"I think what we're seeing is how the Mobile Transporter operates in its natural environment of outer space," said Ben Selari, a NASA mission manager at the Johnson Space Center in Houston.The $190 million device is designed to carry the station's robot arm up and down the rail to set positions called work sites. There the platform can plug itself into power and data sockets so the arm can be operated.
Without this capability future elements of the space station can't be added as the Canadarm2 is the only way to reach the far ends of the outpost.
Beginning at 8:22 a.m. EDT (1222 GMT) Monday, Expedition Four flight engineer Carl Walz entered the command to begin moving the transporter from its launch position to what NASA calls work site No. 4, about eight feet (2.4 meters) away.
"We're about to start the space railroad here if you're all ready on the ground," Walz said to mark the event.
Moving less than one-inch per second, the transporter crawled down the rail as expected until it reached work site No. 4, but as the platform tried to latch itself to the site an error message flashed on computers and everything was shut down and put in "safe" mode.
"Our hearts really didn't skip a beat," Walz said later in the day. "We knew it wasn't going to keep rolling right off of S-Zero. We were extremely confident we were in a safe configuration."
It didn't take long for engineers to come up with a theory and a solution.
Small strips of iron in the rail tell magnetic sensors within the platform when it is over a work site, allowing automatic software to command the latches to close and firmly lock the Mobile Transporter in place.
Engineers believe that as the latches close the rest of the platform is raising up a tiny amount. As a result, the magnets are losing their connection with the iron strips, which sends a signal to the software that the platform isn't in the right position and everything shuts down.
Flight controllers turned off the automatic sequence and manually sent the orders for the platform to clamp down on the rail. Hours later they unlatched the platform and moved it to the other end of the S-Zero truss to work site No. 5 and then experienced the very same scenario, but this time they were ready for it.
"All in all, all the electronics, all the systems, all the software on the mobile transporter are working very well, there are no glitches, " Selari said. "This is not a software problem."
By Monday evening NASA controllers plan to move the platform back to work site No. 4, where the transporter will stay until the next shuttle mission in May. That's when additional equipment is to be launched and attached to the platform.
Known as the Mobile Base System, the device will be added to the Mobile Transporter and provide the interface necessary for the Canadarm2 to finally connect itself to the moving platform, an event that is targeted for a shuttle mission this Fall.
Meanwhile, Atlantis crewmembers Jerry Ross and Lee Morin worked Monday to prepare for the fourth and final spacewalk planned for this mission, which is scheduled to begin about 10:30 a.m. EDT (1430 GMT) Tuesday.
NASA's pair of spacewalking grandfathers are to step outside for nearly seven hours of work that includes installing a "ladder" between the Quest airlock and the S-Zero truss, removing a bolt on the Mobile Transporter system, moving a blanket that was found to be partially covering a sensitive antenna and stowing some tools out the station for future use.
Officials said Atlantis will undock from the station on Wednesday and land in Florida on Friday. The idea of spending more time docked to the outpost was shot down by managers because that would complicate preparations at the station for a Soyuz taxi crew to arrive April 27.