Our goal was to design and build a very lightweight vehicle that could be towed across the lakebed with a ground vehicle and, later, aloft with a light plane, the way sailplanes are towed. Based on the tiny model used in the filmed flights, the first lifting-body vehicle was also called the M2-F1- the "M" signifying a manned vehicle and the "F" designating flight version, in this case the first flight version.
Months before the M2-F1 was completed, it had already been dubbed the "flying bathtub" by the media. The first time seems to have been on 12 November 1962 in the Los Angeles Times article "'Flying Bathtub' May Aid Astronaut Re-entry." The article included a photo of Milt Thompson sitting in a mock-up of the M2-F1 that, indeed, looked very much like a bathtub.
Paul Bikle decided to run the project locally, financing it entirely from discretionary funds. He thought that a volunteer team at the NASA Flight Research Center, supplemented with local help as needed, could build the M2-F1 faster and cheaper than NASA Headquarters could through a major aircraft company. As history proves, Bikle was right.
The M2-F1 was built entirely in four months. Engineers at the Flight Research Center also kept the cost of designing, fabricating, and supporting the M2-F1 to under $30,000, about the cost of a Cessna. At the time the M2-F1 was built, someone associated with a major aircraft company was cited anonymously as saying that it would have cost an aircraft company $150,000 to build the M2-F1. The extremely low-cost M2-F1 program would have invaluable results later, proving to be the key unlocking the door to further lifting-body programs. 1
A Matter of Teamwork: Building the M2-F1, 1962-1963
After our meeting with Paul Bikle and Al Eggers, we were swiftly swept up into the enthusiastic atmosphere of the lifting-body program. On his return to the NASA Ames Research Center, Eggers asked Clarence Syvertson, his deputy, to coordinate all wind-tunnel tests that we needed in support of our design and flight-planning activities. Meanwhile, at the NASA Flight Research Center, Bikle asked me to put together a team to design and fabricate the first lifting-body vehicle.
Long before I began to put together that team, Dick Eldredge and I had already fairly much agreed that the basic design would include two structural elements, a core...
...steel-tube structure and a detachable aerodynamic shell. However, the real work lay yet ahead of us in the detailed design of the hundreds of parts needed for the actual vehicle.
To do this work, I selected a design-and-fabrication team made up of four engineers and four fabricators, all of whom were aircraft buffs involved with home-building their own airplanes, most of them members of the Experimental Aircraft Association. These individuals had worked together to some extent on previous programs in the Flight Research Center's unofficial "Skunk Works." The group's chief designer was Dick Eldredge. To lead the team, we got Vic Horton, a no-nonsense operations engineer who took pride in keeping to schedules.
Horton picked up a few extra part-time volunteers as the work got underway. Hardware designers, besides Eldredge, included Dick Klein and John Orahood. Meryl DeGeer calculated stress levels in the structure to verify the adequacy of the design. Ed Browne, Howard Curtis, Bob Green, Grierson Hamilton, Charles Linn, George Nichols, and Billy Shuler fabricated the internal steel-tube carriage of the M2-F1 as well as its aluminum sheet-metal tail fins and controls.
 Once we had the initial team, we needed a place to work. We sectioned off a corner of the fabrication shop with a canvas curtain, labeling it the "Wright Bicycle Shop." Indeed, we felt very much like the Wright Brothers in those days, working at the very edge between the known and unknown in flight innovation. In the "Wright Bicycle Shop," we put the drafting boards next to the machine tools for maximum communication between designers and fabricators. This strategy worked extremely well. A fabricator could lean over a designer's drafting board and say, "I could make this part faster and easier if you would change it to look like this."
I think our project was Bikle's favorite at the time. We would see him at least once a day, and we got a great deal of extra attention from him. A few chose at the time to grumble about Bikle acting as if he were the super project engineer on the M2-F1, but I think that we thrived as a team from his presence. For one thing, I have never since seen on later projects enthusiasm or morale as high among team members as existed on the M2-F1 project. In fact, it really isn't an exaggeration to say that we had trouble keeping the team from working through lunch, during the evenings, or on weekends.
The M2-F1 project also benefited from Bikle's experience and suggestions. While we didn't have to use his suggestions, we did need to have good reasons for not using them. A time when one of his suggestions helped us a great deal-and there were many such times-was when we had everything else thought out and had begun trying to decide how to build the aerodynamic shell.
The core of the dilemma had to do with the shell's weight, which we hoped to keep under 300 pounds, wanting a vehicle of minimum weight so that the M2-F1 would fly slowly enough that a ground vehicle could tow it aloft. Dick Eldredge and I were thinking about building the shell out of fiberglass, but we weren't sure we could keep the weight within necessary limits.
We knew that our vehicle design lent itself easily to being built in two different locations by two different teams, the two main assemblies being joined later. We knew that we could build the internal steel-tube carriage, tail surfaces, and controls in our NASA shop while the outer shell was being built elsewhere by a second team. But where and by whom?
Bikle suggested that we talk to a sailplane builder named Gus Briegleb, who operated an airport for gliders and sailplanes at El Mirage dry lake, 45 miles southeast of Edwards Air Force Base. Bikle also suggested that he might be able to find enough money in his discretionary fund to contract Briegleb to build the shell for us out of wood.
One of the nation's last artisans building aircraft out of wood, Briegleb had founded the Briegleb Glider Manufacturing Company during World War II to design and build wooden two-place trainer gliders for Army pilots being trained to fly troop-assault gliders. The two-place trainer gliders were used to train these pilots to fly in formation on a tow-line and performing precision dead-stick landings after release from Navy R-4D tow-planes (same as the Air Force C-47). The troop gliders were used extensively during the Allied invasion of France, with the Briegleb Glider  Manufacturing Company being one of only a few companies manufacturing the trainers.
In 1962, when we contacted him, Gus Briegleb was trying to keep alive the art of fabricating wooden airplanes by selling kits of a high-performance sailplane that he had designed. Between selling these kits and operating the glider-sailplane airport at El Mirage, Gus was making a living, but he definitely was not getting rich.
Briegleb responded enthusiastically when we approached him about building the M2-F1 shell out of wood. Although wood eventually gave way to aluminum sheet-metal in the production of aircraft for a good number of excellent reasons, wood is still one of the more efficient structural materials for aircraft in terms of fatigue life, vibration damping, and strength-to-weight ratios. Briegleb initially proposed to build the shell out of wood for only $5,000.
Thinking that sum was too low, Bikle asked Briegleb if he had considered overhead, profit, and unforeseen problems that were likely to arise during the building of the shell. A builder, not a businessman, Briegleb admitted he had not considered these things. Bikle said that he could authorize up to $10,000 for the wooden shell, that being at the time the limit for small purchases at the NASA Flight Research Center. Briegleb agreed to meet the 300-pound target weight and the strength specifications that Dick Eldredge and I had determined from airload calculations, and he agreed to deliver the shell four months from the date the contract was signed.
 When Briegleb got into the detailed design process, he found that the shell would have to be far more complicated than he had originally thought to keep it to the specified weight. He had underestimated the hours needed to build the shell by at least a factor of three. The shell had to be made with two internal keels to carry the loads to the steel-tube frame. Hundreds of small wooden parts made up these built-up wooden keels. To support the outer skin shape, the keels also had multiple internal crossbracings made of miniature wooden box beams of webs and spar caps, all nailed and glued together.
When we saw the predicament that Briegleb was in, we sent him some help: Ernie Lowder, a NASA craftsman who had worked on building Howard Hughes' mammoth wooden flying-boat, the Spruce Goose. Despite having Lowder as a full-time fabricator, Briegleb says he still ate quite a bit of the $10,000 contract. Nevertheless, Briegleb was very proud of his work, and so were we. He delivered the shell to us on time, at cost, and slightly under the 300-pound weight limit. I think we gained a great advantage by being able to use the last of America's finest wooden-airplane craftsmen to build the shell of the M2-F1.
As Briegleb's team built the outer shell, NASA craftsmen built the internal steel-tube structure. The steel-tube carriage was finished first, in about three months, and while the wooden shell was still being fabricated at El Mirage, the carriage was being rolled around on the landing gear. Eldredge and I had designed the M2-F1 so that it took only four bolts to attach Briegleb's shell to the internal structure.
Team Three for Analysis
Once the two teams were in place and building the two main structures of the M2-F1, I realized we also needed a third team to do the analysis on aerodynamics, control rigging, and characteristics of stability and control to support flight tests. Using the wind-tunnel data on small-scale lifting-body models that we were beginning to get from the NASA Ames Research Center, I could determine the basic stick-to-surface gearing in pitch for the outer elevon surfaces and the upper body flap. Rotating the lifting body nose-up to moderate angles of attack amplified to high angles the flow on the aft sides of the bulbous M2-F1 shape.
Tufts of yarn on the small-scale wind-tunnel model had indicated that its outer elevon surfaces experienced about twice the change in angle of attack experienced by the model's nose. Consequently, I specified gearing for the outer elevons to move three times more than the body flap with fore and aft travel of the pilot's control stick. I did this so that, when a differential roll side input was made from the pilot's stick, there would be no risk of stalling an elevon surface, causing reversal of the roll or loss of control of the vehicle during the roll.
Determining control rigging and gearing for turn control was not as obvious as that for pitch control. The M2-F1, and almost all of the later lifting bodies, have extremely high dihedral-that is, with wind from the side (called "sideslip"), the vehicle wants to roll in the opposite direction. Because of this characteristic, rudder deflections  actually resulted in roll rates higher than those produced by differential elevon deflections. Since lifting bodies also have extremely low roll damping from having no wings to resist roll rates, and since Dutch roll results from the extremely high dihedral inherent in most lifting bodies, we had a potentially dynamic problem in stability and control if we did not do the right thing in designing the control system.
Obviously, we needed help from the experts in stability and control at the NASA Flight Research Center, all of whom were currently working on the X-15 program. In its later stages after three years, the X-15 program still had number-one priority at the Center. Because the X-15 program was so well organized and ran so smoothly by that time, many aspects were getting to be routine, even though there were still some surprises showing up during the speed and altitude buildup as the flight envelope was being expanded. Our unofficial lifting-body project was able to recruit the help it needed, despite the on-going X-15 program, thanks to Bikle's policy that the NASA Flight Research Center had an equal responsibility to aeronautical research directed to the future.
 My first volunteer was Ken Iliff, now the Chief Scientist at NASA Dryden, who at the time was a bright and enthusiastic twenty-one-year-old engineer just out of college and doing a mundane analytical task in reducing X-15 flight data. Iliff poked his head in the office where Eldredge and I were working and, after inquiring what we were doing, asked if there was anything he could do to help us out. "Sure!" I replied quickly, not one to refuse any help I could get.
After explaining to Iliff that we planned to get a high-speed ground-tow vehicle to tow the full-scale M2-F1 model across the dry lakebed, I asked him to take a stab at calculating what the rotation and lift-off speeds would be on ground-tow, information we needed in determining the requirements for the tow vehicle. We could have a problem, I explained, if the aerodynamic pitch controls were not strong enough to lift the nose, overcoming the nose moments from the wheel drag and the tow-line force.
Iliff got busy. He calculated rotation speed to be 59 miles per hour and lift-off speed to be 85 miles per hour. Later, when we actually ground-towed the M2-F1, we measured rotation speed at 60 miles per hour and lift-off speed at 86 miles per hour. Needless to say, we were impressed with this young engineer.
Although he continued to maintain his obligations to the X-15 program, Iliff became more and more involved in our little lifting-body program. He started looking at the stability and control characteristics of our strange bird-just in case we did try to fly the M2-F1 following the full-scale wind-tunnel tests. Iliff sought help from his mentor, Larry Taylor, another engineer then studying pilot-control problems on the X-15 who was experienced in applying some of the latest techniques in analyzing stability and control problems on new aircraft configurations. Although Taylor had applied some of those techniques to the X-15 with success and gained the credibility of a number of his aerospace peers, some of the old-time flight-test engineers, including Paul Bikle, considered Taylor a radical practicing a kind of engineering witchcraft.
Taylor claimed he could use mathematics to describe the piloting characteristic of a test pilot, then predict the outcome of a planned flight. He called this the "human transfer function." Bikle disagreed, saying there was no way to predict how a pilot would perform on any one day, emphasizing that a pilot's performance was impacted by events in his personal life, such as having a spat with his wife or partying the night before a flight.
I felt both viewpoints had validity. I agreed with Taylor's viewpoint that there are fundamental differences in how individual pilots react to a difficult control task. In a stressful situation that leads to problems with pilot-induced oscillation, the gains of some pilots rise much faster than those of other pilots. An aircraft can go out of control if it has a tendency to oscillate in a particular direction, especially if the pilot tries to stop the oscillation by chasing the aircraft with the controls. Sometimes the airplane will halt the oscillations on its own if the pilot will slow down or stop moving the controls.  However, this is not the usual or most natural reaction for a pilot during a stressful situation, for as arm and leg muscles tighten up from stress, control movements usually increase.
The master sorcerer of mathematical voodoo, Larry Taylor was at the time passing his mystical art on to his apprentice, Ken Iliff, especially a strange engineering plot called "root locus" that many pilots then thought was pretty far-out stuff. The three categorical ingredients of this mathematical potion were the airplane's aerodynamics, inertial data composed of weights from all parts of the airplane, and flight conditions such as speed, altitude, and angle of attack. The root-locus plot gave results for different types of pilots, ranging from the totally relaxed pilot who does nothing with the controls to the high-gain pilot who moves the controls rapidly.
One magical point on the plot called the pole represented the "do-nothing" pilot. Another magical point called the zero represented the high-gain pilot or autopilot. A line connected these two points, representing all pilots between the two extremes. If the line moved into the right side of the plot, the pilot/aircraft combination was deemed unstable, predicting loss of control of the aircraft. Despite the fact that in the early 1960s even a number of engineers considered the root-locus analysis to be some sort of witchcraft, today root locus is a common mathematical tool used by stability and control engineers.
According to Bob Kempel, then a stability and control engineer at the NASA Flight Research Center with considerable influence on the design of control systems for experimental manned and unmanned NASA aircraft, root locus is a tool by which engineers can predict potential instability prior to flight so that a possibly catastrophic situation can be avoided by either pilot training or modification of the flight control system. "The intent of the engineer," says Kempel, currently active in control-system designs, "is to provide the test pilot with a pilot/airplane combination that. . .will remain stable, regardless of pilot gain," work-load variations, or emergency control situations.
Well tutored by Taylor in this technique, Iliff set off to predict the M2-F1's qualities during flight. He modeled the lifting body mathematically for free-flight as well as for flight while on tow. He found that the tow-line force was quite high in opposing the high drag of the lifting body, adding a high level of static stability to the system, much like towing a high-drag target behind an aircraft.
About this time, two more volunteers showed up whose help would be invaluable on the M2-F1. Bertha Ryan and Harriet Smith, two junior engineers who did not have strong obligations to the X-15 program, asked me what they could do to help. In getting Ryan and Smith as well as so many other volunteers, I was enjoying a bit of luck. The 50 percent of the work force at the NASA Flight Research Center not committed to the X-15 program wasn't being taxed fully in support of other official NASA programs. Even in those days, bureaucratic methods of operation caused tremendous lags to appear between approval and funding cycles. Furthermore, peaks and valleys in workloads occurred at the field stations whenever NASA Headquarters approved,  turned down, or canceled a program, no matter how well the field managers scheduled work.
I was one of those Johnny-on-the-spot opportunists who would move in with my small program to take advantage when valleys appeared in workloads. Most supervisors liked to keep their people busy, and it didn't hurt the lifting-body project one bit to have the local director interested enough in our project to send us new volunteers. Bikle had encouraged Ryan to work with us, knowing that since she owned her own sailplane, she would have some practical as well as analytical skills useful to the project.
Although engineers today are as often women as they are men, women engineers were not common in the early 1960s. After they volunteered on the M2-F1 project, I explained to Ryan and Smith that Milt Thompson wanted some sort of simulator for practice before flying the M2-F1. Good friends, Ryan and Smith thought the task would be fun. They also liked the idea of working as an all-woman simulation team-perhaps one of the first for those times-with Ryan preparing the aerodynamic data input and Smith mechanizing the simulation. Neither of them had ever set up a flight simulator before, but they felt that while the task would be challenging, they could also learn quite a bit by doing it. Actually, all of us were fairly naive about simulators in those days, even though a simulator had been set up for the X-15.
 When Eldredge and I had designed the M2-F1 control system to be flexible, we had thought we were being clever, never realizing that we created a veritable Pandora's box. Instead of having just one version of the M2-F1 to set up on the simulator, we had as many as five, one for each way our variable control system could be hooked up in its swashplate design.
We had six control surfaces on the rear of the vehicle that we could hook up in any combination to the pilot's stick and rudder pedals for pitch, roll, and yaw control-two vertical rudders, two outboard elevons that we called "elephant ears," and two horizontal body flaps at aft top. We also had a removable center fin, but no lower body flaps.
Most of the simulators used at that time were purely analog, requiring 30 or 40 hand-adjusted electrical potentiometers (called "pots") to be set up for each simulation session. It was very easy to make a mistake while setting up these pots, especially by setting a switch to give a minus instead of a plus sign, or vice versa. The only way to guarantee a correct simulation was to require a verification process for each simulation session. Despite their inexperience in setting up a simulation, Ryan and Smith were very methodical. They kept good notes and records, working hard at doing a good job.
Since pitch control did not seem to be a problem on the simulator, we spent much of our time trying to determine the best way to control roll and yaw on the M2-F1. Early on, we decided to eliminate the center fin as well as the differential control on the body flap. The center fin only made the already high dihedral even higher. Besides, we already knew from small-scale wind-tunnel tests that we had plenty of directional stability from the two vertical side fins. By making the body flap single-pitch rather than split, it could be used like an elevator, eliminating the need for the center fin as a fence against adverse yaw from a body-flap elevon system. The shop team had already fabricated a two body-flap system, but by the simple expedient of bolting the right and left flaps together, they made one large flap.
We had narrowed the lateral-directional control system down to two basic possible schemes. In the first control scheme, right stick deflection would move the outer elevons for roll to the right, and the right rudder pedal would move both vertical rudders to the right. In the second control scheme, right stick deflection would move both vertical rudders to the right, and the right rudder pedal would move the outer elevons for roll to the right. Working with us as a part of the analytical team by flying the simulator in the ground cockpit, Milt would give us a pilot's rating for each of the configurations we investigated. His rating system was on a scale of one to ten, depending on the difficulty of changing and holding headings.
Eldredge and I had fairly much made up our minds in favor of the first control scheme, intuition having told us that elevons or ailerons should be controlled by the stick while rudders should be controlled by the rudder pedals. We were shocked when Milt told us that he preferred the second control system. His reasoning was that roll rates resulting from the rudders being deflected were twice as high as those resulting from differential elevon deflection. Milt felt that he could control the vehicle by using.....
 ....proper piloting technique, and he said he would rather have the higher roll rates available to him if he needed them.
If any research pilot could use proper piloting technique, it was Milt Thompson. He was a cool, disciplined pilot who could think well during emergencies or under other stressful conditions. He had already proven several times during the X-15 program that he could and would work closely with engineers in solving potential flight problems. He also liked to fully understand the idiosyncrasies of an aircraft before he flew it. In my opinion, Milt Thompson belongs up there with Chuck Yeager in any estimate of historic greatness for test pilots. Milt Thompson not only had the same stick-and-rudder skill and coolness under fire that Chuck Yeager had, but he also had a certain elegance in thinking when dealing with engineers. Milt had such an air of modest dignity and credibility about him-what today might be called "charisma"-that when he said he preferred the second control system for the M2-F1, we listened to him, even though we didn't necessarily like his choice.
At this point, Iliff did a root-locus plot for both control systems. He determined that there was no problem involved with using the first control system, with its use of the elevons for roll control. However, he found there could be a large problem with the second system which used the rudders for roll control. With the second system-the one Milt preferred-the M2-F1 could be driven unstable in Dutch Roll, resulting in loss of control of the vehicle, if the pilot's gains were too high. Although Taylor was doing a good job in verifying the root-locus technique on the X-15 program, it was still too new to be accepted by others as a valid design or planning tool. Despite Iliff's conclusions, Milt still insisted on using the second control system. His plan for the first car-tow tests was to gently rotate the M2-F1 nose-up until it was flying a few inches off the lakebed before he made any rudder or control-stick inputs. Then, he would move the controls very slowly to test them out. If things didn't look good, he would set the vehicle back down on its wheels, and we could try the other control system.
While the simulator is a wonderful tool in designing aircraft and planning flights, simulator results must be interpreted very carefully. A heavy smoker, Milt would sit in the simulator's cockpit totally relaxed, a cigarette in one hand, flying with the other hand. Under those conditions, unlike those of actual flight, he had no tendency toward driving the Dutch Roll mode unstable, as Iliff had predicted he would in actual flight.
During the month between the completion of the internal structure and the completion of the wooden shell, Vic Horton decided to test the ground stability and control of the internal structure with landing gear. The wheels and nose gear assembly were taken from a Cessna 150 light aircraft. The pilot steered by foot pedals through the nose gear. Milt being away on a trip, X-15 research pilot Bill Dana volunteered to sit in the pilot's seat while the structure was towed by automobile across the dry lakebed.
Dana was soon having a great time, sashaying back and forth like a water skier at thirty miles an hour on a 300-foot tow-line behind the automobile. Having good control of the steering, Dana was building a lot of confidence. Then, he pulled far over to  one side and pulled the tow-release to test it out. Unfortunately, he had been holding a large amount of rudder pedal to compensate for the side pull of the tow-line.
Suddenly, the vehicle veered sharply and started to roll over. Dana countered with the rudder pedal. A wild oscillation began, the M2-F1 steel-tube skeleton doing a wheely to the right, then a wheely to the left. Finally, Dana lost control, and the M2-F1 flipped over. Fortunately, the fabricators had built a strong rollover structure, and both pilot and vehicle came out of the episode without injury. Dana was embarrassed by the incident, and we kidded him mercilessly for years, saying we'd call on him again if we ever needed to run an unmanned structural test.
Final assembly of the M2-F1 began when the wooden shell arrived from El Mirage. We lowered the steel-tube internal structure, minus the landing gear, through a large rectangular cutout in the top of the wooden shell. We inserted the landing gear legs through holes in the shell and bolted them to the inner steel structure. Four bolts on the two wooden keels attached the shell to the inner steel structure. The aluminum tail surfaces, built in the NASA Flight Research Center shop, were then bolted onto the wooden shell, and controls were hooked up by push-pull rods. Finally, we attached to the shell a Plexiglas canopy, made by Ed Mingelle of Palmdale for the M2-F1 after Bikle recommended that we go to him since he was a specialist in making custom canopies for sailplanes. Exactly four months from the day when Bikle had told me to begin building, the completed M2-F1 rolled out of the "Wright Bicycle Shop."
NASA's Muscle Car: Ground-Towing the M2-F1
Dick Eldredge and I had designed the M2-F1 to weigh 600 pounds. However, like most prototype airplanes, it had grown in weight during fabrication, the completed vehicle weighing in at 1,000 pounds. From Iliff's calculations of the M2-F1's tow force and lift-off speed, we knew that to do taxi tests with the M2-F1 before the wind-tunnel tests at NASA Ames, we needed a ground-tow vehicle with greater power and speed than any of NASA's trucks and vans could provide. 2
First, we needed a ground vehicle that could tow the M2-F1 at a minimum of 100 miles per hour. Secondly, we also needed a ground vehicle that, at that speed, could handle the 400-pound pull needed to keep the 1,000-pound lifting body airborne. In meeting these needs, we ended up with what was probably the first and only government-owned hot-rod convertible.
Once again, a volunteer came along who had the know-how that we needed. Working in operations at the NASA Flight Research Center at the time was Walter ...
...."Whitey" Whiteside, a retired Air Force maintenance officer who was also a veteran dirt-bike rider and expert hot-rodder.3 Whitey volunteered to help us out by finding, purchasing, modifying, testing, maintaining, and driving the high-powered ground-tow vehicle that we needed.
At the time, the Pontiac Bonneville seemed the best choice, this model so named because it had been the big winner the year before in Utah at the Bonneville Salt Flats time trials. With Boyden "Bud" Bearse's help in the procurement department, Whitey was able to make a special order from the factory for a Pontiac Bonneville ragtop convertible with the largest engine then available, a four-barrel carburetor, and four-speed stick shift. NASA engineers at the Flight Research Center equipped the Pontiac with its tow rig and airspeed measuring equipment.
Whitey took the car for modification to Bill Straup's renowned hot-rod shop near Long Beach, where the straight-piped Pontiac was modified to run a consistent 140 miles per hour. There, auto-shop technicians also applied their hot-rod wizardry to the Pontiac, producing maximum torque at 100 miles per hour as measured on by a  dynamometer. They added a special gearbox, with transmission gear ratios significantly different from those that had helped the Bonneville win at the Salt Flats, enabling the Pontiac eventually (once drag slicks were installed) to tow the 1,000-pound M2-F1 to 110 miles per hour in 30 seconds. The Pontiac's souped-up engine got about four miles to the gallon. Whitey got full support from the NASA fabrication shops headed by Ralph Sparks (Sparky). Sparky and his right-hand man, Emmet Hamilton, took responsibility for keeping the Pontiac running and making any modifications required by Whitey.
For the safety of the driver and two onboard observers, Whitey had roll bars added to the NASA muscle car. He also had radios and intercoms installed. The front passenger bucket seat was reversed and the back seat was removed, replaced by another bucket seat so that a second observer could sit facing sideways. Of course, the Pontiac had to have government plates, the NASA logo on both sides, and racing stripes. And just so no one would be encouraged to think the car was someone's personal toy paid for with government funds, the hood and trunk of the Pontiac were spray-painted high-visibility yellow so that the convertible looked just like any other flight-line vehicle.4
When the car was finished at the hot-rod shop, Whitey drove it back to the NASA Flight Research Center. A motorcycle fanatic and hot-rodder who loved speed, Whitey found it difficult to hold back once he got the Pontiac outside Los Angeles and on the highway across the desert. Realizing he would get his chance later to open up on the dry lakebed, he was being particularly careful to hold the Pontiac's speed to the posted speed limit when he saw in the rearview mirror the red light of a California Highway Patrol (CHiP) vehicle closely tailing the Pontiac. Pulling over to the side of the highway, Whitey wondered what he'd done wrong. It turned out that the officer was merely curious, having never before seen a government-owned convertible, especially one with a souped-up engine. After a careful up-close look and Whitey's explanation of how the car would be used, the officer drove away, shaking his head in amazement.
The Pontiac also caught the eye of other drivers whenever Whitey took it out onto little-traveled desert highways northeast of Edwards AFB through Four Corners, often into Nevada with its then anything-goes speed limits, to calibrate the car's speedometer, as typically done with research airplanes. Laughing, Whitey recently recalled one particular time when he headed out on just such a venture with one of the base's pilots  in the car. As the Pontiac rumbled along, engine-exhaust system roaring as the speedometer moved above 100 miles per hour, Whitey glanced at the silent pilot, only to find him ashen-faced and trying to disappear into the seat. 5
When we had the M2-F1 completed and ready for wind-tunnel testing at NASA Ames, we were still divided on which basic roll-control scheme to use. Bertha Ryan, Harriet Smith, and Milt Thompson backed their interpretation of simulation results, saying the rudders would give the best roll control. Ken Iliff and Larry Taylor countered with what their root-locus plots showed-that using the rudders for roll control would lead to pilot-induced oscillation. On the other hand, I thought that the outboard elevon surfaces simply looked right for roll control, and I believed that rudders were meant for yaw-not roll-control. In the end, we agreed to use the scheme Milt Thompson preferred, with the pilot's stick hooked to the rudders for roll control, as long as we could reconfigure to the other scheme if that one didn't work.
Of course, we had no official approval to flight-test the M2-F1, which was supposed to be merely a full-scale wind-tunnel model. Sitting in the cockpit, Milt Thompson reasoned, "Maybe it wouldn't really be flying if we just lifted it off the lakebed a couple of inches." Boosting our confidence was the data we had from the earlier small-scale wind-tunnel tests. When approached, Bikle said, "Go for it, but be careful."
We were very careful as we began on 1 March 1963, making several runs in car-tow at lower speeds, gradually working up to the nose lift-off speed of 60 miles per hour on 5 April 1963. During these runs, Milt became familiar with the cockpit and with visibility out the top, through the nose window at his feet, and out the side window level with his feet, these windows necessitated by the anticipated high angle of attack. He also became adept at nose-gear steering and using the differential brakes and tow-line release.
After a week of these cautious towings at lower speeds, Milt said he was ready to try a lift-off. Following Milt's radioed directions, Whitey took the Pontiac and the M2-F1 on tow up to 86 miles per hour, the 1,000-foot tow-line giving Milt plenty of maneuvering room.
Slowly Milt brought the nose of the little lifting body up until the M2-F1 got light on its wheels. Then, something totally unexpected happened. The M2-F1 began bouncing back and forth from right to left. Milt stopped the bounce by lowering the nose, putting weight back on the wheels. Several times he again brought the nose up until the M2-F1 was light on its wheels, and each time the vehicle reacted the same way, Milt ending the bounce by lowering the nose as he had the first time.
Later, in our little debriefing room, Milt said that he felt that if he had lifted the M2-F1 off its wheels, it would have flipped upside down in a roll. We started theorizing about the cause of the problem. Milt felt it had something to do with the landing  gear, wondering if there wasn't enough damping in the oleo-type shock system. Ken Iliff suggested that maybe Milt was feeding the roll motions with the stick or rudder pedals. "Absolutely not," replied Milt, adding that he had made sure during lift-off that he wasn't making roll or yaw control inputs.
We planned to get a little data for analyzing the problem by installing an instrumentation system in the M2-F1 after we returned from wind-tunnel testing at NASA Ames. Before that, however, using a ground-chase vehicle, we made some 16-mm movies taken from the rear of the M2-F1, having painted references stripes on the rudders so we could determine their positions. The movies showed that the rudders were moving back and forth during lift-off. When Milt saw the movies, he concluded that slop and inertial weights in the rudder system-and not the pilot-were causing the rudders to move.
Larry Taylor suggested that we construct data from the movie frames. Using a stop-frame projector, we could determine right and left rudder positions and body roll angle on the M2-F1 by its position against the horizon in the back-ground. We projected the filmed images of the M2-F1 onto a large sheet of paper we had hung on the wall. Using a protractor, we measured the roll angle and positions of both rudders in each movie frame. Using the frame rate of the projector, we then produced plots or time histories of the rudder movement and roll angle. Producing flight data in this way was hard, mundane work. Ken Iliff, Larry Taylor, and I took turns working with the data until we had in hand the results that Iliff and Taylor needed to analyze the problem.
In the hangar, we examined the rudder control system, finding it exception-ally stiff. No way could the rudders be moved without moving the pilot's stick. We examined the weight distribution of the rudder system, looking for how inertia could cause the rudders to move during vehicle roll. We still could not find the cause of the rudder motions.
Ken Iliff compared the phase relationships between rudder position and roll angle, giving Larry Taylor his findings. The control motions were typical of what a pilot would put in to combat roll oscillations. Finally, Larry Taylor and Ken Iliff put together a strong statement, saying they had no doubt that "knowingly or unknowingly" the pilot was working to combat the roll and that continuing to try to fly the M2-F1 with the control system driving the rudders from the pilot's stick would, during roll control, lead eventually to loss of control of the vehicle. They insisted that the current control system be abandoned and the other control system driving the elevons from the pilot's stick be hooked up for the next series of car-tow tests.
We couldn't share their conclusion and recommendation with Milt Thompson at the time, Milt being away on a trip. We had only one week left after Milt's return for car-tow tests before we were scheduled to go into full-scale wind-tunnel tests at NASA Ames. Given the strength of Taylor and Iliff's conviction about the control systems, I didn't want to waste time doing more car-tow tests with the original control system, so I asked Vic Horton to change the control system as Taylor and Iliff had recommended so that, when Milt returned, the M2-F1 would be ready for more car-tow tests.
 After I made this decision, I noticed that the group had lost some of its harmony and camaraderie. Tension began to build between group members as they began to realize that a pilot's life could be at stake in this disagreement within the group over which roll-control system we used in the M2-F1. Milt Thompson was such a personable guy and worked so closely with us almost daily that emotions started emerging whenever critical decisions had to be made. I began to think that maybe it was better to have the research pilot more distant from the project people.
By backing Iliff and Taylor's recommendation, I had alienated Bertha Ryan and Harriet Smith to some extent. Ryan read me the riot act for not including Milt in the decision to change the control system, saying that, after all, it was his life at stake. I replied that Milt Thompson still had veto power as the pilot and that, if he insisted we do so, we would change back to the original control system. Ryan seemed satisfied by what I had said, but harmony on the project remained strained from that point. Even bigger conflicts would come later in the lifting-body program as the project grew.
As soon as Milt Thompson was back, I told him about the change made in the roll-control system. He was disappointed, wanting to do some more testing while using the previous control system, but he accepted the change, saying he still thought the problem was caused by the landing gear and that, when the new control hookup didn't solve the problem, we could go back to the original hookup.
With Milt Thompson onboard, we again hooked up the M2-F1 to the Pontiac and, with Whitey at the wheel of the Pontiac, off they charged across the lakebed. Cautiously, Thompson rotated the nose of the M2-F1 until there was very little weight left on the wheels. He continued to rotate the nose until the wheels were about three inches above the lakebed. The M2-F1 remained steady as a rock. We made another run, this time to an altitude of three feet. Thompson was gently maneuvering the M2-F1 right and left behind the Pontiac, but the lifting body showed no tendency to oscillate.
By now, Whitey had gone to Mickey Thompson's hot rod shop in Long Beach to replace the Pontiac's rear tires with drag slicks, a change that increased the car's towing speed to 110 miles per hour. Normally, drag racers use the wide, high-traction, threadless tires generally known as "slicks" because torque from the drive train to the lower gears is greatest at the start of the very short race known as a "drag," when tire slippage is most likely to occur. Our experience was exactly the opposite, with the height of drag found at the high-speed end of a tow. At about 90 miles per hour minus the slicks, the tires on the Pontiac would start slipping. Adding the drag slicks on the rear wheels of the Pontiac increased the towing speed enough to allow Milt Thompson to climb to twenty feet in the M2-F1, release the tow-line, and get about ten seconds of free flight before the flare landing.
Using the new control system, the M2-F1 handled well, both on and off tow in flight. Milt Thompson seemed to be happy with the control system. Neither Ryan nor Smith ever suggested later that we go back to the original control system. Not being an "I-told-you-so" sort of guy, I never again brought up the topic. And never again did  we discuss control-rigging within the group, other than how to reduce stick forces with aft stick positions.
The Pontiac towed the M2-F1 for the first time on 1 March 1963, and before April was over, it had towed it a total of 48 times. While the Pontiac was prominently a part of the M2-F1 adventure, it was no secret that the car didn't exactly resemble the usual flight-line vehicle. According to Whitey, whenever someone from NASA Headquarters was visiting the Flight Research Center, Paul Bikle would slip away momentarily to phone him, telling him to hide the car. Whitey would pull the Pontiac behind a shed and throw a cover over it, the Pontiac "grounded" until the visitor left. 6
What happened to the NASA muscle car once the M2-F1 program ended? Near the end of 1963, the Pontiac was shipped to NASA Langley Research Center in Virginia and used in tests at Wallops Island. There was some regret expressed at the NASA Flight Research Center when the Pontiac left, fairly much captured in a comment printed at the time in the X-Press, the NASA newspaper at Edwards Air Force Base: "No longer can we drive along the lakebed and pass the airplanes in flight." 7
1 Stephan Wilkinson, "The Legacy of the Lifting Body," Air & Space (April/May 1991), p. 51; Hallion, On the Frontier, p. 149
2 Hallion, On the Frontier, p. 150.
3 Milton O. Thompson, At the Edge of Space: The X-15 Flight Program (Washington, D.C.: Smithsonian Institution Press, 1992), p. 52.
4 For other details on the Pontiac and its modification, see Hallion, On the Frontier, pp. 150-151; Wilkinson, "Legacy of the Lifting Body," p. 54. For some details about ordering the Pontiac from the factory, intvw., Walter Whiteside by Darlene Lister, 21-22 June 1996. Both Hallion and Wilkinson identify the source of the modifications to the Pontiac as Mickey Thompson's shop. However, in an interview with Robert G. Hoey and Betty J. Love on 22 July 1994, Walter Whiteside was adamant that the source was Bill Straup's shop and that Mickey Thompson was the source for only the wheels and tires. This last interview is the source for several of the details in the present narrative.
5 Whiteside intvw. for incident with the pilot.
6 Whiteside intvw for the grounding of the Pontiac.
7 As quoted in Hallion, On the Frontier, p. 150n.