The Northrop X-4 "Bantam"

A tailless transonic experiment:

Tucker, Charles

Click on Picture to enlarge

In the fall of 1947, I received a phone call that was one of those calls every test pilot dreams about: Northrop was looking for a pilot for its new X-4 Bantam, and was I interested? It isn't often that a test pilot gets to make the first flight of a new design, especially one as radical as the X-4, so I couldn't believe my good fortune when Northrop gave me the nod.

At the time, I had only been flying for five years, but it had been a busy five years. I graduated from the Army Air Corps flying school in January 1942 as a "pursuit" pilot. With its typical logic, the Army had me flying a C-47 from West Palm Beach, Florida, to China via South America, Africa, India and over the "Hump" to China. I made seven cargo flights over the Hump in the C-47 before transferring to Gen. Chennault's 14th Air Force in China in August 1942. I flew 75 combat missions in P-40s and scored four victories before ground fire hit my cooling system during a strafing run in July 1943, and I was forced to bail out. After a week of trekking, I made it home and then became ill from the bad food and water I had while evading the Japanese, and I was evacuated to the U.S. for treatment in October 1943.

After recovering, I was assigned to the 412th Fighter Group. This unit was equipped with the Bell P-59 Airacomet, P-38s, P-63s, P-51s, the Lockheed XP-80 and various other aircraft. Later, the Air Corps ran accelerated service tests on the P-80 at Muroc Army Airfield. Wright Field pilots were flying these airplanes, and I was invited to join this elite group in those tests.

After the War, I went to work for Lockheed as a civilian production test pilot on P-80s, and that lasted until September 1947. I wound up with quite a number of hours on that aircraft. It was then that Northrop hired me to fly the X-4, as I had quite a bit of jet experience and was rather small of stature; my size was important because the X-4 cockpit was small. The X-4 hadn't yet been completed, so they started me flying the XB-35 and YB-49 Flying Wing. By the time the X-4 was ready to fly, I had already completed the stall program on the YB-49 and had spun the airplane.

The X-4 was designed to explore the flight characteristics of tailless aircraft in the transonic speed range. After WW II, all airframe manufacturers were aware of Germany's efforts with jet propulsion and flying-wing aircraft. As Northrop was a flying-wing specialist, the government selected it to build an airplane with no horizontal tail. As far as I know, no German designs influenced Northrop's design. After the initial problems had been sorted out, the aircraft performed especially well, and was fun to fly; it gave Northrop, NACA and the Air Force much-needed data about tailless-aircraft performance in high-speed flight.

The aircraft (two were being built) were not finished when I arrived at Hawthorne. The first time I saw them, the beauty of the little birds stunned me; I could hardly wait to get my hands on one. I watched their assembly and closely studied the control systems as they were installed. Northrop used the hydraulic actuator system from the YB-49 bomber for the elevons of the X-4. Using the control system from a heavy bomber in an airplane as small as the X-4 resulted in high stick friction and an excessive breakout force of approximately 12 pounds. These problems were primarily caused by the 3/16-inch-size bomber control cables and the 400 pounds of high cable tension in conjunction with the small, control-cable pulleys and the resultant acute-cable radii required in the small airframe of the X-4. These problems would later cause me much grief, as the friction in the system made it impossible to make small control inputs, thus making the airplane almost uncontrollable.

The rudder was electrically controlled. The rudder's engineering team had been thinking about high-speed flight, and they designed the actuating motor with a very slow response rate of approximately 25 degrees per second. While this is fine for high-speed flight, it is not at all desirable for low-speed flight such as during takeoff and landing. Once again, engineers were thinking like engineers, and it worried me; I wished more engineers were pilots.

The first X-4, No. 6676, was transported to Muroc in November 1948. The slow rudder response made itself felt immediately by making the airplane very difficult to control in the yaw axis. On the ground, this translated into an inability to track straight on the runway. I argued repeatedly that this rudder system was inadequate, but the engineering team thought I was just being "sour grapes" and should be able to handle anything they could build; after all, I was a test pilot, wasn't I? As the engineers refused to alter the rudder control system, I had a line painted on the lakebed for use as a yaw reference indicator. I felt this "centerline" would be beneficial in tracking the yaw movements because there was little else for visual reference on the lakebed.

I continued to make taxi runs on the lakebed, and over time I increased the speed in small increments. The slow rudder response was almost intolerable. The aircraft would yaw right, so I would apply left rudder and wait. As soon as the aircraft began to go left, I would apply right rudder and wait. This induced an oscillation that was difficult to damp. Still, the engineers thought it wasn't a serious problem. Quite a number of those guys were young and relatively inexperienced; some of them had their own agendas, and none were pilots. I continued making taxi tests until I had learned enough about controlling the slow response time that the engineers thought I should try a takeoff. I was certain that the aircraft wasn't ready for that phase of the program, but the X-4 project was way behind schedule, and the company was losing money on it daily. I was furious about the rudder, but I also wanted to get this first flight off and running. As a conscientious employee, with my employer's well-being foremost in my mind, I consented to try a takeoff.

On December 15, 1948, 1 had made three takeoff attempts, and all were aborted because of the yaw/controllability problem. Although I wasn't comfortable with the airplane, I decided to make one more attempt. I lined the little airplane up on the centerline and opened the throttles. The Westinghouse J30 engines spooled up, and the airplane accelerated, but this time, I kept the little beast going straight by judicious brake use. It didn't take much to keep it going down the centerline. Eureka! I was so fixated on keeping the airplane straight that I was completely unprepared for what happened next.

While I had been concentrating on keeping it straight, the X-4 had been accelerating like a rocket and was now well past the calculated liftoff speed. I pulled back on the stick, and the airplane leapt off the ground and began to climb at a very high angle. I pushed the stick forward, and the nose immediately dropped almost straight down. The airplane had been so fast when I rotated that it had literally shot straight up. If it hadn't done that, I wouldn't have had enough altitude to recover from the dive that followed and would have dug a very deep hole in the desert floor.

Now I was in trouble-airborne in a completely unstable airplane. Any attempt to change pitch, even the tiniest stick movement, resulted in an immediate, almost violent vertical climb or dive. Jamming the stick between my hand and knees, I managed to finally turn the flight recorders on. I tried moving the stick very gently to see what I could get away with. Not much! The high breakout force and high stick friction made the very light and gentle control inputs I needed almost impossible. I really had my work cut out for me. Literally hanging on for dear life, I thought of bailing out as I struggled with the airplane, but I felt that as long as I was still in the air, I had a chance to save it. Relaxing my grip on the stick brought an immediate reaction-up, or down, depending on my slightest movement. The X-4 was so sensitive that any thought of raising the landing gear was out of the question. I was lucky that the air was smooth.

After I discovered that I could keep the airplane straight and level only by holding the stick in both hands and clamping my hands between my knees, I made it a point not to touch the rudder. I let the airplane climb so that, if necessary, I would be able to eject, but I really didn't want to do that. It took me roughly 30 minutes at between 200 and 250mph to make a 180-degree turn back to the runway.

Nearing the runway, I decided that I could chance taking one hand off the stick momentarily to make a small power reduction for descent. I descended a bit, took one hand off the stick to reduce power, descended a bit, reduced power, again and again, until the airplane just flew onto the lakebed. At that point, being on the ground felt very good! I muttered lots of unprintable things about engineers as I taxied back to the waiting group. To put it mildly, I was very upset that they had sent me out in such a poorly prepared airplane, and I was extremely vocal about it. I also got pretty drunk that night!

As a result of that flight, a number of changes were made to the airplane. The control cables were reduced to 1/16 inch, and the cable tension was reduced to approximately 150 pounds. This lightened the breakout force to a more normal 1 to 1 1/2 pounds, and it also reduced the stick friction to normal levels. The method of rudder actuation was changed. The electrically operated system was removed, and rudder control was changed to a conventional cable-operated system with 1/16-- inch cables. This cured the low-- speed yaw control problem completely.

Also, the center of gravity was not correct, and that is what caused the divergent pitch oscillations. It was changed from 22 percent of the mean aerodynamic chord to 19.7 percent. Fitting 1/4-inch-thick lead sheets inside the nose skin accomplished this nicely. The airplane was so small and so crammed with equipment that this was the only place to add weight. These changes completely altered the plane's flight characteristics and made it a delight to fly. From this point on, not too much was structurally changed on the airplane.

Having the proper center of gravity, or CG, is the most important aspect of airplane control and stability. The CG was off because when the initial calculations were made, they forgot to take into account the effect of the fuselage that housed the fuel tank-and me. The engines and fuel system of No. 6676 were always problematic. As a matter of fact, this airplane was retired after 10 flights; there was just too much wrong with it. By this time, the second airplane, No. 6677, was ready to fly. It was a much better airplane than 6676, and the rest of the X-4 program was flown with it; No. 6676 became a source of spare parts.

Test flights in experimental research aircraft are very rigidly controlled. There is a mission to perform, and no time is allocated for anything else; any unauthorized maneuvers are officially frowned on. When the mission requirements have been met, however, and you are on your way back to the base, you have a bit of latitude. Two examples come to mind: the split flaps/speed brakes on the X-4 were huge and very effective. I returned from test flights at between 35,000 to 40,000 feet. Right over the end of the runway at Muroc, I opened the flaps/speed brakes and pointed the airplane straight down. In this configuration, with full power, the airplane never exceeded 350mph. Then I called the tower to tell them that I was over the end of the runway for landing. They called back and asked me to confirm my position because they couldn't see me. I told them I didn't understand that, as I was right over the end of the runway. They then said, "OK, you are cleared to land." I got away with this several times, until one day, one of the tower people happened to look way up. The game was up, and we all laughed about it.

Another time, Col. Al Boyd, the Muroc base commander, was flying chase for me in an F-86. (Chase pilots were instituted after the YB-49 crash earlier in 1948. I can't explain why they waited until then to institute the chase-plane program. Perhaps manufacturers were protective of their newest technology; perhaps they were concerned about the cost of the program. And there was probably the question of who would fly the chase planes.) We had finished the scheduled test flights and were on the way home at about 40,000 feet over the western edge of the Mojave Desert. I could see the big hangars at the base in the distance. With nothing on my mind except getting home as quickly as possible, I pointed the nose of the X-4 at the hangars. This gave me about a 20-degree dive angle from my altitude, so I retarded the throttles somewhat, as I knew I would pick up quite a bit of speed. The airplane accelerated, and the anticipated Mach buffet began at about 35,000 feet. As an experiment, I triggered the stick-mounted switch a bit to open the flaps/speed brakes at what I calculated to be about 1 to 2 inches. By the time I got to 30,000 feet, the buffet was gone and the airplane accelerated smoothly. When we got back on the ground, Al told me that he had given the F-86 all the power it had, and his airplane was shaking and complaining, but he couldn't catch me. I figured I must have been going pretty fast, but the Mach meter in the X-4 only went to Mach 1.0, so I had no idea how fast I was actually going. A couple of weeks later, Walt Williams, head of the NACA group at Muroc, called me and told me that they had finished examining the telemetry data from that flight, and he wanted to congratulate me on going Mach 1.02 in the X-4. I felt pretty good about that, although breaking Mach 1.0 hadn't been on my mind at the time; I had just wanted to get home. No one else ever came close to that speed in the X-4.

The X-4 was the first X-Series airplane equipped with an ejection seat. The canopy and windshield were one-piece blown Plexiglas with no fixed windscreen. That assembly was hinged at the upper rear, and in case of ejection, would be blown off the aircraft. This canopy bothered me because if I was ever forced to eject at a high speed, I was concerned about what would happen to me in all that wind; so I designed a full-face shield helmet. This was the first full-face shield helmet ever designed, and I was subsequently awarded a U.S. patent for it; it is the precursor of today's standard military flight helmet. In my design, the faceplate opened from side to side, instead of up and down as in today's helmets.

As mentioned at the beginning of this article, Northrop, NACA and the Air Force gained much useful data on transonic flight from the X-4. The airplane was built to investigate whether or not a tailless configuration was suitable for supersonic aircraft. When this investigation had been completed, based on its experience in flying the airplane, NACA deemed the tailless configuration unsuitable. The Navy actually built a fighter without a horizontal tail, but it was not successful. When you pull G in a tailless airplane, it wants to pitch up, which is not good. Tailless is OK for straight and level flight but not for maneuvering.

Of course, the X-4 had no stability augmentation system; those had just begun to be designed and installed in some aircraft for testing. These systems were looked on as "black magic" by the Air Force, which prohibited them for any aircraft design submitted for service with that branch. Air Force bureaucracy had a lot of inertia, and to change its thinking was difficult. At that time, it thought that an artificial system could fail and then leave the pilot in a very bad situation, usually at a most inopportune time. Air Force thinking during the late '40s and '50s was that an aircraft must be inherently stable to be suitable for Air Force consideration. When I flew the YB-49, the directional damping was below par but with practice, I could hold the airplane on a heading. Northrop put the first stability augmenter on that airplane with great success, but it was not accepted by the Air Force with any enthusiasm.

Today, we fly high-speed aircraft that are so unstable that they are uncontrollable without sophisticated stability augmentation systems. What makes today's fighter aircraft so maneuverable is this very instability, which is kept in check by those stability augmentation systems. A human pilot can't move fast enough to supply all the control inputs needed to fly the airplane. The F-117 Stealth Fighter is perhaps the epitome of this instability. I doubt that a human pilot could get this airplane off the ground without the stability augmentation that is incorporated into the airplane's design.

The flight I made with partially opened flaps/speed brakes that enabled the X-4 to exceed Mach 1.0 opened the investigation into the usefulness of blunt trailing edges. The X-4, after it had been turned over to NACA and the Air Force, was flown with the flaps/speed brakes partially blocked open. They also altered the elevons to a blunt trailing-edge configuration. These experiments proved that the blunt trailing edge did delay the onset of Mach buffeting. This knowledge was put to good use with the X-15 research aircraft; take a good look at the tail surfaces of the X-15 in photographs some time.

How did the X-4 fly once all the initial bugs had been corrected? Beautifully! It was highly responsive to control inputs and actually very easy to fly-just like a little fighter. Of the two X-4s built, the first one (No. 6676) made 10 flights and the second one (No. 6677) made 102 flights between 1948 and 1953.

Today, No. 6676 is on display at the Air Force Academy at Colorado Springs, Colorado. Number 6677 is at the Air Force Museum at Wright Patterson Air Force Base in Dayton, Ohio.

The first five X planes (X-1, X-2, X-3, X-4 and X-5) were instrumental in opening the doors of high-speed flight. Each of those airplanes was designed to test a specific aspect of high-speed flight, and all performed admirably. Today's aviation owes a lot to these pioneering aircraft. I am glad to have been a part of it.

Air Age Feb 2002

Type	Tailless aircraft prototype
Manufacturer	Northrop
Maiden flight	15 December 1948
Number built	2

Click on Picture to enlarge

X-4 Bantam

The Northrop X-4 Bantam was a small twin-jet airplane that had no horizontal tail surfaces, depending instead on combined elevator and aileron control surfaces (called elevons) for control in pitch and roll attitudes. The hope of some aerodynamicists was that eliminating the horizontal tail would also do away with stability problems at transonic speeds resulting from the interaction of supersonic shock waves from the wings and the horizontal stabilizers. This hope was unrealized.

Development

Click on Picture to enlarge

Northrop N-9M

Northrop XB-35

Northrop YB-49

Two X-4s were built by the Northrop Corporation, but the first was found to be mechanically unsound and after 10 flights it was grounded and used to provide parts for the second.

While being tested from 1950 to 1953 at the NACA High-Speed Flight Research Station (now Edwards Air Force Base), the X-4's semi-tailless configuration exhibited inherent longitudinal stability problems (porpoising) as it approached the speed of sound. It was concluded that (with the control technology available at the time) tailless craft were not suited for transonic flight.

Me-163 Komet

de Havilland DH.108 Swallow

It was believed in the 1940s that a design without horizontal stabilizers would avoid the interaction of shock waves between the wing and stabilizers. These were believed to be the source of the stability problems at transonic speeds up to Mach 0.9. Two aircraft had already been built using a semi-tailless design—the rocket-powered Me-163 Komet flown by Germany in World War II, and the British de Havilland DH.108 Swallow built after the war. The United States Army Air Forces signed a contract with the Northrop Aircraft Company on June 11, 1946, to build two X-4s. Northrop was selected because of its experience with flying wing designs, such as the N-9M, XB-35 and YB-49 aircraft.

The resulting aircraft was very compact, only large enough to hold two Westinghouse J30 jet engines, a pilot, instrumentation, and a 45-minute fuel supply. Nearly all maintenance work on the aircraft could be done without using a ladder or footstool. A person standing on the ground could easily look into the cockpit. The aircraft also had split flaps, which doubled as speed brakes.

Operational History

Click on Picture to enlarge

Preparing for flight

The first X-4 (serial number 46-676) was delivered to Muroc Air Force Base, California, in November 1948. It underwent taxi tests and made its first flight on December 15, 1948, with Northrop test pilot Charles Tucker at the controls. Winter rains flooded Rogers Dry Lake soon after, preventing additional X-4 flights until April 1949. The first X-4 proved mechanically unreliable, and made only 10 flights. Walt Williams, the head of the NACA Muroc Flight Test Unit (now Dryden Flight Research Center) called the aircraft a "lemon". The second X-4 (serial number 46-677) was delivered during the halt of flights, and soon proved far more reliable. It made a total of 20 contractor flights. Despite this, the contractor flight program dragged on until February 1950, before both aircraft were turned over to the Air Force and the NACA. The first X-4 never flew again, used as spare parts for the second aircraft.

The NACA instrumented the second X-4 to conduct a short series of flights with Air Force pilots. These included Chuck Yeager, Pete Everest, Al Boyd, Richard Johnson, Fred Ascani, Arthur Murray and Jack Ridley. The flights were made in August and September of 1950. The first flight by a NACA pilot was made by John Griffith on September 28, 1950.

The initial NACA X-4 flights, which continued from late 1950 through May of 1951, focused on the aircraft's sensitivity in pitch. NACA pilots Griffith and Scott Crossfield noted that as the X-4's speed approached Mach 0.88, it began a pitch oscillation of increasing severity, which was likened to driving on a washboard road. Increasing speeds also caused a tucking phenomena, in which the nose pitched down, a phenomenon also experienced by the Me 163A Anton prototypes in 1941. More seriously, the aircraft also showed a tendency to "hunt" about all three axes. This combined yaw, pitch and roll, which grew more severe as the speed increased, was a precursor to the inertial coupling which would become a major challenge in the years to come.

To correct the poor stability, project engineers decided to increase the flap/speed brake trailing edge thickness. Balsa wood strips were added between the flap/speed brake halves, causing them to remain open at a 5 degree angle. The first test of the blunt trailing edge was flown on August 20, 1951, by NACA pilot Walter Jones. A second test was made by Crossfield in October. The results were positive, with Jones commenting that the X-4's flight qualities had been greatly improved, and the aircraft did not have pitch control problems up to a speed of Mach 0.92.

The balsa strips were removed, and the X-4 then undertook a long series of flights to test landing characteristics. By opening the speed brakes, the lift-to-drag ratio of the aircraft could be reduced to less than 3 to 1. This was for data on future rocket powered aircraft. The tests continued through October 1951, until wing tank fuel leaks forced the aircraft to be grounded until March 1952, when the landing tests resumed. NACA pilots Joe Walker, Stanley Butchard, and George Cooper were also checked out in the aircraft.

The thickened flap/speed brake tests had been encouraging, so balsa wood strips were reinstalled on both the flap/speed brake and the elevons. The first flight was made by Jones on May 19, 1952, but one of the engines was damaged during the flight, and it was August before a replacement J30 could be found. When the flights resumed, they showed that the modifications had improved stability in both pitch and yaw, and delayed the nosedown trim changes from Mach 0.74 to Mach 0.91. Above Mach 0.91, however, the X-4 still oscillated.

Click on Picture to enlarge

F7U Cutlass

In May of 1953, the balsa wood strips were again removed, and the X-4's dynamic stability was studied in the original flap/speed brake and elevon configuration. These flights were made by Crossfield and John B. McKay. This was the final project for the X-4, which made its 81st and final NACA flight on September 29, 1953. Both aircraft survived the test program. The first X-4 was transferred to the United States Air Force Academy, Colorado Springs, Colorado, before being returned to Edwards Air Force Base. The second X-4 went to the National Museum of the United States Air Force at Wright-Patterson Air Force Base near Dayton, Ohio, where it remains on display.

Click on Picture to enlarge

Boeing Bird of Prey

The X-4's primary importance involved proving a negative, in that a swept-wing semi-tailless design was not suitable for speeds near Mach 1, although the F7U Cutlass proved to be something of a counterexample—the developed version was the first aircraft to demonstrate stores separation above Mach 1. Aircraft designers were thus able to avoid this dead end. It was not until the development of computer fly-by-wire systems that such designs could be practical. Semi-tailless designs appeared on the X-36, Have Blue, F-117, and Bird of Prey, although these aircraft all differed significantly in shape from the X-4. The trend during its test program was already toward delta and modified delta aircraft such as the Douglas F4D, the Convair F-102A derived from the XF-92A, and the Avro Vulcan.

X-4 Specifications

Click on Picture to enlarge

General characteristics

Crew: 1
Length: 22 ft 3 in (7.1 m)
Wingspan: 26 ft 10 in (8.2 m)
Height: 14 ft 10 in (4.5 m)
Wing area: ft² (m²)
Empty weight: 5,600 lb (2,540 kg)
Max takeoff weight: 7,820 lb (3,550 kg)
Powerplant: 2× Westinghouse J30 turbojet, 1,600 lbf (7.1 kN) each

Performance

Maximum speed: 640 mph (1,035 km/h)
Range: miles (km)
Service ceiling: 44,000 ft (13,400 m)
Rate of climb: ft/min (m/s)
Wing loading: lb/ft² (kg/m²)

The X-4 was designed to test a semi-tailless wing configuration at transonic speeds. Many engineers believed in the 1940s that the such a design, without horizontal stabilizers, would avoid the interaction of shock waves between the wing and stabilizers. These were believed to be the source of the stability problems at transonic speeds up to Mach 0.9.

Two aircraft had already been built using a semi-wingless design - the rocket-powered Me-163 Komet flown by Germany in World War II, and the British de Havilland DH.108 Swallow build after the war. The Army Air Forces signed a contract with the Northrop Aircraft Company on June 11, 1946, to build two X-4s. Northrop was selected because of its experience with flying wing designs, such as the N9M, XB-35 and YB-49 aircraft.

The resulting aircraft was very compact, only large enough to hold two J30 jet engines, a pilot, instrumentation, and a 45-minute fuel supply. Nearly all maintenance work on the aircraft could be done without using a ladder or footstool. A person standing on the ground could easily look into the cockpit. For control without horizontal tail surfaces, the X-4 used combined elevator and aileron control surfaces (called elevons) for control in pitch and roll attitudes. The aircraft also had split flaps, which doubled as speed brakes.

The first X-4 (serial number 46-676) was delivered to Muroc Air Force Base, Calif., in November 1948. It underwent taxi tests, and made its first flight on December 15, 1948, with Northrop test pilot Charles Tucker at the controls. Winter rains flooded Rogers Dry Lake soon after, preventing additional X-4 flights until April 1949. The first X-4 proved mechanically unreliable, and made only 10 flights. Walt Williams, the head of the NACA Muroc Flight Test Unit (as Dryden was then known) called the aircraft a "lemon." The second X-4 (serial number 46-677) was delivered during the halt of flights, and soon proved far more reliable. It made a total of 20 contractor flights. Despite this, the contractor flight program dragged on until February 1950, before both aircraft were turned over to the Air Force and the NACA. The first X-4 never flew again, serving as a spare parts bin for the second aircraft.

The initial NACA X-4 flights, which continued from late 1950 through May of 1951, focused on the aircraft's sensitivity in pitch. NACA pilots Griffith and Scott Crossfield noted that as the X-4's speed approached Mach 0.88, it began a pitch oscillation of increasing severity, which was likened to driving on a washboard road. Increasing speeds also caused a tucking phenomena, in which the nose pitched down. More seriously, the aircraft also showed a tendency to "hunt" about all three axes. This combined yaw, pitch and roll, which grew more severe as the speed increased, was a precursor to the inertial coupling which would become a major challenge in the years to come.

In May of 1953, the balsa wood strips were again removed, and the X-4's dynamic stability was studied in the original flap/speed brake and elevon configuration. These flights were made by Crossfield and John McKay. This was the final project for the X-4, which made its 81st and final NACA flight on September 29, 1953. Both aircraft survived the test program. The first X-4 was transferred to the Air Force Academy, Colorado Springs, Colo., before being returned to Edwards Air Force Base. The second X-4 went to the Air Force Museum, Wright-Patterson air Force Base, Ohio.

The X-4's primary importance involved proving a negative, in that a swept-wing semi-tailless design was not suitable for speeds near Mach 1. Aircraft designers were thus able to avoid this dead end. It was not until the development of computer fly-by-wire systems that such designs could be practical. Semi-tailless designs appeared on the X-36, Have Blue, F-117A, and Bird of Prey, although these aircraft all differed significantly in shape from the X-4.