james oberg logo



space shuttle



Getting back to Earth from outer space requires reliable hardware, as the Columbia tragedy of Feb 1 reminded us. But the return to Earth of the Russian ‘Soyuz TMA-1’ spacecraft on May 3 also made the point that reliable software can be almost as critical.

The three-ton landing capsule based on a decades-old design was carrying home the three crewmen of the International Space Station after five months in space. A new crew had blasted off on a new Soyuz the week before and had taken command of the station.

High anxiety surrounded the landing, both due to the fatal shuttle accident three months earlier and to the uncomfortable fact that this Soyuz had been equipped with new computers, new software, and new landing rockets. In the past, upgraded models of the Soyuz had always been tested on at least one unmanned flight before cosmonauts used them, but schedule and budget pressures forced the Russians to skip that step. Instead, they insisted, all new systems had been thoroughly tested piecemeal and in simulations on Earth.

So when contact was lost with the descending Soyuz about fifteen minutes before landing – the same point at which Columbia had disintegrated – an ominous ‘deja vu’ sensation gripped many workers, officials, and crew relatives at Moscow’s Mission Control. Waiting recovery teams in Kazakhstan searched the skies and the ground in vain. For two hours there was no word on the fate of the crew.

A software glitch on the new control computer had apparently confused the spacecraft’s navigation system, making the planned ‘steep glide’ approach to the aim point impossible. Instead of enduring a planned maximum G-force of 4 to 5 for about a minute, the crew suddenly faced twice that deceleration as the spacecraft plunged steeply into denser layers of the atmosphere before reaching the ground 450 kilometers short of the waiting rescue teams. The autopilot followed this so-called ‘ballistic trajectory’ by slowly rolling the capsule, neutralizing the small aerodynamic lift forces that a functioning computer could (and usually did) use to steer the spacecraft to its aim point much farther downrange.

At first, the spacecraft’s vendor insisted that an American crewman must have pushed the wrong button and ‘confused’ the guidance computer. But the Russian cosmonaut who was actually flying the Soyuz was adamant that no procedural errors had occurred. Halfway back to Earth, for no apparent reason, the computer had suddenly moded itself to ‘rendezvous mode’ and began searching for the space station as if to redock with it.

Piled atop this malfunction was an independent failure that added to the initial anxiety in Mission Control. As the main parachute deployed out of its container inside the Soyuz, one shroud line pulled out a wire antenna for radio communications with ground searchers. But that shroud line snapped, disconnecting the antenna and cutting off voice contact.

Russian space engineers have six months – until the planned return of the current station crew in the Soyuz TMA-2 spacecraft – to determine if these software and antenna incidents were generic design flaws or just one-time sloppy mistakes in the fabrication process. The two men now aboard ISS didn’t need this new worry about their own Soyuz while they concentrate on caretaking the orbital outpost.

The Soyuz was skimming horizontally across the edge of the upper atmosphere at close to 25,000 ft per second (Mach 25, or 25 times the speed of sound), using air drag to shed its tremendous velocity. Normal plans would have called for choosing an altitude where the air was just thick enough to slow the craft by about 150 ft per second every second. In terms of earth surface gravity, whose acceleration is the classic 32 ft per second per second, that gives a deceleration force of about five G’s.

But this flight plan requires the Soyuz to fly with the upper edge of its heat shield tilted slightly forward, to gain a small amount of ‘lift’ and keep from dropping into the lower, thicker layers which will slow it faster – and hence decelerate it more strongly. To accomplish this, the spacecraft was designed to have its ‘center of mass’ off center, closer to the outer hull above the heads of the crew. By rolling – rotating along its long axis – the heavy side ‘up’, the necessary lift-producing tilt can be created and controlled.

But this requires that the guidance computer recognize what direction is ‘up’ and where the spacecraft is in relation to its aim point far ahead. This is not a particularly complex calculation but must be done with high precision and reliability. Gemini, Apollo, and Soyuz spacecraft perfected this technique in the mid- to late-1960’s.

What happened this time was that the autopilot suddenly announced to the crew that it had forgotten where it was and which way it was headed. “The auto system switched to backup”, a NASA source told msnbc.com, “which surprised them”.

Bowersox was even more dramatic during an interview on the aircraft carrying him from the landing area back to Moscow. “The first thing we saw was signs on our displays that the entry was going to be off nominal,” Bowersox reported. “And when we saw those signs our eyes got very wide.”

The crew knew that without guidance commands, the autopilot would stabilize the spacecraft by a simple-minded backup procedure. It would send commands to steering thrusters to perform a slow roll, turning the spacecraft’s ‘heavy side’ continuously around the dial. This had the desired effect of ‘nulling out’ any now-unsteerable lift and follow a ‘ballistic’ descent.

But this also meant that without the lift to stretch their flight path, they would fall faster into thicker air. That in turn would impede their forward speed even more ‘aggressively’ (Bowersox’s word), resulting in a deceleration about twice as high as normal. They would also reach the ground far short of the planned landing zone.

Historically this is a very rare type of failure – several times between 1967 and 1975, but never afterwards. Suspicion immediately focused on this new Soyuz’s ‘improved’ guidance computer. “Ken suspects a software problem”, a NASA source told msnbc.com.

Software problems in the Soyuz guidance computer aren’t just a matter of landing randomly back on Earth – they are potentially fatal. In 1988, a confused guidance computer nearly jettisoned the Soyuz T-6’s rocket engine section while the crew was still in orbit, a malfunction that would have doomed the men to a slow death by suffocation. Only the alertness of one of the pilots detected and aborted the insane command.

This time, since Budarin, Bowersox, and Pettit knew that their backup descent profile was safe – if rougher – they kept their hands off the manual controls. “They didn’t do anything,” msnbc.com was told. “[They] just let the auto system control.”

Apparently the rest of the descent went as planned, and the parachutes and soft-landing engines did their job. As in about half of all Soyuz landings, the landing module wound up on its side, probably pulled over by a gust of wind in its parachute just at touchdown.

The three men, who knew they were far off course, were able to open the hatch themselves and get out – it’s a much easier drop to the ground when the capsule is on its side. They then waited two hours to be spotted by a search plane, and several hours more for the arrival of the first helicopter.

While Russian recovery teams struggled to locate the off-course spacecraft and its three crewmen, outside observers also faced a daunting search for what was really going on. Behind the blizzard of premature 'all-is-well’ announcements and the subsequent blasé official attitude – “Hey, they’re alive, aren’t they?” – to the alarming malfunction, and hindered by the Saturday-Sunday overnight hours when most standard information sources are nearly comatose, hard-core space watchers struggled to find out the real news, good or bad.

Along with a cadre of space program journalists, they mostly depended on a steady stream of reports from ‘NASA TV’. This is the satellite channel NASA usually uses for education, press conferences, and general ‘preaching-to-the-choir’ cheerleading. On this occasion, it came through magnificently as spokesmen quickly passed on raw reports they received from colleagues in Russia.


oberg corner piece

home | profile | articles | books | lectures | jim speaks | humor
links | email

Copyright 2010 James Oberg. All Rights Reserved
Site Designed and Maintained by YoeYo.com

oberg corner piece