMVAD) . The tests began in February 1993. For this purpose, the GEC -I-NIGHTS helmet with integrated residual light amplifiers for night vision was first tested. Since the weight was unacceptable and the X-31 never flew at night, the residual light intensifiers were removed. The helmets were used in simulators and worn on acclimatization flights in T-38s. Because the aircraft's test program was pushed back, GEC offered the new Viper helmet in September 1993. Since the helmet was about a kilogram lighter, it was accepted as the new HMVAD. The 3D audio system, developed by the Armstrong Laboratory of the USAF, required the installation of special headphones. The system, which could display the flight path and / or angle of attack via audio signal, was previously tested on an AV-8B and OV-10.

With only six months from placing the order to the flight test, there was no time to develop HMD symbols. At the beginning of 1993 advice was given and GEC was given 60 days to program symbology. At the same time, test pilots evaluated the displays in the simulator in order to incorporate improvements as quickly as possible. In addition to the normal HMD displays, two new ones were also tested, which should improve orientation when flying at a high angle of attack: Arc Segmented Altitude Reference (ASAR) from DASA and the theta reference system from USAF. At the ASAR a kind of “U” was displayed in the lower viewing area. If the "U" covers a semicircle, the pitch angle is zero, with less positive, with more negative. Markers at the ends indicated the horizon. The theta reference system displays a mini-globe in the lower field of view, the lower hemisphere of which has dashed longitudes , while the upper hemisphere has solid lines. By moving the mini-globe, the orientation of the aircraft relative to the earth could be read, with N / E / S / W overlays on the axes. In addition, another angle of attack indicator was displayed in the middle of the field of vision: two triangles with the same base, the tips of which rise with increasing angle of attack. Between 0 ° and 30 ° angle of attack the triangles lay on top of each other, if the angle of attack was more than 30 ° the lower one stopped at 30 °, while the other triangle tapered further upwards. In order to know in which direction a specific target was even at extreme angles of attack, a circle was faded in in the upper field of vision, with "N" for north and a gap for the current direction of view of the pilot. During the flight tests, the target circle for the helmet sights was also modified. The target circle was then dashed when the target was outside the ± 30 ° launch envelope of the simulated missile.

During the flight tests, the pilots' feedback gave the following picture: The display of the relative flight altitude and the compass direction in air combat was superfluous because the enemy machine was sufficient as a reference. The ASAR was too imprecise, the theta reference system was found to be good and simple. Speed, altitude and angle of attack should always be displayed. For the angle of attack, the division into 30/50/70 ° was sufficient. Vertical speed and specific excess power were also displayed and found to be superfluous.

The tactical evaluation of the X-31 was not only based on flights against real machines. In addition, DASA was also working on using augmented reality to display a virtual enemy for aerial combat in the HMVAD. This should save costs compared to real aircraft, at the same time there would be no risk of collision in close combat. It can also be used to practice close aerial combat against machines that are not available to your own side.

According to the realities in air combat, the machine should be visible up to 3000 m (6000 m with display) and its position in space recognizable up to 4000 m. Due to the limited computing power, the “enemy combat aircraft” should consist of as few vectors as possible, and so the grid structure looked like an MBB Lampyridae without an air inlet. Below 60–80 m, the virtual firefly could no longer be displayed on the HMD because the textures were not sufficient to recognize structural details. Due to the limited computing power - precisely because of the lack of texture - it was difficult to estimate the approach rate below 200 m. To increase the fun of the game, the virtual fighter plane was moved by a target maneuver generator . Based on the movements of one's own aircraft and taking into account the flight mechanics and performance of the virtual opponent, the displayed aircraft carried out tactical maneuvers (sequences), as would be expected from a real opponent. Predefined maneuvers (keeping course, turning with a constant radius, etc.), aerial combat maneuvers from an offensive or defensive position and close combat maneuvers could be selected. The Target Maneuver Generator directed the virtual opponent and selected his thrust and fire commands. The display, which could be controlled in six degrees of freedom, was updated 20 times per second. The principle proved to be suitable and was adopted by the pilots as a training tool.


Installation of the RM-12 engine of the Saab 39 Gripen

Three GE F404s were modified as jet engines for the X-31 Enhanced Fighter Maneuverability. The General Electric F404 was chosen because it is relatively insensitive to the turbulence that usually occurs at high angles of attack. Thanks to the air intake of the X-31, the engine was still able to deliver full power even at extremely high angles of attack. At the start of the test flights, 18 flights were carried out to check the airworthiness of the combination of the X-31 airframe and F404 engine. A total of 80 flights were carried out to validate the flight envelope .

Before the start of the VECTOR (Vectoring, Extremely Short Takeoff and Landing, Control and Tailless Operation Research) test flights, it was checked whether the RM-12 of the Saab 39 Gripen could be built into the X-31. The RM-12 is a variant of the General Electric F404. This attempt, as part of phase 1, to get the X-31 back into the air, was successful: the engine fitted into the fuselage. The next step to equip the engine with the thrust vector nozzle AVEN (Axisymmetric Vectoring Exhaust Nozzle) failed because of Sweden's lack of funds.

The installation of the Eurojet thrust vector nozzle in the GE F404 by the Spanish company ITP was also rejected. At the end of 2000 it was to be agreed to install the Eurojet EJ200 with a thrust vector nozzle in the X-31 within the framework of a German-Spanish-American agreement and to test it. The costs of around 60 million US dollars to install an EJ200 with SVS in the X-31 were to be borne mainly by the Spanish government, the rest by Eurojet Turbo GmbH. However, the US Navy would be responsible for managing flight tests, and the Europeans were concerned about the EJ200's technology transfer to the United States. So this plan was not implemented either. Ultimately, the X-31 always flew with the General Electric F404, which had the thrust nozzle removed. Instead, three CFC- PETALs were mounted on the rear of the aircraft to deflect the exhaust jet. The range of motion of the petals inwards and outwards was + 35 ° / −60 °.

Test program

In simulator flights it was determined that the launch ratio could be increased by at least a factor of 2 through super maneuverability, even if one's own forces were numerically inferior. The simulations were carried out in 1979 by IABG . The design of the LVJ-90 / Jäger-90 / Prälo-Eurofighter with and without thrust vector control (SVS) was compared. In 125 simulation runs the dogfight was simulated with cannon only, in 331 the fight with cannon and short-range air-to-air guided missile (SRM). In air combat with only an on-board cannon, the thrust vector machine was able to fire more often than first (2.5: 1), get into the firing position more quickly (10: 1) and achieve a launch ratio of 4.2: 1. With the help of SRM, the thrust vector machine was able to fire more often than first (2: 1), get into the firing position more quickly (2: 1 LFK, 8: 1 BK) and achieve a launch ratio of 5.4: 1.

F-18 and X-31 in parallel flight

During the dogfights over the NAS Patuxent River between October 1991 and 1995, IABG carried out further computer simulations between October 1991 and April 1993 with regard to test setup and expectations. After the pilots to flying with post- stall technology (PST) were familiar, ungeskriptete combat maneuvers were flown against other machines. The test setup was also changed, for example by limiting the angle of attack of the X-31 to a maximum of 45 ° or setting the maximum speed for PST to 265 kts. Various missile envelopes have also been adopted, for example a restriction on shooting from high angles of attack or the use of a helmet visor. The "opponents" were F-14B / D and F-18C from the VX-4, as well as F-15C and F-16B 52 from the 422 TES. Some of the results are confidential.

It was published that a 70 ° angle of attack achieved better results than a 45 ° angle of attack. If PST was deactivated, the X-31 mostly lost to the degraded F-18C, which was supposed to perform similarly (15% won, 46% lost, 39% undecided). With PST the X-31 clearly dominated the battle against the F-18C from a neutral starting position (91% won, 3% lost, 6% undecided). The launch ratio was around 1: 2 in favor of the F-18 if the X-31 waived SVS, and 8: 1 in favor of the X-31 if it used PST. If the take-off position began in slow parallel flight, the X-31 was able to clearly dominate over all of its opponents; launch ratios of over 100: 1 were achieved. If the battle began in a fast parallel flight, it was still 16.6: 1. When evaluating the data, it was found that the X-31 pilots had "shot" their opponents mostly from the 2 o'clock to 5 o'clock position (or the other way around) with the on-board cannon. The range of the shot was up to 3000 ft, with hitting less than 2000 ft. In combat with short-range air-to-air guided missiles (SRM), the simulated missile was mostly fired when the enemy showed his 1 o'clock to 5 o'clock position (or mirror-inverted) of the X-31. The furthest shots at 10,000 ft were fired in the 1–2 o'clock range, from 2–5 o'clock from less than 4000 ft. To avoid the SRM shot, the enemy was forced to be closer than 2000 ft to the X-31 come where he fell victim to the PST. At a distance, the X-31 with the SRM could shoot from high angles of attack before the enemy came within minimal range. From the data, Rockwell concluded that all future dogfighters would have vector thrust control.

X-31 without vertical tail; the 1994 picture has been retouched

On March 17, 2004, a flight "without vertical stabilizer" was carried out for the first time. Radar signature, air resistance, fuel consumption and weight were named as advantages. The test flights were carried out for the JAST program and were intended to evaluate the controllability of the configuration for precise air-to-ground missions. The flight control software was designed by DASA (more precisely Hermann Beh and Georg Hofinger) and expanded the SVS's area of ​​application to include take-off, landing and ground attacks. The X-31 with vertical tail became unstable at an angle of attack of more than 25 ° around the vertical axis, so that ultimately only this effect had to be reinforced. For this purpose, the stabilizing effect of the rudder was switched off and a destabilizing feedback loop was installed, which controlled the (aileron) rudders. The SVS was then responsible for stabilization. The degree of destabilization could be selected dynamically by the pilot on a panel during the flight, the software was flexible. From 30% to 80% without vertical stabilizers, the flight control computers were able to simulate everything in 10% steps. The lower part of the SV envelope was then lowered from at least 14,000 ft to 2200 ft MSL. Due to poor handling qualities, the actuators of the thrust vector petals had to be replaced. After the envelope expansion, landing and marching flights were tested. It was found that the destabilizing effect of the flaps and landing gear was synergistic, i.e. the disruptive effect was greater than the sum of the individual effects. Flight maneuvers went smoothly. Simulated landing approaches on an "aircraft carrier" were also tested. Due to the high landing speed of the X-31, however, an angle of attack of 12–13 ° was necessary, which made the view from the cockpit unacceptably worse. Since increasing the air resistance of the X-31 was out of the question (due to stability issues, etc.), the approach profile was changed. In the following air-to-ground tests, dive aircraft attacks were flown against simulated MANPADS and mobile anti- aircraft missiles with a 45 ° descent angle. The aim was to fly attacks outside the threat destruction range and to keep the targets precisely in the HUD. To do this, a dive was started from 18,000 ft and 250 KCAS, the simulated weapon was disengaged at 12,000 ft at 400 KCAS and pulled into a climb in a 4 to 4.5 g maneuver. The second exercise consisted of a 15 ° descent with simulated fire at ground targets with the on-board cannon. The target area was represented by a light field in which five to seven targets alternately lit up per attack. Due to the quick target changes, the pilots (Kim, Luftwaffe and Loria, USMC) were forced to work aggressively with pedals and sticks. The third exercise was a low-level pop-up attack. From 1000 ft and 400 KCAS, a climb was pulled, relieved and a 4 g roll flown onto the target at an altitude of 2500 ft in order to bring it into the HUD. The aircraft was then relieved, held in level flight for 15 seconds, and then the use of weapons was simulated at 1500 ft at 400 KCAS. Then 4.5  g was pulled into the climb. The aircraft only passed all air-to-ground tests adequately; the poor roll rate in particular was criticized. The analysis found that the limits of the FCS on the lateral movements of the control stick were unnecessarily high. After a revision of the control laws and the more flexible choice of engine thrust, simulations were carried out which showed better results. Previously, less than 50% of the engine thrust margin was used, now it was possible to maneuver with a full afterburner. The test flights carried out later were correspondingly positive. Most of the maneuvers were flown with 50–60% vertical stabilization. With the pop-up, however, only 33% of the thrust vector was available for changing the path, the rest was needed by the computer for stabilization. As a result of the test series it was concluded that fighter-bombers without vertical stabilizers are possible. It was suggested to build an aircraft without vertical stabilizers in order to be able to better use the advantages (radar signature, air resistance, weight). This led to the McDonnell Douglas X-36 . In order to increase redundancy and damage tolerance and to be able to better balance asymmetrical external loads , twin-engine aircraft were recommended.

X-31 VECTOR approaching for landing

After a six-year break, the X-31 took off again in early 2001 for the VECTOR (Vectoring, Extremely Short Takeoff and Landing, Control and Tailless Operation Research) test flights. The VECTOR program should demonstrate a reduction in landing speed of at least 40% in around 60 test flights (Extremely Short Take-Off and Landing, ESTOL). Further data should also be collected on configurations without vertical stabilizers. A differential GPS from IntegriNautics and the new air data system from DASA, which consisted of eleven concentric holes in the nose, were installed. First, test flights were flown to validate the air data system, and Mach 1.18 and 70 ° angle of attack were achieved. Because of the system that worked with holes and pressure differences, the data for the boom that was mounted on the nose could be dispensed with. Then the ESTOL tests began: In order to reduce the landing distance, the approaches to the runway should be made with an angle of attack of up to 40 ° in order to reduce the landing speed by 40–50%, to around 90 kts (170 km / h). The ESTOL landing approaches were flown automatically by the flight control computer. The pilot just flew the aircraft into a launch box, activated landing mode and took his hands off the control inputs. The machine then flew standing on the thrust jet, controlled by SVS, at a 40 ° angle of attack towards the runway and tilted forward 2 ft above the ground, shortly before touching down. Several test landings were carried out. From a 12 ° angle of attack in 2 ° steps (which were partly skipped), finally 40 ° angle of attack was reached in the landing approach. The limit of 40 ° angle of attack was set for redundancy reasons on the X-31; in terms of controllability, higher angles of attack would have been possible. The loads on the landing gear when hitting were in the green area. The aircraft had a camera in its nose so that the pilot could follow the flight path on the displays. In a series machine, the image should be projected into the HMD. The US Navy was interested in increasing the landing weight of carrier aircraft.

Technical specifications

Three-sided view of the X-31
Parameter Data
crew 1
g limits 7.2 g 1
length 13.2 m
span 7.3 m
height 4.4 m
Wing area 21 m²
Empty mass 5443 kg
Max. Takeoff mass 7303 kg 2
Wing loading
  • minimum (empty weight): 259 kg / m²
  • maximum (max. take-off weight): 348 kg / m²
Engine 1 × GE F404-GE-400
  • with afterburner: 1 × 71 kN
  • without afterburner: 1 × 47 kN
Top speed Mach 1.3
Thrust-to-weight ratio
  • maximum (empty weight): 1.3
  • minimum (max.start mass): 0.97
1Limited. Breaking load 15.1  g , because it is 2.1 times (110%) oversized
2 includes 4100 pounds (1860 kg) of fuel

Web links

Commons : Rockwell-MBB X-31  - Album with pictures, videos and audio files



  1. a b The modern demonstration of “super maneuverability” usually consists of pulling into the stall during flight shows in order to perform the most spectacular turns possible. The loss of speed is very high. This has little to do with the original idea of ​​Wolfgang Herbst or Messerschmitt-Bölkow-Blohm (MBB). At that time, Herbst postulated the following maneuver characteristics: 1) 5 seconds PST duration on average, 2) 10% of the total close combat time in PST, 3) low G-loads of around 1  g , 4) lower maneuver speeds of around 0.1 Mach ( W. Herbst: Supermaneuverability. 1983, Messerschmitt-Bolkow-Blohm; and Introduction of the RTO HFM Lecture Series Human Consequences of Agile Aircraft. 2001). The PST maneuvers should therefore only be used for transients, for example to straighten the nose instead of a slow turn in order to fly an even tighter curve ( autumn maneuver ). The thought of heaving a 20-ton machine into the stall and then letting it fly on due to its inertia was completely alien to him.
  2. A 8 = nozzle throat area, A 9 = nozzle exit area
  3. The minimum turning speed with the maximum load factor. The highest turning rate is achieved here.

Opinion polls

In the period from April 1997 to October 1998, the NATO RTO Working Group 27 conducted two opinion polls among fighter aircraft pilots. The first “Operational Need” survey dealt with the question of what usefulness pilots would ascribe to certain abilities of an agile combat aircraft. 23 US pilots (5 NASA, 13 USAF Air Warfare Center, 5 USAF), 11 from the Swedish Air Force, 3 from the German Air Force and 2 from France were involved. The flight experience was 900 to 9000 and an average of 2589 hours. The pilots represented the crème de la crème of pilots with flight experience from X-31, F-18 HARV, F-15 ACTIVE, F-16 MATV, Harrier, F-22 Raptor, F-18, MiG-29, Rafale, Gripen and Typhoon. After an open discussion in an anonymous questionnaire, the usefulness of 12 g - flight envelope , helmet visor, negative G-loads and flight with high angles of attack / thrust vector technology , as well as the performance of two American anti-G suits on a scale of 1 should be assessed to 7.
The pilots rated the usefulness of the helmet visor the highest, followed by flight at high angles of attack / 3D thrust vector technique. This was followed by the 12 g envelope and the negative G loads. The evaluation was sometimes very different depending on the country : Swedish pilots rated everything (in absolute numbers) somewhat lower, but the 12 g envelope was the highest. US pilots, who made up the bulk of the pilots in the survey, consequently dominated the survey with their “national preference” for helmet visors, high angle of attack / 3D thrust vector technology and 12 g envelope, although the gradations were small. German pilots also gave the helmet visor the highest value, but rated the 12 g envelope and high angle of attack / 3D thrust vector technology as equally useful. The German pilots rated negative G-loads better than their Swedish colleagues, but worse than the US pilots.
In the discussion, perceived problems and desires of various systems were addressed: good field of view with HMDs, problems with uncomfortability and G-LOC at 12  g , uselessness of the HUD if one does not look directly ahead, disorientation when flying at high angles of attack, fear, off Accidentally stalling and then quickly losing energy (which is why the X-31 was equipped with a force feedback limit) and the demand for carefree maneuvering. The experienced pilots were satisfied with the HOTAS technology; operating 50 functions was not seen as a problem. Touch screens were seen as an immature technology and the reliability of voice control was questioned. Automatic collision avoidance was required.
In the second “Situational Awareness” survey, 29 pilots were asked cognitive and physiological questions about situational awareness. Of these, three came from Germany, twelve from Sweden, eight from France, five from the Netherlands and one from the United States. The flight experience averaged 2490 hours on types such as F-16 Falcon, MiG-29, JAS 39 and Mirage 2000. The questions concerned, among other things, the usefulness of a two-person cockpit (52% negative, 38% positive, 10% undecided ), Use of HUD, HMD, 3D audio system and voice input, as well as questions about pilot training and the acceptance (positive) of automatic maneuvers.

Individual evidence

  1. a b c d e f g h i j k Holger Friehmelt et al .: X-31A TACTICAL UTILITY FLIGHT TESTING . In: AGARD Flight Vehicle Integration Panel Symposium on Advances in Flight Testing . September 1996 ( [PDF]). X-31A TACTICAL UTILITY FLIGHT TESTING ( Memento of the original from March 4, 2016 in the Internet Archive ) Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice.  @1@ 2Template: Webachiv / IABot /
  2. a b c d e f g h i j X-31: breaking the stall barrier. (PDF) In: Flight International. July 11, 1987, accessed April 21, 2014 .
  3. a b Partners find X-31 funding. In: Flight International. July 1, 1998, accessed April 21, 2014 . or X-31A receives funding. In: Flight International. September 24, 1988, accessed April 21, 2014 .
  4. a b c d e MBB stresses its involvement in X-31. In: Flight International. June 20, 1987, accessed April 21, 2014 .
  5. a b c d e f g h i j k l m n o p q Flight beyond normal limits. (PDF) In: Flight International. May 9, 1990, accessed April 21, 2014 .
  6. a b c X-31 design frozen. In: Flight International. September 5, 1987, accessed April 21, 2014 .
  7. a b c d e f g Rockwell plans X-31 testing. (PDF) In: Flight International. January 7, 1989, accessed April 21, 2014 .
  8. a b c Rockwell and MBB roll out X-31A. In: Flight International. March 7, 1990, accessed April 21, 2014 .
  9. X-31A first flight five months late. (PDF) In: Flight International. October 17, 1990, accessed April 21, 2014 .
  10. Relocated X-31 flight tests resumed. In: Flight International. May 13, 1992, accessed April 21, 2014 .
  11. X-31 post-stall maneuvers accomplished. In: Flight International. May 19, 1993, accessed April 21, 2014 .
  12. X-31A vertical tail to be removed intrials. (PDF) In: Flight International. October 6, 1993, accessed April 21, 2014 .
  13. a b Rockwell and DASA fly 'tailless' X-31. (PDF) In: Flight International. March 30, 1994, accessed April 21, 2014 .
  14. Eurofighter looks to use key X-31 technology. (PDF) In: Flight International. March 30, 1994, accessed April 21, 2014 .
  15. X-31 thrust-vectoring option for Eurofighter. (PDF) In: Flight International. April 23, 1994, accessed April 21, 2014 .
  16. a b Eurofighter rejects new control proposal. In: Flight International. January 25, 1995, accessed April 21, 2014 .
  17. X-31 wrecked after test-flight crash. In: Flight International. January 25, 1995, accessed April 21, 2014 .
  18. ^ Pitot icing suspected in X-31 crash. In: Flight International. February 8, 1995, accessed April 21, 2014 .
  19. a b X-31 crash pilot 'badly briefed'. In: Flight International. June 7, 1995, accessed April 21, 2014 .
  20. a b Turning heads. (PDF) In: Flight International. June 28, 1995, accessed April 21, 2014 .
  21. Europe and USA hold thrust-vectoring talks. In: Flight International. May 15, 1996, accessed April 30, 2014 .
  22. Sweden joins talks to extend X-31 demonstrator project. In: Flight International. May 22, 1996, accessed April 21, 2014 .
  23. Sweden funds JAS39 demonstrator. In: Flight International. October 1, 1997, accessed April 21, 2014 .
  24. Partners poised to restart X-31 VECTOR. In: Flight International. February 18, 1998, accessed April 21, 2014 .
  25. ^ X-31 to fly again under tri-national MoU. In: Flight International. June 23, 1999, accessed April 21, 2014 .
  26. a b Sweden pressed over funding next phase of X-31 programs. In: Flight International. September 15, 1999, accessed April 21, 2014 .
  27. US Navy options for X-31 follow-on. In: Flight International. April 18, 2000, accessed April 21, 2014 .
  28. a b c X-31 to be used for thrust vectoring Eurofighter test. In: Flight International. November 14, 2000, accessed April 21, 2014 .
  29. a b c X-31 prepares for extreme STOL. In: Flight International. March 6, 2001, accessed April 21, 2014 .
  30. US Navy seeks more funds for X-31. In: Flight International. August 14, 2001, accessed April 21, 2014 .
  31. a b c X-31 completes first ESTOL testing. In: Flight International. April 15, 2003, accessed April 21, 2014 .
  32. VECTOR in high AoA landing. In: Flight International. May 13, 2003, accessed April 21, 2014 .
  33. Press release EADS: German-American experimental aircraft X-31 at the ILA in Berlin. May 10, 2004  ( Page no longer available , search in web archivesInfo: The link was automatically marked as defective. Please check the link according to the instructions and then remove this notice.@1@ 2Template: Toter Link /  
  34. Carballal et al .: Loads and Requirements for Military Aircraft . In: NATO AGARD-R-815 . January 1997 ( [PDF]). Loads and Requirements for Military Aircraft ( Memento of the original from February 20, 2013 in the Internet Archive ) Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice.  @1@ 2Template: Webachiv / IABot /
  35. ^ A b c Jim Schefter: X-31 / How they're inventing a radical new way to fly . In: Popular Science . February 1989.
  36. X-31 Enhanced Fighter Maneuverability Demonstrator Photo Gallery Contact Sheet. In: NASA. May 23, 2000, accessed April 28, 2014 .
  37. a b c Controlled Flight Beyond Stall The X-31 Aircraft Program. (PDF) In: HAW Hamburg . May 2007, accessed April 28, 2014 .
  38. a b c d Tischler et al .: Advances in Aircraft Flight Control . Crc Pr Inc, 1996, ISBN 0-7484-0479-1 , pp. 321-343 .
  39. a b c d C.J. Loria et al .: X-31 Quasi-Tailless Evaluation . In: Aerospace Applications Conference . February 1996, p. 253-276 vol . 4 .
  40. a b c Steven C. Boehmer: X-31 Helmet Mounted Visual & Aural Display (HMVAD) system . In: Proc. SPIE 2218, Helmet- and Head-Mounted Displays and Symbology Design Requirements . June 1994.
  41. ^ A b Hans W. Pongratz: X-31 helmet-mounted display virtual adversary symbology development and simulation . In: Proc. SPIE 2465, Helmet- and Head-Mounted Displays and Symbology Design Requirements II . May 1995.
  42. X-31 VECTOR PROGRAM PHASE 1 BEGINS. In: NASA. March 9, 1998, accessed April 30, 2014 .
  43. Flight test targets set for US / German X-31. In: Flight International. May 2, 2000, accessed April 21, 2014 .
  44. Different approach. In: Flight International. October 23, 2001, accessed April 21, 2014 .
  45. a b c T.J. Lyons: Operational Need . In: Defense Technical Information Center . 2000 ( [PDF]). Operational Need ( Memento of the original from March 4, 2016 in the Internet Archive ) Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice.  @1@ 2Template: Webachiv / IABot /
  46. TJ Lyons & JY Grau: 2. “OPERATIONAL NEED” AND “SITUATIONAL AWARENESS” SURVEY . In: Human Consequences of Agile Aircraft, RTO TECHNICAL REPORT 15 . 2001 ( [PDF]). 2. “OPERATIONAL NEED” AND “SITUATIONAL AWARENESS” SURVEY ( Memento of the original from September 25, 2015 in the Internet Archive ) Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice.  @1@ 2Template: Webachiv / IABot /