HOUSTON - As NASA prepares for its final service call to the Hubble Space Telescope, it's also preparing for something never attempted in the history of the shuttle program: a rescue operation so dramatic that Hollywood would be hard-pressed to come up with a more outlandish plot.
If the Hubble repair crew due for liftoff on Monday got into the deepest sort of orbital trouble, yet another shuttle would have to be launched into orbit as little as a week later. NASA hasn’t launched two piloted spacecraft so close together in more than 40 years. But that's just the first act of the drama.
The rescue shuttle, Endeavour, would have to pull within about two dozen yards of the stranded shuttle Atlantis, and then help Atlantis' crew members make their way across a lifeline to refuge. Then Endeavour, full to capacity, would have to leave Hubble as well as Atlantis behind and return home — but not before Atlantis' controls are set for a self-destruct sequence.
The rescue mission, known as STS-400, would be NASA's last resort for saving the lives of Atlantis' astronauts in case of emergency. If Atlantis suffers irreparable damage to its thermal protection system — perhaps during ascent, perhaps from a space debris impact, perhaps from other less likely but not impossible hazards — it would no longer be able to return safely to Earth.
Because Atlantis is in an orbit different from that of the international space station, it wouldn't be able to reach that safe haven, even though the station will periodically zoom below the shuttle. The only hope for survival would be STS-400’s arrival.
STS-400 would be arguably the most perilous journey ever planned for space travelers. And no matter what the outcome, the mission would probably bring the 28-year space shuttle program to an early end.
The possibility of a rescue mission is the reason why Endeavour and Atlantis are both sitting on launch pads at NASA's Kennedy Space Center — a double vision that's been seen only a handful of times in the past and almost certainly will never be seen again.
Two crews are on alert for launch: The seven astronauts for the final Hubble servicing mission, known as STS-125, and the four astronauts for the STS-400 rescue mission. When STS-125 launches, the countdown for STS-400 will be at L-minus-7 days and counting. As STS-125 proceeds, the STS-400 countdown continues to L-minus-3 days, and then enters a "hold" just before fueling.
In a perfect world, the STS-400 team would just mark time until Atlantis heads back to Earth, after which Endeavour would be put back into preparation for a flight to the space station in mid-June. But if STS-400 is needed — and the need might not be discovered until the final few days of the STS-125 mission — the countdown would resume, and Endeavour would launch three days later. A day after that, using an abbreviated rendezvous path, it would be hovering back to back, 75 feet (23 meters) from Atlantis.
Already on board are the extra spacesuits and other equipment needed to ferry STS-125's seven astronauts across the gap into Endeavour. That could take two days, maybe three. Just before he left, Atlantis commander Scott Altman would configure the shuttle's flight deck so that ground controllers could bring it down safely into the ocean by remote control. Then it would be time to check the rescue ship itself for damage, do any necessary repair work and head back to Earth.
The rescue operation will almost certainly never be needed. But if it is needed, all the preparations and all the training must already have been performed. There just won’t be time to throw it together on the spot.
Preparing for possibilities
In the wake of 2003's Columbia catastrophe, NASA has always designated one shuttle crew to rescue another crew in the event of an emergency, so the concept itself is no big deal. But on every previous shuttle mission, and on all remaining shuttle missions after this one, a stranded crew could hold out aboard the international space station. There, air and food and power supplies give them a good chance of lasting 70 or 80 days before needing a pickup.
Consequently, the designated rescue vehicle did not need to be standing by for quick launch. Usually it was not even "stacked" to its external tank and solid-rocket boosters. If the mission ever needed to be called up, ground processing teams could go to double shifts and prepare it in time.
Not this time. Like the doomed Columbia, Atlantis will be following an orbital path that makes reaching the shelter of the space station impossible. The only supplies the crew can use to extend their lives and await rescue are those that they bring along with them.
So for this mission only, the potential urgency of the situation requires the rescue ship to be ready to go within a few days of a launch decision. The last time NASA had two different manned space vehicles in a similar situation was the dual flight of Gemini 7 and Gemini 6 in December 1965.
Off-and-on Hubble mission
After the loss of Columbia, NASA looked into whether the next-in-line shuttle mission — then about two months in the future — could have been rushed into space within two weeks. It would have been practically impossible, they concluded, and extremely risky to the second vehicle and its crew.
So when a rescue capability was decreed mandatory for subsequent shuttle mission, the lone non-station mission — the Hubble repair — was left without a plausible safety net and thus was canceled.
Engineers at the Kennedy Space Center didn't give up on a rescue plan, however. If they couldn’t rush the regularly scheduled next flight to launch, perhaps they could give special treatment to the next mission hardware that would allow it — this one time only — to be pre-positioned. First they convinced themselves it could be done, and then they convinced NASA Administrator Mike Griffin. The Hubble mission was back on the schedule.
“We’d had plenty of time to work out all the details needed for processing two vehicles,” said Angie Brewer, flow director for Atlantis' STS-125 mission. “The main thing was to be smart with the resources, and to schedule tests so they didn’t conflict with tests for the other vehicle.”
Atlantis and Endeavour have different processing teams in separate wings of the Orbiter Processing Facility at Kennedy Space Center. They followed the standard countdown plans on schedule. Resources that they shared, such as special testing and servicing facilities and teams, had to be “deconflicted,” in NASA jargon. But there was adequate time to lay out a schedule for that, and it all worked out.
"They worked smart and had a plan in advance," said Candrea Thomas, a spokeswoman at NASA's Kennedy Space Center. "There was no more overtime than usual."
The reasons why Atlantis' crew can't reach safe haven at the space station — or, for that matter, why Columbia's astronauts couldn't have done so even if they knew that their ship was mortally wounded — have not always been explained well. But the principles are fairly straightforward. Understanding them opens up an appreciation of some benefits of the different orbits, as well as the inconveniences they can cause.
Shuttles, like other satellites orbiting above the atmosphere, stay in orbit by virtue of their tremendous speed: about 25,000 feet per second (17,000 mph or 27,000 kilometers per hour). Gravity still pulls them toward Earth’s center — they still "fall down." But they are moving forward so fast that Earth’s curvature drops the ground out from under them. They fly over the horizon and just keep falling.
Changing the direction of the satellite’s flight path requires sideways thrust. The "orbital plane" can be shifted by using rocket power. It’s just very costly, because the added velocity is usually so tiny compared with the inherent velocity of the orbiting object.
A boost of 1 foot per second to the side results in an orbit that diverges about 400 feet from the previous flight path. To turn the orbit by as much as one degree (about 70 miles) requires an impulse of about 800 feet per second, which is more than enough to return to Earth, or fly hundreds of miles higher.
So when one object orbits at an angle of 28 degrees to the equator (which is typical for a due-east shuttle launch from Kennedy Space Center), and another object is orbiting at 52 degrees (where the space station flies), even the best case demands a plane change of 24 degrees. and hence a thrust almost as powerful as the original blastoff from Earth’s surface.
Neither Columbia in 2003 (in a slightly higher orbital inclination) nor Atlantis in 2009 has anywhere near the rocket power to make that "left turn in space" and match orbits with the space station — no matter how often their paths criss-cross one over the other. It’s almost close enough to wave, but never to shake hands.
But where orbital planes taketh away, they also giveth. If it's required, STS-400 would get a valuable bonus for its rescue mission. Blasting off from Earth, from a point that is being carried eastwards by Earth’s rotation, it is much easier to reach a Hubble-type orbit.
First of all, the shuttle can take full advantage of the "kick in the pants" it starts out with by launching directly into the direction of Earth’s rotation. Secondly, there's a longer launch opportunity because the flight paths for STS-400 and STS-125 stay closer together for a longer time. The "launch window" for liftoff to the station is usually only four or five minutes, but the corresponding window for a launch to Hubble (and Atlantis, if a rescue is needed) is almost an hour long.
Some days a countdown can be terminated by a serious technical flaw, or meteorological messiness. But other countdowns can be (and have been) scrubbed because of a brief snag — say, a thundercloud, or an intruding pleasure boat in an offshore impact zone, or a software data set that needs to be quickly reloaded. Such problems couldn't be fixed in five minutes, but could have been in 50 minutes.
So the rules of orbital mechanics are what make the STS-400 rescue mission necessary. But the same rules make the mission a little easier than they might otherwise have been.
NBC News space analyst James Oberg spent 22 years at NASA's Johnson Space Center as a Mission Control operator and an orbital designer.