The ground support required for an X-15 flight was quite extensive. At the launch lake, we had a rescue helicopter, a fire truck, and a NASA emergency vehicle with X-15 crew members aboard. At the intermediate lakebeds, we had a fire truck and a NASA emergency vehicle with X-15 crew members. At Edwards, we had the works. After landing at Edwards, the pilot was greeted by a convoy of vehicles that converged on him almost before he stopped. The Edwards fire chief led the pack in his bright red pickup. He was not officially the leader of the convoy, but he had the fastest vehicle. He would simply outrun everyone. There was no rigid convoy formation or speed limit.
Everyone was on his own once the mobile control van started moving. The fire chief routinely hit speeds approaching 100 MPH as he drove across the lakebed. There were at least ten or twenty vehicles plus a helicopter that roared across the lakebed to rescue the X-15 pilot whether he wanted to be rescued or not. Ironically, if we made it home to Edwards we usually did not have a problem. We normally needed the most help if we did not make it home to Edwards. It was, however, a spectacular show with all those vehicles racing across the lakebed raising a dust cloud that rose 1,000 feet into the air. Sitting in the X-15, it was somewhat frightening to see all these vehicles converge on us. We really hoped and prayed that their brakes worked properly. On routine flights, this was the most dangerous part.
All of this emergency ground support equipment was supplied by NASA and the Edwards Flight Test Center. This obviously involved a lot of logistics to get the support vehicles to their appropriate locations prior to each flight. The launch lake rescue helicopter normally flew up to the launch lake early on the morning of the scheduled flight. They normally had to be on station at 8:00 A.M., which required that they leave Edwards at four or five in the morning to get to a launch lake 200 to 300 miles from Edwards. The fire trucks and the NASA emergency vehicle were flown up to the launch lake in a C-130.
Edwards had two C-130s dedicated to X-15 support, and they were obviously critical to the operation. They were sometimes late accomplishing their support mission, but we did not often cancel a flight because of lack of support. The C-130s would deliver the launch lake vehicles and then fly back to Edwards to pick up another rescue vehicle and a paramedic team. They would then fly back up range to assume a position about halfway between the launch lake and Edwards. From that orbit position, they could react and cover an emergency landing at the intermediate lakebeds. If a call came over the air indicating an emergency landing at one of the intermediate lakebeds, the C-130 headed to the designated lakebed. The C-130, with its paramedics, would also cover any crash or ejection anywhere along the ground track.
This total ground support system was tied together through VHF and UHF radio communications, as well as single side band radios. The ground support vehicles and crews were completely dependent on radio communications to alert them to an emergency landing at their particular location because they normally could not see the X-15. They waited to hear the call that the X-15 was going to make it home or make an emergency landing. That decision was made in the first minute and a half after launch.
Everyone was happy when the call came that Edwards was the landing destination. When any other landing destination was called out, everyone immediately tensed up. An emergency landing in the X-15 was an honest-to-God emergency. The call for a landing at an intermediate lakebed was the call that the emergency lakebed crews dreaded. They knew they could have a potential disaster on their hands in the next few minutes. But of course that is the reason they were on station in the first place. That is when they really earned their pay.
The final phases of the flight involved an unpowered approach and landing. The unpowered approach ideally involved a 360 degree spiraling descent. The starting position was directly over the desired touchdown point at 35,000 feet altitude on the runway heading, at 250 to 300 knots airspeed. This position was referred to as high-key. From this position, we began a turn using a nominal thirty-five degree bank while maintaining 250 to 300 knots airspeed. The turn was usually made to the left, although it could be made in either direction. At the completion of 180 degrees of turn, we were approximately 4 miles abeam of the intended touchdown point at 18,000 to 20,000 feet altitude headed in the opposite direction of the landing runway. This position was referred to as the low-key position. The turn was continued from this position through another 180 degrees to line up with the landing runway about 5 miles short of the runway. At this point we were set up for the landing.
The X-15 simulator initially established the geometry of the approach pattern. The pattern was then verified in flight using other aircraft such as the F-104 configured to match the X-15 gliding performance. The rate of descent in the landing pattern averaged about 12,000 feet per minute which meant that the landing approach averaged about 3 minutes in duration.
The nominal approach described had many variations to compensate for off nominal initial energy conditions at the high-key position or starting point. If we were high on energy at high key, we could widen the pattern by using a shallow bank angle or conversely, we could tighten the pattern if we were low on initial energy. We could also adjust our energy in the pattern by using a higher or lower airspeed, or by using speed brakes to kill off excess energy. The approach pattern could also be abbreviated if we were very low on initial energy. We could use a 180-degree approach or a straight-in approach. If there were multiple runways, we could use 270-degree or 90-degree approaches. We normally planned all of our flights to have a lot of excess energy so that we always had enough to make the preferred 360-degree approach. We did on occasion have to make other approaches due to unanticipated problems.
The possible variations in the approach seemed limitless. Successful approaches were made starting from speeds as high as Mach 3 at high key and altitudes as high as 70,000 feet. A successful approach was also made from an altitude as low as 25,000 feet at high key. It is amazing how much versatility there is in an unpowered approach. We ultimately achieved consistent high-key energy conditions by initiating energy adjustments 50 miles before reaching high key.
Very precise pinpoint landings can be made in aircraft that are unpowered as demonstrated for many years in gliders and sailplanes. The X-15 and the other early rocket aircraft pilots utilized these same basic approach techniques to make over 500 successful landings with a high degree of accuracy. Overall landing accuracy in the X-15 was within 2,000 feet of the intended touchdown point. Not bad for a hypersonic glider. We did miss by as much as 4,000 to 5,000 feet on some emergency landings. Now I understand where that old statement, “missed by a mile,” came from.
The culmination of an X-15 flight was the unpowered landing. Unpowered landings were routine in the early rocket aircraft. Prior to the X-15 program, there were over 250 unpowered landings made in the X-1s, the X-2 and the D-558-II rocket aircraft. Unpowered landings were a way of life at Edwards. The X-15 just happened to be the latest of the series.
Even before the X-15 aircraft was constructed, it was obvious that a special unpowered landing technique had to be developed. Some preliminary inflight simulations of an X-15 using an F-104 to duplicate the subsonic lift/drag ratio revealed that the old power off landing techniques were not adequate. Landing flare initiation was hard to judge due to the steep approach angle and the high rate of descent. The ensuing landings were not consistently successful in the flight simulations and many wave offs were initiated prior to touchdown. The Lockheed test pilots were having similar problems demonstrating a flame out approach technique for the F-104. It also had a very low subsonic lift/drag ratio and came down like a streamlined brick without power. The Lockheed pilots crumped at least two airplanes trying to demonstrate power off landings.
A proposed landing technique was developed by a NASA Ames pilot, Fred Drinkwater. He developed the technique using an F-104 aircraft. He varied lift/drag ratio using flaps and speed brakes to investigate a range of low lift/drag ratio unpowered approaches. He simulated the X-15 with
a lift/drag ratio of 4 as well as simulating other more inefficient unpowered aircraft. He found that unpowered landings of very low lift/drag ratio (L/D as low as 2.5) vehicles could be made successfully using a high-speed approach and two different aim points short of the runway to assure touchdown accuracy.
Using Fred’s technique, the pilot literally dove at the first aim point at an angle that would result in a 300 knot stabilized airspeed. On reaching a pre-computed altitude, he would initiate a programmed flare maneuver and pick up the second aim point on a much shallower 3-degree glide slope. He maintained this glide angle until he descended through a height of 100 feet above the runway, at which time he made the final flare maneuver. This technique produced some very accurate touchdown locations but somewhat inconsistent touchdown sink rates, when the NASA pilots at Edwards evaluated it.
As a result of these evaluations, the Edwards pilots modified the technique to eliminate the second aim point and reduce the inner glide slope to a degree or less. This produced both accurate touchdown locations and very acceptable sink rates of less than 1 foot per second. The X-15 landing flare was initiated at about 1,000 feet above the ground at an airspeed of 300 knots. We attempted to come level 50 to 100 feet above the runway. We lowered the landing flaps as soon as we came level and then deployed the landing gear as we decelerated through 230 knots. During this float period, after the flare, we were adjusting height above the runway and rate of sink to touchdown at a low-sink rate as the airspeed decreased through 200 knots.
This modified technique became the accepted approach and landing technique for all the early pilots in the program except Crossfield. He developed his own technique using ground simulation and flight simulation in an F-100 aircraft. Scott had to use a drogue chute on the F-100, in addition to speed brakes to simulate the low lift/drag ratio of the X-15. Scott’s approach and landing technique utilized lower airspeeds, which made the landing flare maneuver more difficult to execute properly.
With the higher-speed approach, the pilot could make the landing flare maneuver and come level to the runway with a lot of excess airspeed. For example, by starting the flare at 300 knots, the pilot could complete the flare and come level with more than 260 knots of airspeed. After flare, the pilot had 60 to 70 knots of airspeed to lose before touching down at 190 to 200 knots. We were decelerating at about 4 knots per second, so we had between 15 and 18 seconds to adjust height and rate of sink for a smooth landing. This does not seem like much time but it was completely adequate. The sink rates at touchdown using the high-speed approach averaged less than 2 feet per second. Crossfield’s technique provided very little float time after flare and his average sink rate was substantially higher using his lower speed approach. He later began using the high-speed approach and his sink rates decreased significantly.
The landing and slideout completed an X-15 flight operation. The aircraft slid 1.5 to 2 miles before it finally came to a stop. As the aircraft slid to a stop, the recovery convoy converged on it to assist in getting the pilot out and in deactivating the aircraft. Residual propellants and peroxide were jettisoned on the lakebed and the aircraft was then towed on a dolly, back to the hangar to be inspected prior to starting over a whole new flight operation.
One of the more risky jobs during an X-15 operation was to set out the landing smoke flares. These were used to provide information for the X-15 pilot on the direction and relative velocity of the surface wind on the lakebed runway. The smoke flares spewed colored smoke for approximately 1 minute after they were ignited.
Some flares were normally ignited at least 1 minute before landing to enable the pilot to set up his approach and landing to compensate for the wind. Additional flares were ignited within a half minute of landing to ensure that smoke would still be visible during final touchdown. These flares were positioned on the edge of the runway about halfway down the lakebed runway. The critical nature of this task was to position and ignite the flares at the proper time and then get the hell out of the way since the X-15 could not be accurately steered on the ground, particularly in a crosswind, and thus, we were not certain where it was going to end up. Nothing is more frightening than to see the X-15 bearing down on you at over 200 MPH throwing up a 200-feet-high rooster tail of dust.
The usual procedure was to drive a carryall vehicle out to the proper location about 10 minutes prior to landing and then begin tossing flares out 1 minute or so prior to landing. You tossed the last flare at about 20 seconds to touchdown and then headed out at full throttle perpendicular to the landing runway. This usually provided reasonable clearance. Every once in a while though, someone would panic and stall the engine while attempting to accelerate away from the runway. That could result in some world-class record hundred-yard dashes by the vehicle occupants.
In summary, the unique features of an X-15 rocket aircraft operation were the air launch several hundred miles from Edwards, the tremendous acceleration during the short engine burn time, the unpowered hypersonic glide to the landing site, and the deadstick landing. The flights seldom exceeded 10 minutes in duration from launch to landing. This meant that on a pilot’s first flight, he had to learn to fly the X-15 sufficiently well enough in 10 minutes to make a deadstick landing on his first try. There was no go-around capability. It had to be done right the first time or no cigar.
In preparation for a flight, we flew the X-15 ground simulator and we flew F-104s. The ground based simulator was used to practice the major portion of the flight, beginning with launch and proceeding through boost, hypersonic flight, and energy management maneuvering to arrive at the landing site. The F-104s were used for in-flight simulation of the landing approach.
The ground simulator was a replica of the X-15 cockpit with all of the instruments, gages, switches, and controls. This cockpit was connected to a computer that translated the pilot’s control actions into meaningful instrument readings, simulating what the aircraft would do in-flight in response to these control inputs. Flying on instruments in the simulator was essentially doing just as you might do if you were flying in the clouds. We used this simulator to practice each flight. But before the pilot could practice a flight, someone had to plan it or lay it out in minute detail. This was the job of the flight planner.
The research engineer would first request certain maneuvers at a specific flight condition of speed and altitude. He might want a maneuver to measure the stability and control characteristics, or he might ask for a maneuver to apply g loads to the airplane in order to measure the stresses induced in the structure of the airplane. The flight planner would take all of these requirements and lay out a flight plan to obtain these maneuvers at the desired flight condition. This flight plan would define the distance in miles required to reach the desired flight conditions for each maneuver and to obtain the desired data. Based on this distance, the flight planner would select a suitable launch lake from those which we had available. The flight planner would then select the intermediate lakes to be used for emergency landings, and then begin finalizing the flight plan in great detail.
Flight planners spent their entire working day in the simulator. In fact, they spent their entire life in the simulator. We would occasionally find a flight planner who had died a natural death in the simulator. Because they lived in the simulator, they were the experts on the airplane. They knew all of its innermost secrets. They would determine what thrust level to set the engine at after launch. They determined the climb angle, the pushover time, the throttle back time, the engine shutdown time. They would define the maneuvers to be accomplished after engine burnout and the maneuvers required to decelerate into the approach pattern at Edwards. They would also define the cutoff times or decision points for the various emergency lakebeds.
Once they had defined all of this, they took the pilot under their wing and taught him to fly the desired flight plan. They worked with the pilot for days and weeks practicing for a particular flight. The pilot would fly the flight over and over again getting the timing down to t
he second. During these training sessions, the pilots memorized the flight conditions second by second. At any given second during the boost phase, the pilot knew what the altitude, velocity, and rate of climb should be. He also knew what the g level should be, the angle of attack, the pitch attitude, the heading, the roll attitude, the dynamic pressure, and so on.
Once the pilot knew the basic flight plan by heart, the flight planner would simulate emergencies for the pilot. Hours and hours were spent practicing emergency responses. Typically, the pilot spent a minimum of 15 to 20 hours on the simulator practicing for an 8 to 10 minute flight. A pilot could conservatively practice each flight six to ten times an hour, so over a 20-hour practice session, he could fly the mission 120 to 200 times. The pilot was almost jaded by the time he finally got in the airplane, but that training saved his butt on many occasions. He could fly the flight completely on his own if, for example, he lost radio communications. In fact, we did lose radio communications quite often. Mike Adams added something new to the simulation routine. He would occasionally get bored and start practicing the mission while flying upside down.
The simulator was continually updated with data obtained on previous flights to ensure the validity of the simulator. The original simulator was created using predicted data obtained in the wind tunnel. As the flight program progressed, various maneuvers were performed to obtain aerodynamic derivatives. These flight determined derivatives may or may not have agreed with the original predictions obtained from wind tunnel tests. If the wind tunnel and flight data disagreed, we would attempt to determine which data was correct by repeating the flight maneuvers and in some cases asking for a rerun of the wind tunnel tests. In many cases, the pilot influenced the final decision by comparing the simulator responses, using both old and new sets of data, with the actual aircraft responses. The data that created the best match between the simulator responses and the actual aircraft response was normally selected to update the simulator and ensure its continuing validity. By continually updating the simulator, we were able to more accurately predict potential problems as we continued to expand the flight envelope.
At the Edge of Space Page 9
