How a space docking feels can depend on which side of the interface you’re facing—whether you’re the docker or the dockee. But when the 100-ton shuttle Atlantis linked up with the 100-ton space station Mir in June 1995, neither crew had any doubt about what was happening.
Just before contact, with the two spacecraft perfectly aligned, the Atlantis crew had pushed a button to fire thrusters that gave them a last nudge into the Russian docking mechanism. For the shuttle astronauts, it was the noise of the thrusters more than anything that signalled their arrival at Mir. “You could hear the booming of the forward jets,” recalls Charlie Precourt, copilot of mission STS-71. The contact itself is “absolutely imperceptible,” says Kevin Chilton, who commanded the third shuttle-Mir docking mission nine months later. You know something’s happening from “all those cannons going off all around you,” but there was no bumping or jostling inside the shuttle cabin.
On the Russian side it was a different story.
The impact felt “like a big hug,” commander Vladimir Dezhurov recalls. “A real man’s hug.” The Mir began quivering, then calmed down. When the station finally stopped shaking, says Dezhurov, “we understood the docking had occurred.”
Over on Atlantis, where shock absorbers in the docking system dampened the force of impact, the mechanism “bounced like a baby carriage,” Precourt says, but the back-and-forth motion was too subtle to be sensed directly. “The only way we could tell there was any rebound at all was to look in the camera.”
The first orbital hook-up of U.S. and Russian spacecraft in two decades had come off without a hitch.
Docking has been part of the spaceflight repertoire for more than 30 years, and as often happens, NASA has made a complex and challenging operation look boring and routine. In practice it is anything but. Robert “Hoot” Gibson, who commanded Atlantis during the first shuttle-Mir mission, calls space docking “a cross between air-to-air refueling and a carrier landing.” When the two spacecraft are still at a distance, it seems easy. “But the closer you get, the tighter you control, and the smaller the allowable errors can be,” he says. With an unlucky combination of equipment problems and human error, things can go spectacularly wrong, and that’s reason enough to regard each space docking with apprehension and respect.
When Neil Armstrong completed the world’s first orbital docking, connecting his Gemini VIII capsule to an Agena target vehicle in 1966, his joy was soon overshadowed by a life-threatening out-of-control tumble that led to an emergency splashdown in the Pacific (the fault lay in a stuck thruster on the Gemini, not in the docking technique). On the very next flight, a shroud covering the target’s docking port failed to open fully, making docking impossible.
Docking problems frustrated Russia’s first space station mission in 1971 and nearly aborted NASA’s first Skylab mission two years later. When the Russians added the Kvant science module to the Mir station in 1987, an errant trash bag got stuck in the docking interface, preventing an airtight seal until spacewalking cosmonauts removed it. Other failures and close calls convinced both U.S. and Russian space engineers that nothing about space docking would ever become routine.
Bumping two large masses together in orbit without damaging or breaking anything makes for a tricky physics problem. Vehicles docking on Earth have at least some of their motion already constrained at the time of contact. Freight cars move along rails, ships float on water, even aircraft have aerodynamic stability. But in space, position and orientation can vary in all three dimensions, and can change at different rates. All these variables—Precourt calls orbital docking an “eight-degrees-of-freedom problem”—have to be controlled simultaneously to make sure the final contact happens within the mechanical limits of the docking hardware. On Earth we also encounter natural damping forces—friction, air and water resistance, the restraining forces of rails or cables. In space, all the energy has to be absorbed and damped out within the vehicles themselves.
Because these are problems imposed by the laws of physics, it’s no surprise that in the mid-1960s, U.S. and Russian engineers came up with essentially the same design for docking mechanisms. Both countries built systems that worked like this: The chase vehicle extended a long, stinger-like probe with capture latches at its tip. On the target vehicle was a cone-shaped receptacle. When the tip of the probe entered the wide end of the cone, it was naturally guided to the back, where another latch mechanism was waiting. The engagement of these two latches was called “soft docking.” The docking probe then retracted, drawing the two vehicles together so that facing rings could be latched together for a “hard dock.”
This was the basic design used for NASA’s Apollo lunar missions and for the Skylab space station. It also became standard for Soviet vehicles and, with one exception, has served all Soyuz, Progress, and science module dockings with Russian space stations to this day.
The inescapably “male” and “female” nature of the probe-drogue system has led to countless earthy jests by astronauts and cosmonauts over the years. The major drawback is equally obvious: Only mechanisms of different types can successfully mate. For short spaceflights this wasn’t a problem, since each vehicle could easily be outfitted with mission-specific hardware. But engineers knew that at some future point it would make sense to build spacecraft that could dock with any other vehicle in orbit.
The “androgynous” docking mechanism sprang from this anticipated requirement. When Nixonian détente thawed relations between Moscow and Washington in 1971-1972, the resulting plan for the symbolic Apollo-Soyuz orbital docking gave space engineers the opportunity to build and test an androgynous docking mechanism. The new design had an immediate political advantage: Neither the Russian nor the U.S. spacecraft would appear dominant. Though arguably for the wrong reasons, space engineers were enabled to do the right thing.
Based on preliminary sketches by virtuoso spacecraft inventor Caldwell Johnson (a self-made NASA engineer who had co-designed the Mercury capsule in the 1950s), as well as a symmetric ring-to-ring system designed at about the same time in Moscow, U.S. and Russian engineers—led by docking expert Vladimir Syromyatnikov— joined forces and came up with a new design. Each vehicle would have a docking ring with three open “petals” extending out from it. The petals were for alignment only: They fit slot-and-groove style between the petals of the other vehicle’s ring, so that the facing rings could fit together only in the prescribed way. During docking, the ring on the active vehicle (complete symmetry was sacrificed) would extend outward on shock absorbers and be rammed (slowly!) into the passive vehicle’s ring. The petals would then fit like fingers sliding together, and latches on the active vehicle’s petals would catch latches on the target docking ring. Finally, after the motion from initial contact was damped out, the extended ring retracted to pull the two vehicles closer together. At that point the heavy latches around both rings would engage to achieve a hard dock.
The new system worked fine on the one mission it flew (Apollo-Soyuz), and its advantages over the probe-drogue set-up immediately became clear. For one thing, the damping mechanism allowed it to handle much more massive vehicles. True, it demanded more accurate alignment from the pilots, but neither pilots nor engineers saw that as a problem.
By the time the Russians were designing the Mir space complex in the mid-1980s, they needed exactly this kind of system for the Buran space shuttle, which was to mate with the station. The shuttles were too massive for the probe-drogue design, and the Russians would now be using a variety of different docking combinations—Soyuz to Mir, Soyuz to Buran, and Buran to Mir. The androgynous system was the only one that could satisfy all these requirements.
The Russians called their design “APAS,” for “androgynous peripheral aggregate of docking” (“docking” in Russian is stykovka). They improved the Apollo-Soyuz design in several significant ways, most visibly by turning the guide petals inward rather than outward. The system was perfectly designed for the Buran-Mir dockings. But the Russian shuttle was scrapped before the system got the chance to prove itself.
Meanwhile, U.S. space designers had been developing their own docking mechanisms for the shuttle and the Freedom space station. The only principle guiding this complicated, clumsy system seemed to be that it not look like the Apollo-Soyuz design. By the early 1990s, however, the political winds had changed, and it was no longer unacceptable for Americans to acknowledge Russian space expertise. After a brief review, the Russian system designed for Buran-Mir was adopted for shuttle-Mir and the space station, with Rockwell and RSC Energia doing the modification work.
When Gibson and Precourt were tapped to fly the first docking mission, they knew they were in for a challenge. No space shuttle docking hardware had ever worked properly on its first attempt in orbit. The highly public embarrassments of failed first attempts to dock with the Solar Maximum satellite in 1984 and the Intelsat satellite in 1992 (both of which involved hardware carried by spacewalking astronauts, not vehicles), as well as several less publicized but equally frustrating failures with other space hardware, reminded everyone how easily things could go wrong.
Even after all the hardware had been analyzed and tested piece by piece, experienced engineers knew they weren’t finished. At the insistence of veteran space docker John Young, NASA added a special program for “end-to-end testing” at the Kennedy Space Center’s Orbiter Processing Facility, where the shuttle is still horizontal. The docking assembly was installed in the shuttle’s payload bay with all the flight hardware in place. Test engineers rigged up a mockup of the passive mechanism on Mir and lowered it by crane at docking speeds of only inches per second. They verified in the cockpit that the instrument panel performed as advertised throughout the whole sequence.
One value of these tests was to raise the crew’s comfort level with the post-contact damping process, the time between initial capture and hard docking, when the two giant vehicles would be only loosely joined together. During this time, Mir’s attitude control system is switched off so as not to introduce motions that could bend the docking mechanism. But even in this “free drift” mode, the Russians had worried that random twisting of the two large masses might never settle down. Noted Precourt: “This would prevent us from drawing the ring back in.”
Based on the ground tests, the crew came up with a solution. Precourt explains: “We interrupted the auto[matic] sequence at the first point we saw ring align, stayed there about a minute, waited until motion stopped, and then we retracted.” With the rings on Mir and the shuttle perfectly parallel, the hard dock could proceed.
Even though their hardware was different, the shuttle-Mir dockers knew they had much to learn from the previous generation of astronauts. Precourt spent time chatting with six-time spaceflight veteran John Young, now a special assistant to the Johnson Space Center director. Precourt was especially interested in the difference between simulation and reality. “In a simulator, a lot of the sensations aren’t there, but in flight you are subject to a lot of distractions,” he says. Young told him to trust the simulators, which was good advice—the crews who’ve docked with Mir say they are extremely faithful to the actual experience. If anything, says Precourt, the real flight “was a lot smoother than most of the sims, in terms of everything working.”
Before mission STS-71, the astronauts “flew” over 200 approaches in a variety of simulators. Docking with Mir requires a very slow closing speed—barely more than an inch per second during the final approach. It also demands great precision. The docking rings have to be parallel within two degrees in each axis, and the targets have to be aligned within three inches of each other. The astronauts have various tools to help them measure the alignment. A metal “stand-off cross” extends on a rod above and parallel to a black painted cross on the Mir target. If the crosses appear in TV views to line up perfectly, the pilot knows he’s on track. The TV cameras also have grid markings to make it easier for the astronauts to check their alignment.
One concern had been the disorienting view caused by the camera’s being at a distance from the pilot’s eyeballs. “You’re not looking at the real world,” explains Precourt. “It’s not like landing an airplane with a view straight out the front windshield.” It’s more like closing your eyes, holding your hands out, and trying to touch your fingertips, he says. But even though it took some getting used to during training, it turned out not to be a problem.
Gibson and Precourt, as well as every docking crew after them, learned in the simulators to hit the marks every time, even when jets and instruments and computers failed. On the STS-71 docking, the angular errors were measured in tenths of degrees, almost too small to be noticed. The arrival time was nearly perfect too: They were only two seconds off.
Experience has shown that on-time arrival doesn’t matter all that much. “I always argued against getting hung up on the docking time as if it were critical,” says Kevin Chilton. “I wanted to dock a minute later or a minute early just to show it’s not important.” He ended up docking “pretty much on time” anyway.
In fact, so far every docking has been a model of precision. “When you think about it,” says Precourt, “it’s pretty amazing that you’d have two vehicles flying in space that are subject to bending and moving, yet the relative position of the docking ports can be precisely known when we arrive.”
With at least five more shuttle-Mir missions planned, and with dockings to the international space station scheduled to begin in 1998, orbital docking is finally becoming, if not routine, then at least no cause for great anxiety. Engineers working on the space station have come up with a few modifications to the shuttle-Mir design but not many. They plan to fine-tune the orbiter’s damping mechanism to further reduce the energy transferred to the station at contact. The station also will have a few of the old-style probe-drogue ports, since a variety of Russian, American, European, and Japanese vehicles will have to dock with it.
Dockings have now taken place with four different configurations of the shuttle and Mir (approaching the Russian station, with all its protruding solar arrays, modules, and vehicles, is “like docking with a porcupine,” says STS-79 commander Bill Readdy). The STS-74 crew brought up a new docking module to attach to Mir last year, which provides greater flexibility and places the docking interface at a distance from the main station. This addition, plus the station’s different configuration and greater mass, may account for the fact that Mir crews are now feeling less of a jolt than Dezhurov and his companions experienced. Readdy says that when Atlantis pulled up to the docking port last September, Shannon Lucid and her cosmonaut crewmates hardly felt a thing.
The STS-74 astronauts even came up with a soundtrack to accompany all the slow, graceful maneuvers in space. A Strauss waltz had already been appropriated by Stanley Kubrick, and besides, it evoked Vienna, not Moscow. So Ken Cameron and his Atlantis crew went with Tchaikovsky’s “Swan Lake” for their final approach and docking.
Precourt is now back in Houston training to command another Mir rendezvous mission; he will be the first astronaut to make a second such docking. On his first trip to Mir he spent time with his cosmonaut hosts inside the attached Soyuz spacecraft, and he has been through the complete cosmonaut training program for Soyuz dockings to Mir. As it stands, he has the inside track to become the world’s most experienced space docker.
Still, he keeps worrying about what could go wrong, what might be done ahead of time to reduce the risk, and what he might have to do should an unforeseen problem arise in orbit. His biggest fears are shuttle failures that could cause a sudden increase in closing speed during final approach. He’s also thought out another failure scenario. “I’ve told folks that I really think we’re going to see a bounce-off,” he says. “At some point there’s going to be a mechanism that doesn’t work for us right. The Russians have had it happen to them.” He pauses thoughtfully. “I hope we’re adequately prepared to deal with that.”