PROGRAMMABLE TECHNOLOGIES WEB SITE

A scientific study of the problems of digital engineering for space flight systems,
with a view to their practical solution.


FGB Launch and Software Error

The following exerpt is posted with permission of the author, James Oberg (http://www.jamesoberg.com) from:

Star-Crossed Orbits - Inside the US-Russian Space Alliance
James Oberg
© 2002 McGraw-Hill
pp. 245-247
ISBN: 0071407960

James Oberg's Reference: This incident of Nov 22, 1998, was reported on two different Moscow TV stations on August 4, 2000.  Oberg reports that as far as he can discover, neither NASA nor the Russian Space Agency or any other www site has provided any information on this event.   I am currently looking for a technical report or more details on this incident.  Russian or English language reports are fine. -- rk

On November 22, 1998, the FGB-the first piece of the International Space Station-was carried into orbit atop a Proton rocket. Another flaw in the design very nearly stopped the program dead in its tracks within hours of launch. Once the FGB had reached orbit, champagne corks popped at Baykonur, but at Russia's military space control center southwest of Moscow, it was eyeballs that were popping. As the FGB passed overhead on its first orbit, controllers radioed up some routine instructions for the autopilot to prepare to raise its orbit. The FGB sailed on, not responding. The commands were not even acknowledged.

The controllers at Krasnoznamensk (also called Golytsino) were accustomed to crisis. Military teams there controlled 120 of Russia's operational satellites, all but Mir. Most of these satellites had far exceeded their design lifetimes and were limping along, acting up, and even completely breaking down. But the FGB wasn't supposed to start off acting as if it were dying.

The FGB command system used a military radio link, called KOMPARUS. Russia never told NASA the command codes, and NASA eventually abandoned plans to develop independent monitoring, command, and control capabilities. During our orbital design activities, my team had again and again pounded on the Russians to at least disclose simple things that we needed to know. One of the most important was a list of the locations of the Russian ground tracking sites, so that we could compute the first and last opportunities for communication each day. It took a year to get a serious response, and when it did arrive, I doubted whether it was accurate. And in the year between my departure from NASA and the launch of the FGB, two of the stations either broke down or were shut down for lack of funds.

Understandably, there were no NASA observers at Krasnoznamensk. Probably, few NASA workers even knew that it existed, and none of them knew exactly where it was located. No Americans had ever been invited to inspect it. On the other side of Moscow, at the famous TsUP, a control room had been modified to monitor the FGB's telemetry. But the raw data on the screens there came originally from Krasnoznamensk.

So on the day the FGB was born, and appeared to be about to die, a lieutenant colonel named Nazarov was facing the crisis alone. He was no stranger to stress. Along with the rest of Russia's military officers, he was severely underpaid (the equivalent of $70 a month), and his pay was years in arrears. But he was on duty.

The FGB was in its preliminary "parking orbit," just above the atmosphere.   Without rocket burns within a day or two, it would start to tumble, then quickly fall back and burn up. A miracle was called for.

As soon as the FGB passed out of radio range, Nazarov began attempting a long-range diagnosis of the problem. The FGB passed over Russia twice more.  Soon it would drift too far west to be contacted for almost a day. With one communications pass remaining, Nazarov formulated a theory.

Exactly how his intuition arrived at this approach was never made clear. It would be a year and a half before the Russians even disclosed that the event had occurred at all. But the idea probably came from a casebook of previous command failures of FGB-class vehicles. NASA was never shown this list, of course, and wouldn't have been able to do anything with the list even if it had seen it.

Carefully, Nazarov sketched out an alternative command code scheme that he thought the autopilot might accept. Ninety minutes later, the FGB had circled around Earth and once again appeared over Russia, for the last time that day. Nazarov sent up his modified command codes.

They worked. The autopilot saluted in electronic style, accepted the commands, and turned the spacecraft so that its maneuvering engines pointed in the correct direction. It moved farther out into space, away from the upper atmosphere. The spacecraft, and the entire program with it, was saved.

The Russians decided not to tell NASA about the scare. What NASA didn't know couldn't hurt it, they probably figured, not for the first (or the last) time. A software error in the FGB's autopilot was soon confirmed (as Nazarov had guessed) and quietly corrected by radio commands from Earth.

It's for cases such as this that NASA develops backup procedures. With its worldwide tracking capability, NASA could have stepped in to tackle the problem if Nazarov and his team had failed. But the Russians made that impossible by not providing NASA with the command codes and never informing NASA of the crisis. In order to maintain control, they subjected the project to a near-lethal risk in the very first hours of its flight. And they got away with it: There's no evidence that NASA ever complained, even after the agency found out what a close call it had been.

For Nazarov, the incident also had a happy ending. He got a medal, a thank-you note, and a cash award of several months' pay, plus all of the back pay that was owed to him. It made him a very happy man, even though he had been the last man on Earth to celebrate the success of the FGB on launch day. In truth, he had been the first man to be justified in celebrating, since all of the other festivities took place only because the celebrants were not aware of what was really going on.


Home
Last Revised: November 29, 2002
Digital Engineering Institute
Web Grunt: Richard Katz
NACA Seal