PREFACE:  To better understand how an X-15 flight was conducted in the 1960′s the following video is suggested.

Story from the book “Flight Testing at Edwards”

by Johnny Armstrong

On October 3 1967 the X-15A-2 research aircraft achieved a maximum Mach number of 6.72 piloted by Major Pete Knight. The events that led up to the flight really began five years earlier when, on its 31st flight, the number two X-15, S/N 66671, suffered major damage during an emergency landing at Mud Lake, Nevada.

X-15 #2 Crash Landing at Mud Lake

Ten months prior to that emergency landing I had returned to Edwards as a civilian employee for AFFTC. Following my four years at Edwards in flight test engineering wearing the AF blue, I had accepted a job with NASA at Huntsville Alabama during which time I had worked on flight test of the Saturn rocket vehicle. Then near the end of 1961 I was contacted to see if I would be interested in returning to Edwards to work on the X-15. That was an easy yes. I had been missing airplanes so much that I would run out of the house to see the Southern Airways DC-3 fly over. So here I was back at Edwards working as the AFFTC flight planner on one of the most exciting programs of all times.

The ship 2 X-15 had entered the three-ship X-15 program in September 1959. In 1961 it had flown nine flights reaching a maximum Mach number of 6.04 and a max altitude of 217,000 feet.

X-15 crash landing at Mud Lake

NASA pilot Jack McKay had just launched from the B-52 mothership at 45,000 feet, lit the YLR-99 rocket engine and was on his way when Pete Knight, who was the flight communicator in the NASA control room, radioed for him to check his throttle position. Jack verified it was at the full throttle position and Pete advised him that the engine was only putting out 30 percent thrust.

This kind of situation is exactly why all X-15 flights were launched within gliding distance of a suitable dry lakebed. When Pete radioed that it was going to be a Mud Lake landing, Jack began a preplanned series of actions to make an emergency landing. He shutdown the engine and set up the turning pattern to Mud Lake, jettisoning the anhydrous ammonia and liquid oxygen propellants along the way. Due to several of Mr Murphy’s law things adding together, the strut of the left main landing gear failed on touchdown resulting in the aircraft turning sideways and then rolling upside down trapping Jack in his seat with his head next to the lakebed surface. Prior to rolling over Jack had had the forethought to jettison the canopy. The emergency crew that was pre-stationed at the site immediately moved into action. The Air Force helicopter pilot had the good judgment to hover over the X-15 blowing away the pungent fumes of the anhydrous ammonia still venting from the aircraft. The ground crew was able to dig a hole in the lakebed to provide enough clearance to extract Jack from the cockpit and get him on the way for medical attention. Jack would live to fly the X-15 again but he was an inch shorter due to crushed vertebra.

Six months after the Mud Lake landing accident a contract was signed with North American to repair and modify the aircraft to an advanced configuration to accomplish future experiments. The modified aircraft was returned to Edwards in February 1964 and made its first flight in June of that year. The main modification to the aircraft was the addition of two jettisonable external tanks to hold added engine propellant to provide a Mach 8 capability. This additional performance was desired for testing of a hypersonic ramjet engine then under development by NASA Langley. The airplane had also been lengthened by 29 inches to add room for a tank to provide hydrogen fuel for the ramjet engine. Other modifications included additional hydrogen peroxide tanks for propellant turbopumps for the longer engine burn time of about an additional 60 seconds, an openable hatch behind the cockpit for a star tracker experiment, and a modified canopy windshield. Additionally, an ablative coating was developed to be applied to the aircraft to protect the structure from the higher heating that would be encountered during the higher Mach number flights.


The envelope expansion effort began with Flight 43 when Major Rushworth flew the first flight with external tanks. Consistent with the incremental approach of an envelope expansion program this flight was flown with the external tanks empty. Now there was a new constraint for the X-15 flight. The launch point not only had to be the necessary distance from Edwards depending on the max Mach to be flown, and also within gliding distance of a prepared dry lakebed runway, but also had to have a suitable area for the tanks to impact upon being jettisoned after they were empty. For this flight the X-15 was launched near Cuddeback dry lake about 60 miles North of Edwards on a heading that would result in the external tanks impacting on the Edwards bombing range. The flight was flown at 50 percent thrust to slow the acceleration. The flight went about as planned with the tanks being jettisoned at Mach 2.2 and 70,000 feet. The tanks impacted on the bombing range. The recovery chute system work properly for the ammonia tank and it was recovered in repairable condition. The LOX tank recovery chute did not deploy and the tank was destroyed on impact. With this flight the handling qualities of the X-15 with external tanks were verified to be adequate although somewhat worse than the simulator. It also validated the tank ejection and separation.

We See No Flow

Now we were ready to fly with full external tanks to Mach 6.0. It was to be the 45th flight of ship 2. It was also going to be last flight for Major Rushworth who had received his orders for his next assignment. (Rushworth flew more flights in the X-15 than any other pilot; a total of 34.) In retrospect, with twenty-twenty hindsight, flight 45 was destined for failure. The propellants were pumped to the engine from the internal tanks of the X-15 and the propellants in the external tanks were just pressure transferred into the internal tanks. From simulation studies we knew that the pilot would lose control of the X-15 rather rapidly if one of the tanks failed to transfer propellant. This was simply due to the moment that would be produced about the roll axis that could not be counteracted by the X-15 roll control that was produced by the differential deflection of the horizontal tails. We developed procedures in the simulator to be followed if one of the tanks failed to feed after launch. It essentially consisted of immediately setting up at conditions of angle of attack and dynamic pressure where tanks separation was predicted to be good. This consisted of first shutting down the engine, maintaining angle of attack to let the dynamic pressure decrease below 400 psf and then pushing over to low angle of attack and jettisoning the tanks. Then a landing would be made at Mud Lake. We practiced this over and over and over again in the simulator with Jack McKay serving as NASA 1, the flight controller. A few seconds after launch we would simulate the failure and Jack would make the canned call “we see no flow” and Rushworth would then initiate his abort actions.

We knew that the instrumentation available to establish that the tanks were feeding was an indirect indication, but it was felt that adequate pre-launch test would validate the instrumentation. This instrumentation was a pressure transducer across an orifice in the helium lines that pressurized each of the external tanks. (We had better instrumentation under development but it was not available for this flight.) We had verified the pressure transducer operation during a planned captive flight with propellants in the external tanks. At the four-minute-to-launch point on flight 45 we verified that the pressure sensors were working satisfactorily by jettisoning a small amount of propellant. After launch on flight 45 the ammonia-helium pressure did not respond. As a result, 18 seconds after launch Jack announced over the radio the dreaded call, “We see no flow on Ammonia Bob.” Bob immediately began setting up to jettison the tanks. He reduced the throttle to minimum thrust and pushed over to the desired 6 degrees angle of attack then hit the tank jettison switch. He forgot that he had not shut down the engine and proceeded to do so. The tanks separated cleanly although at much higher dynamic pressure than desired. The emergency landing at Mud lake was as routine as an emergency landing in an X-15 could be. Post flight analysis led to the conclusion that the fuel was, in fact, feeding from the ammonia tank properly. I carried the lesson learned from this flight with me for the rest of my active flight test career. Do not make critical flight decisions on inadequate/marginal instrumentation!

The next flight (49) was flown with the ventral fin on to Mach 5.2. It was primarily to familiarize Pete Knight with the handling qualities in this configuration that would be flown on the following flight with external tanks. Flight 50 was the first successful flight with propellant in the external tanks. Improved propellant transfer sensors (paddle switches) were installed for this flight. The flight was flown as planned and reached a max Mach of 6.3. Flight 51 was flown to evaluate the aircraft’s handling qualities with a dummy ramjet installed on the fixed portion of the lower ventral fin. The dummy ramjet was jettisoned during final approach as was normal for the movable lower rudder due to lack of adequate ground clearance for landing. The flight was flown without external tanks and reached a Mach number of 4.8. Next was a flight with the ablative coating and dummy ramjet installed but without external tanks. The flight was flown to Mach 5 and the ablative generally performed as expected although there were indications of high heating in the lower ventral fin adjacent to the ramjet that we did not fully appreciate until the next flight. The ablative coating was refurbished and the aircraft prepared for its “all up” flight with external tanks, ablative coating, and dummy ramjet.

Fastest airplane flight of the Century

All was ready on the morning of 3 October 1967 as Col Joe Cotton started each of the eight engines on the NB-52B mothership affectionately named “Balls Eight” from its S/N 008. Pete had already been in the small cockpit of the X-15 for over an hour performing the pre-flight checks with the ground crew led by Charlie Brown and Larry Barnett and a host of test support personnel in the NASA control room. After takeoff of the B-52/X-15 with Maj Cuthill following in an F-104 as chase 1, it took about 50 minutes to reach the launch point abeam Mud Lake approximately 170 nautical miles north of Edwards.

X-15A-2 Launch

” I reached up and hit the launch switch and immediately took my hand off to go back to the throttle and found I had not gone anywhere. It did not launch. So I probably just got my hand off of it, because I reached up and hit it again and it launched the second time. Launch was very smooth this time.”

As the X-15 was falling from the B-52 he lit the engine and locked on to 12 degrees angle of attack. He was pushed back into his seat with 1.5 g’s longitudinal acceleration. The X-15 rounded the corner and started its climb. During the rotation as normal acceleration built up to 2 g’s Pete had to hold in considerable right deflection of the side arm controller to keep the X-15 from rolling to the left due to the heavier LOX in the left external tank. When the aircraft reached the planned pitch angle of 35 degrees his scan pattern switched from the angle of attack gage to the attitude direction indicator and a vernier index that was set to the precise climb angle. The climb continued as the fuel was consumed from the external tanks, then at about 60 seconds he reached the tank jettison conditions of about Mach 2 and 70,000 feet. He pushed over to low angle of attack and ejected the tanks. He was now on his way and would not be making an emergency landing at Mud Lake. “We shut down at 6500 (fps), and I took careful note to see what the final got to. It went to 6600 maximum on the indicator. As I told Johnny before, the longest time period is going to be from zero h dot getting down to 100 to 200 feet per second starting down hill after shutdown.” Final post flight data recorded an official max Mach number of 6.72 equivalent to a speed of 4534 miles per hour. From there down Pete was very busy with the planned data maneuvers and managing the energy of the gliding X-15. He approached Edwards higher on energy than planned and had to keep the speed brakes out to decelerate. On final approach he pushed the dummy ramjet eject button and landed on Rogers lakebed runway 18. He indicated he did not feel anything when he activated the ramjet eject and the ground crew reported they did not see it. Pete said that he knew something was not right when the recovery crew did not come to the cockpit area to help him out of the cockpit, but went directly to the back of the airplane. Finally when he did get out and saw the damage to the tail of the X-15 he understood. There were large holes in the skin of the sides of the fin with evidence of melting and skin rollback. Now we are talking Inconel-X steel that melts at 2200 degrees F. Later analysis would show that the shock wave from the leading edge of the ramjet’s spike nose had intersected the fin and caused the aerodynamic heating to increase seven times higher than normal. So now maybe we knew why the ramjet was not there

X-15 dummy ramjet search

“I did not feel anything when I pushed the button for the ramjet. I understand that there were people saw that it did come off and others say that they never saw it, so I don’t know where it is.” So said Major Pete Knight during the postflight debriefing.

The flight records indicated that the ramjet instrumentation ceased functioning 25 seconds after the engine was shutdown and the airplane reached Mach 6.72. So at about Mach 6 during the deceleration/glide the burn through had taken place. The obvious conclusion then was that the ramjet departed at that time and had gone it’s merry way from over 90,000 feet to the desert floor below; over 100 miles from Edwards.

Lower Ventural Heat Damage

Later that afternoon as several of us were reviewing the data records we noted an abnormal decrease in the sensitive longitudinal acceleration trace (indicating a sudden decrease in drag). Although it was a small change, it was instantaneous. We decided to go on the assumption that this could be where the ramjet departed the airplane. Correlating the time of day with the flight parameters we found that it was at about the 180 degree point during the turn over the south area of Rogers Lake bed at about Mach 1.0 and 32,000 feet. The airplane was in a 57 degree left bank and, more importantly, pulling 1.6 g. Now I was confident that this could have been the time that the dummy ramjet began it’s independent trajectory. So next, I time-correlated the radar data and found the spot where this event occurred and the heading of the aircraft at the time. This then was the Initial Conditions of the ramjet flight. Next, I drew a line on a map along the heading at the time I suspected it separated from the X-15. I could say that I did a detailed calculation of the drag coefficient for a tumbling ramjet, then a 5th order curve fit of the potential trajectory, corrected for winds; – but actually I just made an engineering estimate (guess) at a downrange distance. It turned out that the estimated resting place was right on the Edwards AFB bombing range. Placing a mark on the map at my selected impact point, I then drew a line perpendicular to the estimated track. Next I pick out some recognizable ground reference points on the map. As it turned out I selected the Rocket Base on Lehman Ridge west of Rogers Dry Lakebed and a mountain peak in the San Bernardino mountains.

Dummy Ramjet Impact Site

Now it was time to present my theory to the team. There were many disbelievers of the theory who felt the dummy ramjet was way up north of Edwards. However, Bill Albrecht, the NASA Operations Engineer in charge of X-15A-2, was willing to humor and trust me. We contacted Joe Rief, the AFFTC airfield manager, and got permission to go wandering out on the bombing range. Bill and I got in a NASA radio equipped carryall van, radioed Eddie Tower, and headed out on the range. We drove east on the road on the bombing range with me looking for that general area where I had drawn the line on the map. We finally stopped the van and we walked down the road until I could hold my arms out and line up the Rocket Base and the mountain peak. At that point, I had Bill head North-East along my magic line and I headed South-West.

Shock! disbelief! glee! cold chills up my spine! …. after walking only about 200 yards, I saw the ramjet lying on the sand in between the tumble weed bushes in two major pieces.

I hollered with excitement to Bill but he was out of hearing range. So I ran back to the road and got his attention and we managed to back track to where I had seen it. There was a depression in the sand back up the track where it made first contact before coming to rest. We gathered up the nose cone and the machined conical steel pressure probes that were the very leading edge of the ramjet and headed back to the van.

The main body of the dummy ramjet was too large and heavy for us to return. It was almost quitting time at NASA when we carried our trophy up to the pilot’s office. I strutted around like the hen that laid the golden egg. Fitz Fulton, I recall was favorably impressed. The next day Albrecht and I went back out to the site to direct a Huey helicopter to the location. A cable was attached to the main part of the ramjet and it was flown back to NASA.

Inspection of the ramjet revealed that it also had major melting and damage due to the high aerodynamic heating in the area where it attached to the ventral fin. In addition, 3 of the 4 explosive bolts that held the ramjet on had been fired, undoubtedly due to the high temperature. The fourth bolt had structurally failed; which was apparently all that was securing the ramjet on the X-15 from about Mach 6 down to Mach 1.

Dummy Ramjet Damage

The X-15 program winds down

There was major damage to the structure of the fixed lower ventral fin from excessive aerodynamic heating due to shock wave impingement from the nose of the dummy ramjet. The aircraft was trucked to North American Aviation at LAX to be repaired, however it was to never fly again. Six months after the Mach 6.72 flight of ship 2 the Number 3 X-15 crashed near Randsburg, California killing Major Mike Adams. Over the next year the number 1 X-15 flew eight more flights with the last flight occurring on 24 October 1968, the 199th flight of the X-15 program. Several attempts were made to fly the 200th flight but to no avail. On 20 December Pete Knight was in the cockpit of the X-15 under the wing of Balls Eight ready to taxi to the runway when a freak snow storm moved over Edwards and the flight was cancelled. Thus ended the flying portion of the most successful X-plane program in history, however the analysis of the research data from the program continued for many years afterward.

How did I get involved in all this? RPRT. I was fortunate to be the Air Force flight planner during this time period on the X-15 program. I got to develop the flight plan using the X-15 man-in-the-loop simulator, train the pilots and then participate in the control room during the flight. Throughout my flight test career RPRT was good to me and allowed me to participate in some major flight test events and programs. It also presented me with the opportunity to work with some of the top people in the flight test community and this thing we called the test team. The AFFTC/NASA FRC X-15 team was one of the best. RPRT? – RIGHT PLACE RIGHT TIME.

X-15A-2 in USAF Museum sm

The above photo includes the X-15A-2, the X-24B, and the Martin SV-5J labeled like the X-24A.  See the X-24B Runway Landing story to understand.  Photo is from the National USAF Museum web page.
NOTE was posted that the Research and Development display area will be closed beginning in 1 May 2013 do to budget reduction.