Why We Miss the X-15

To commemorate the 40th anniversary of the fastest flight by an airplane—the X-15’s rocket-powered, Mach 6.7 dash on October 3, 1967—U.S. Air Force Chief Scientist Mark Lewis and former Air Force historian Richard Hallion recently met with Air & Space editor Linda Shiner at the National Air and Space Museum to talk about the X-15 program and its legacy.

A&S: It’s been 40 years since the X-15 flew 4,520 mph, a record that’s never been broken. Why did we stop there? Why didn’t we continue to build faster and faster research aircraft?

Lewis: Embedded within that question is the question of whether we should have continued the X-15 or X-15-like programs. And to that the answer is clearly "yes." It’s one of the great tragedies in aerospace that we didn’t. Look at the record of the X-15 program: 199 flights—200, if you count the time an X-15 blew in half during a rocket engine ground test and shot Scott Crossfield forward a few feet. A very clever, well-conceived series of experiments and series of measurements. It was true science. The X-15 produced over 700 technical papers.

The science and the engineering that came out of the X-15 program were absolutely phenomenal. It was bold, it was daring, but it was very, very well grounded in real science, and that’s a very fine line to walk and it’s something I deal with almost every day. We want to do bold, great things, but we shouldn’t attempt to violate the laws of physics.

The X-15 was a culmination of our notion of flight test, the experiments in the air, pushing the envelope, but doing it in a logical, evolving manner.

Hallion: It was a tremendous focal point for activity. It was where you could actually take theory, and ground test, and simulation, and you could match all those against a real world flying system. Where the measurements you got were real world measurements. You didn’t have to figure out what are the scaling factors. You didn’t have to calculate from approximate data. You were getting real data.

A&S: And this data was used in the design of the space shuttle.

Hallion: The X-15 gave us the exact same kind of mission profile coming back in, once it got to Mach 6 and was headed downhill. There was a difference because the X-15 was not a delta [wing design] so you have differences in the actual approach and landing. But the low lift-to-drag ratio approach was identical, the workload on the pilot was identical. It gave us an exact indication of what would be involved in the way of cues, flight path management, energy management—everything came out of the X-15.

But back to the original question, I think where we got off track in aeronautics was the cancellation in 1963 of the Manned Orbiting Laboratory program, we began a pattern of fits and starts that has continued to the present day. We replaced DynaSoar with Manned Orbiting Laboratory, then we didn’t build the Manned Orbiting Laboratory. Had we fulfilled Manned Orbiting Laboratory, several things would have happened. First of all, we would have achieved a space station at a much earlier date, much like the way the Soviet Union achieved a space station. The second thing is you would have had to have had a launch vehicle to get a team of astronauts up to that space station, and we actually had that system: It was called Gemini.

Beyond Gemini and the Titan launch vehicle, there were plans to take a lifting-body shape, the so-called SV-5 shape, which was an Air Force-Martin program. Then NASA came on board as well, and NASA had plans themselves to look at a lifting-body shape that they called the M-2. Northrop was willing for a price of $200 million to build a piloted, demonstration orbiting vehicle. This would have led to a routinely operating hypersonic reentry vehicle that we could have used to support operations of the Manned Orbiting Laboratory.

We turned our back on that as well. When you had then the fulfillment of Apollo in 1969 coming after the shut down of the X-15 program [which ended in 1968], and you had the cancellation of any plans to extend the X-15 program to use a delta-wing configuration and to make use of an experimental scramjet propulsion system or at least to test an experimental scramjet propulsion system—you had the stage set for the plateauing of hypersonic research.

In the 1970s, we had two opportunities to reinvigorate our hypersonics research. One was a program called the National Hypersonic Flight Research facility, which would have built on lifting-body experience to give us a manned hypersonic demonstrator up to about Mach 7. And the second was in support of the space shuttle. That was a proposal by a group of engineers within NASA at the Dryden Flight Research Center to develop a subscale shuttle that could have been flown out to Mach 5 to 6 to collect reentry data in support of the shuttle program. That would also have given us the ability to use that vehicle at some point for plain old hypersonic research, to evaluate materials, systems, propulsion concepts—things of that sort. But when we turned our back on that, and the bills started to come due to attempt to meet the anticipated launch rate for the space shuttle, we simply didn’t have the money. And so the next big step, the NASP, had its own challenges.

A&S: Was the X-15 data used in that X-30 program, the National Aerospace Plane?

Lewis: It’s easy to bash the X-30 program. They spent a ton of money, and there’s been a bit of revisionist history, of people looking back at the X-30 and saying "Oh we got all this great stuff. We got new materials, we got this, we got that." The reality is the X-30 was very much the antithesis, in my mind, of the X-15. It was "Let’s not do a logical, reasoned science effort. Let’s jump to Mach 25 the first time out of the barn. We’re going to build it, we’re going to hop in, we’re going to fly it, it’s going to work. We’re not going to do any ground testing because we have all the computers we need to simulate anything we’d ever need to simulate." And they were completely wrong. It’s still wrong to this day.

Hallion: There was over-enthusiasm about computational fluid dynamics. It was going to replace all ground test facilities and much of flight test facilities; you could do it all by crunching numbers.

Lewis: We had people saying that the X-30 marked the end of the wind tunnel. We no longer needed wind tunnels because we could simulate everything on a computer. And you can contrast it with work today. We’ve gotten into this mode where we don’t really do envelope expansion. We keep shooting for the Next Greatest Thing.

So look at where we are with today’s programs. I have a personal favorite: the Air Force’s X-51 [a small-scale hypersonic vehicle]. It draws as much from the X-15 as from any other program in terms of the materials that are being used and our understanding of the basic physics.

It’s an Inconel structure [as was the X-15]. It uses some space shuttle materials, but our understanding, for instance, of the airflow around a very sharp leading edge—that didn’t come out of the Shuttle program; it came out of the X-15.

A&S: And how did that information, that data, from the X-15 program get into the hands of the engineers working on the X-51?

Lewis: It comes out of the scientific literature. I’ll give you an example—in terms of fluid mechanics. The fluid physics associated with the fin on the X-15 prove that the people who designed the aircraft knew exactly what they were doing.

That fin is basically just a wedge. Why a wedge? At low speed, an airfoil shape is very important. As you reach higher and higher speeds, it actually becomes less important. Here’s what that wedge does for you: It turns out that when you’re flying at high speed, you get a shock wave forming on either side of the wedge. It’s almost like a wake coming off the bow of a ship, just as you get a shock wave off the front of the aircraft. If you deflect that wedge—say you deflect it a degree; you gain a degree on one side, and lose a degree on the other—at high speed, the differential pressure caused by a change in angle of the shock waves is about three times greater than at low speeds. So that becomes an incredibly effective control surface. It’s not an effective control surface at low speed, but it’s extremely effective at high speed. Shock waves are inherently non-linear. That means you change the angle by a certain amount and the change in properties doesn’t scale in linear proportion.

Hallion: Put another way, if you were to use the conventional thickness ratio for that vertical fin, for the same degree of directional stability, you’d have to have a much larger fin. It would have to be huge. It would be so huge it would be structurally impractical. This in a way is fooling the air into thinking that the fin is much, much bigger than it actually is.

Lewis: It was brilliant aerodynamics.

Hallion: It really was. It was a very creative design.

Lewis: I have to say, as much as I love the space shuttle, the space shuttle is a horrible aerodynamic design. The space shuttle was designed so it wouldn’t burn up on reentry, and you get these incredible aerodynamic compromises in its design.

Hallion: It’s a railroad car. You can make a model of the space shuttle by taking a box car and putting on a nose cap, a tail cap with the engines, delta wings, vertical fin, and the Orbital Maneuvering System pods. And you’ve got yourself a space shuttle, because the core of it is this big box, a 65,000-pound payload bay. But it doesn’t really have design elegance.

Lewis: There was another phenomenon on the X-15. There was a famous flight of an X-15 when they were testing the airflow around a dummy air-breathing engine, and the engine burned off. And we understand in gory detail now why it burned off. It was a shock interaction that at the time we really didn’t understand very well.

Hallion: It’s called a shock-shock.

Lewis: Basically, two shocks intersect. One shock wave hits another shock wave, and when they interact, there is a very, very hot jet of gas. And this is why it’s important: We now worry about that when we design a scramjet engine [a Supersonic Combustion ramjet]. In the flight of a scramjet, there will be a shock wave coming off the nose of the aircraft and another shock wave formed from its own lip, or inlet. And when those two shock waves meet, if we’re not careful, we could get the same style of interaction. So one of the very first design principles in selecting the inlet for a scramjet or a high-speed ramjet is "no shocks interact."

A&S: Another of the X-15’s innovations was the all-moving tail. Why was that important at high speeds?

Hallion: The X-15 had a rolling tail. It not only had a pivoting surface that was movable in pitch; it also had differential movements so you could use it as an aileron [to roll the aircraft]. And that’s why it was called a "rolling tail." On the wing, the inset aileron you might think would be used for roll control were actually flaps that would dramatically increase the lift on final approach.

The best configuration for [hypersonic] vehicles tend to be delta-wing blended-body configurations and we were headed down that road. Had we not lost the Number 3 [X-15] airplane [in a November 1967 crash that also took the life of pilot Michael Adams], there was every expectation that it would have been modified as a delta-wing aircraft.

Lewis: There’s an interesting comparison with the Orbital Sciences Antonio Elias, drew from the X-15 dimensions. The wingspan was set by what the B-52 could carry. That’s another example of the X-15 legacy.

Hallion: One of the things the program engineers discovered when the X-15 returned from a mission is that occasionally you’d get dimpling on the structure. Aircraft experience structural loads because they’re generating lift and experiencing other forces. But when you heat the structure, it experiences a whole different kind of structural load. These thermal loads would occasionally buckle the skin near rivets or where there were plates joining. And this little buckle would act like an inlet. At hypersonic speeds, air would ram through that inlet, and damage could occur downstream. So if you look at the skin of the X-15, you’ll see some interesting things. On the leading edge of the wing, you’ll see things that look like little cap strips. Those cap strips were used to prevent hypersonic leakage into the structure, based on what people discovered from actual flight test results.

Looking at the X-15 here from the second floor [of the National Air and Space Museum] under the lights, you can see a sort of beer canning effect. You see that this vehicle really has been through it; it’s withstood that thermal hypersonic environment, and it has that gun-blued look to it. When people make models of the X-15, they make a neat, flat black model, but it’s really not. It’s a heat-treated metal.

You take a look at the gaps and voids [in the skin] of the X-15, which were designed to accommodate the anticipated thermal expansion of the material. Where the X-15 wing joins the fuselage, there’s a significant gap that starts at the leading edge of the wing and reaches back almost to the quarter chord. That’s an open slot all the way back to the spar. Now this is a very robust wing; it’s hard to relate this to the conventional rib and spar construction because the whole thing is almost solid. But the point that I’d make is we have gotten away from the notion of test as test. We use this terrible term "technology demonstrator." It is a term we should walk away from. Technology demonstration is what I do when I go into a high school science class; it’s where you have repeatable experiments through which nothing new is learned. We’re standing in front of the X-15, which is a vehicle of actual flight test. In flight test, you go out and you explore unknowns. You’re not demonstrating anything; you’re learning things. And if you’re going to learn things, you need to be able to study what you’ve accomplished. We have tantalizing bits of data from some of these programs where we don’t actually know what happened in the last seconds of a flight or the last seconds of an experiment. We don’t have the complete understanding that we would only gain if we were actually recovering something on the ground. We need to look at that artifact and study and figure out where we go next.

Lewis: Several years ago Donald Rumsfield got some bad press for using the term "unknown unknown," but engineers use that term all the time. There are unknown unknowns and known unknowns.

Engineers working on the X-15 program faced both types of unknowns. Towards the end of the program, there was experimentation with a variety of high-temperature ablative coatings. And there was a question, "What happens if the ablative coatings out-gas during the reentry and fog up the windows? How’s the pilot going to land?" They didn’t know how bad the effect would be. But the program was characterized by very clever engineering, and they found a very elegant solution. They simply put an iris over one of the windows. And the pilot flew with one window; when it fogged up, he opened the iris and looked out the other window—and that’s how he landed.

Hallion: And that’s another way that the X-15 made a contribution to future programs. The X-15 became a test bed for other systems and concepts. We learned with that spray-on ablator that that was not a good way to configure a reentry vehicle for a human presence in space. Now there may be some future time when we find that we can do that, but this countered what people thought at the time. There was the idea that you could build a really cheap, lightweight, conventional-material vehicle, maybe even out of aluminum, and you could spray it with this ablator, and you could fly it, and you could refurbish the ablator, and sha-zam! you’d be right back in business again. This demonstrated that you weren’t able to do that.

The X-15 so quickly demonstrated its utility that people started using it for things that it was not intended for. They used it to collect micrometeorites, to collect imaging from a near-space environment. You see this by looking at the pods, which were not original to the basic design but were added later in the flight test program.