Authors: David Hitt,Heather R. Smith
Astronaut Bonnie Dunbar was still an undergraduate student at the University of Washington during the early portions of shuttle’s development, and she worked with the school’s dean of ceramics engineering, who had received a grant to work on the tiles for the shuttle’s thermal protection system.
’s earlier manned spacecraft had used ablative heat shields, which absorbed heat by burning up, protecting the rest of the vehicle. Such a system was simple and effective, but for the new, reusable Space Shuttle,
wanted a reusable heat shield, one that could protect the vehicle without itself being destroyed. The solution that was settled upon involved a vast collection of tiles and “blankets” covering the underside of the orbiter and other areas of the vehicle that would be exposed to extreme temperatures.
A worker removes a tile as part of routine maintenance activities on the orbiter fleet. Courtesy
“First of all, tiles are a ceramic material, so by definition they’re brittle,” Dunbar said.
But the reason they have an advantage over metals is that they don’t expand ten times over their thermal exposure range. It’s called the coefficient of thermal expansion. Also, they are an insulator; they don’t conduct heat. We looked at metals, or what they call refractory metal skins, and there are two disadvantages. You still have to insulate behind them, because metals conduct heat. The other is that when you go from room temperature, let’s say seventy-five degrees Fahrenheit, to twenty-three hundred [degrees], you have a large growth. It’s like your cookie pans, I guess, in the oven. So the airframe would distort. The ceramic materials [have] very small thermal coefficients of expansion, ten to the negative sixth, so you’re not going to see a lot of deformation. Also you could, on a very
low density tile, expose the surface to twenty-three hundred degrees Fahrenheit, and the backface, three inches deep, would not see even close to that, less than a couple hundred degrees, till after you’re on the ground. It’s a very slow coefficient of thermal expansion and heat transfer. So ceramics had a definite advantage. We knew that from the work we’d done in the sixties, and in fact, ceramics were already being used as the heat shields on nose cones for missiles and so forth.
So the next big challenge was to put them in a low-density, lightweight form that could be applied to the outside of a vehicle. Apollo vehicles, Gemini, Mercury, were all covered by ablators, which meant that they burned up on the reentry to the Earth’s atmosphere and could not be reused. The tiles were meant to be reusable. They didn’t deform. They didn’t change their chemistry. We had to, though, shape them so that they were the shape of an airplane, so we had all the aerodynamic features there. So we sort of did a little reverse engineering, in that we said, “Okay, here’s what the shuttle looks like; got to maintain that shape. Here’s how hot it gets from the nose to the tail. Most of the heat’s at the nose, on the nose cone, and the leading edges of the wings. We want to make sure the aluminum substructure doesn’t get over 350 degrees Fahrenheit; that’s when it starts to change shape. So how thick does the tile have to be?” So we used all those limits and constraints, then you’d use the computer . . . to calculate how thick each tile had to be. Then we started looking at, well, okay, how big should each tile have to be? Could I just put large sheets of tile on there?
Well, we started looking at what the structure does during launch, and now we’re getting to something called vibroacoustics. There’s a lot of force pressure on the vehicle, a lot of noise, if you will, generated into the structure, and it vibrates. We calculated that if we put a foot-by-foot piece of tile on there, the vibration would actually break it up into six-by-six-inch pieces. We said, “Well, we’ll design it six by six.” So you’ll see most tiles are six by six. Now how close do you put them? We thought, well, you can’t get them too close, because during that vibration they’ll beat each other to death, because they’re covered with a glaze. You’ve got silicon dioxide fibers that are made into very low mass tiles, nine pounds per cubic feet, or twenty-two pounds, and to ensure they don’t erode in the airstream when you reenter, they’re covered with a ceramic glaze. So that’s also brittle, so you can’t get them too close or they’ll break the glaze. You can’t get them too far apart or, during reentry, the plasma flow will penetrate down in those gaps and could melt the aluminum. So that’s called gap or plasma intrusion. So that then constrained what we called the gap. Then from tile to tile,
how high one was compared to the next one, we called step. That became important because if you had too large a step towards the leading edge of the wing, that would disturb the boundary layer, and you would go up the plasma, and instead of having smooth layers, it would start to transition to turbulent, from laminar to turbulent, and turbulent results in higher heating. So that controlled the step. So gap and step were very important to that as well.
“Those were all challenges,” Dunbar said. “We depended on advances in computerized machining capabilities, wind tunnel work with models to help us determine the requirements on step, the manufacturing, just everything. Firing a tile, a certain temperature and time was important to maintaining its geometry. . . . It’s, I think, a real tribute to the program that if you look at follow-on programs, even in
but also in Japan or in Europe or even the Russians, who built the Buran [Soviet shuttle], you’ll find that the system on the surface is very similar to the shuttle tile system. It was a good solution.”
Dunbar said that working on the shuttle during that early time was an exciting opportunity.
This was the next-generation vehicle. Not only was it next generation, it was . . . “transformational” is the word we use now. If you think about it, everything to that point was one use only. Couldn’t bring any mass back. We sent a lot of things into orbit that we had to test and leave there, and it became a shooting star, coming back to Earth. So this transformed our ability to do research. It’s why we have a space station now. We not only learned from Skylab, but we flew [on] Spacelab countless research projects that we could bring back to Earth, get the results out, diagnose problems with equipment. I think it saved the government billions of dollars, because we didn’t throw it away each time. So it was exciting, and we knew what it could do. New technology. It was leading edge on not only the thermal protection systems, but it was the first fully fly-by-wire vehicle, in terms of the computers and the flight control system. The main engines were also a pathfinder as well, and so it was exciting, even if it delayed till ’81. If you think about it, we baselined it to the contractor, to Rockwell, in 1972, I believe. So nine years later we have a vehicle, a reusable vehicle, flying.
Astronaut Terry Hart was the Astronaut Office’s representative in the development of the Space Shuttle main engines.
Since I had a technical background, mostly mechanical engineering, John Young had asked me to follow the main engine development. This was a couple of years before
1. In fact, it was ironic that we showed up [as
astronauts] in ’78, and everyone said we’re one year away from the first shuttle launch, and two years later, we were still one year away from the first shuttle launch, and it was really because of two main areas of technical difficulty. The main engine development was somewhat problematic, with some turbo pump failures that they’d had on the test stand, and the tiles. We had difficulty with the tiles being bonded on properly and staying on. But the main engine was one that John Young wanted me to follow for him, and so I spent a lot of time going back and forth to [Marshall Space Flight Center in] Huntsville [Alabama] and to
, the National Space Technology Laboratories, in Bay St. Louis [Mississippi, currently called the
John C. Stennis Space Center], where
tested the engines. And Huntsville, of course, was where the program office was for the main engines. And that was very exciting. I mean, I was like a kid in a candy store, in the sense that a mechanical engineer being able to kibitz in this technology, with the tremendous power of the fuel pumps and the oxidizer pumps, and the whole engine design, I thought, was just phenomenal. The hard part of that job was when we had failures on the test stand, which were, unfortunately, too frequent. I’d get the pleasure of standing up in front of John Young and the rest of the astronauts on Monday morning to explain what happened. And, of course, everyone was always very disappointed, because we knew this was setting back the first launch and it was a jeopardy to the whole program. But we got through that, and the engines have done extremely well all through the program here, where it was always thought to be the weak link in the design.
Astronaut Don Lind was involved in the early planning and development of the remote manipulator system, the shuttle’s robot arm.
I guess the first significant assignment I had [for the shuttle] was in developing the control system for the remote manipulator system, the
. In the hinge line of the cargo bay doors, there is an arm that’s articulated pretty much like the human arm. It’s about as long as two telephone poles, and it’s designed for deploying and retrieving satellites. Again, somebody had to worry about the operational considerations of that arm. It was built by the Canadians with the agreement through the [U.S.] State Department, and I was assigned to work on that. So I made a lot of trips into Canada to work with those people. The peo
ple who were actually building the hardware were very, very compatible, very easy to work with, and we had a very nice working relationship.
Lind contributed to the development of the three different coordinate systems that were going to be built into the arm’s software.
One coordinate system, obviously, applied when you’re looking out of the window into the cargo bay, and so you want to work in that coordinate system. If you wanted the arm to move away from you, you pushed the hand controller away from you. Also, if you’re trying to grasp a satellite up over your head and you’re looking with the
camera down the fingers at the end of the arm, which is called the end effector, and you want to move straight along the direction the fingers are pointing, you don’t want to have to try to figure out which way you should go, so you shift to a totally different coordinate system. So if you’re looking in the
picture with the camera that’s mounted right above the end effector, you want to push the hand controller straight forward. You want it to move straight forward in the television picture.
Lind also helped answer the question of how the hand controllers were to be configured.
We wanted hand controllers where the translation [movement] motion would be done by one hand controller, which we decided would be the left hand, and the rotational motion controlled by a hand controller which would be handled with the right hand. We decided, as a joint decision, that the hand controller for translation should be a square knob.
Then I said, “Now, remember you’re floating. You’re floating, so you’ve got to hang on to something while you’re translating, and you don’t want your bobbing around to affect the hand controller. So you need to put a square bracket around it so you can hold on to the bracket with your little finger and can use the hand controller.” “Oh yeah, we hadn’t thought of that. Well, how big do you want it to be?” We actually measured my hand and designed the controller and bracket to the physical dimensions of my hand. Obviously, when you make a decision like that, then you have five other astronauts check it out, and they say, “Yeah, that was a really good decision.” I didn’t want the hand controller for the right hand to be mounted square on the bulkhead, because the relaxed position of your arm is not at a square angle; it’s drooping down to the side. And I wanted that position to be the no-rotation position. We set up a simulation, and I stood
up there, and they measured the angle of my arm and then built a bracket to mount that hand controller just exactly the way my arm relaxed. And again, we had several other astronauts check it, and they said yes, that was a fine thing. So the hand controllers were literally fine-tuned to my design.
Other people were worrying about the software, how to implement these coordinate systems. Other people were doing all the very sophisticated engineering. But the human factor was my responsibility, and basically it was a very pleasant experience to work with the Canadians, with one exception. The arm has two joints: like the elbow, and like the shoulder; one degree of freedom in the elbow, two in the shoulder, and three degrees of freedom in the wrist, so there are three literal components to the wrist junction. They had mounted the camera on the middle one. As you maneuver in certain ways, the wrist has to compensate for the rotations of the other joints, and every once in a while the
picture would simply rotate. Not that anything had actually rotated, but the wrist was compensating. I said, “That’s unacceptable.” They said, “No, no, no, no, it has to be there. That’s the cheapest place to put it.” The engineers were all in agreement that this was a mistake, because you could lose a satellite when suddenly the picture rotates and nothing really has happened. But the management people said, “This meets our letter of intent with the State Department. We’re not going to change it.” So in one meeting I had to be very unpleasant. I said, “Now, gentlemen, if we ever lose a satellite because of this unnatural rotation, I will personally hold a press conference and say that you had been warned, and it’s the Canadians’ fault.” They looked at me like, “Ooh, you’re nasty.” At the next meeting, they said, “Well, we’ll change it, and it doesn’t cost as much as we thought in the first place.” Usually you could get good cooperation, but occasionally, particularly with people up in the bureaucratic levels, you had to be a little bit pushy. I try not to be pushy, but that’s one time I did.