Spaceflight Now

Discovery mission extended; moving up launch ruled out
Posted: November 9, 2006

UPDATED to include solar array software details

Discovery arrived at pad 39B this morning. Photo: NASA-KSC
Rollout images available here

The shuttle Discovery was hauled to launch pad 39B today for work to ready the ship for blastoff Dec. 7 on a critical space station assembly and re-wiring mission. NASA managers considered moving launch up an additional day, to Dec. 6, but ruled that out today based on time needed to complete crew training and to develop software intended to prevent damage to the station's new solar arrays.

The software in question is designed to constantly monitor the positions of the sun-tracking arrays and warn flight controllers of possible rocket plume contamination or excessive structural loads, Sources said the new monitoring software - and the training needed to use it - likely cannot be formally certified before Discovery's current December launch window closes.

Instead, engineers hope to have a workable system in place by Dec. 7, although details about what needs to be done are not yet clear.

Mounted atop a powerful crawler-transporter, Discovery and its mobile launch platform began the 4.2-mile trip from the Vehicle Assembly Building to the pad at 12:29 a.m. The MLP was "hard down" at the pad by 9:03 a.m.

Spaceflight Now Plus
The space shuttle Discovery departs the Vehicle Assembly Building in darkness and arrives at pad 39B at sunrise in readiness for its station assembly mission. FREE PREVIEW

Discovery's crew - commander Mark Polansky, pilot William Oefelein, Nicholas Patrick, Robert Curbeam, European Space Agency astronaut Christer Fuglesang, Joan Higginbotham and Sunita Williams - plans to fly to the Kennedy Space Center next week to review emergency procedures and participate in a dress rehearsal countdown Thursday.

A two-day flight readiness review to assess Discovery's processing is scheduled for Nov. 28-29. If no problems develop, launch will be targeted for no earlier than Dec. 7. The launch window that day opens at 9:30:42 p.m. and closes at 9:40:42 p.m. Liftoff will be targeted for the middle of the window, at 9:35:42 p.m.

The goal of the flight is to attach a short spacer segment to the station's solar array truss and to carry out a complex two-spacewalk re-wiring job to switch the station over from interim power to its permanent electrical system. That change over, which requires extensive realtime ground commanding and system power cycling during ongoing spacewalks, makes Discovery's flight the most complex station assembly mission yet attempted.

Mission STS-116 initially was baselined for 11 days with the possibility of extending the flight one day in orbit if on-board supplies permitted. NASA managers recently made that official, baselining a 12-day flight and adding a post-undocking (flight day 11) heat shield inspection to the crew's timeline. Here is a revised summary timeline:

  • FD-1 (12/07): Launch at 9:36 p.m.
  • FD-2 (12/08): Heat shield inspection
  • FD-3 (12/09): Space station docking; P5 truss removal from bay
  • FD-4 (12/10): Spacewalk 1: P5 attachment to P4 solar array segment
  • FD-5 (12/11): P6 port solar wing retraction
  • FD-6 (12/12): Spacewalk 2: Electrical channel 2/3 re-wiring
  • FD-7 (12/13): Crew off duty time; crew news conference
  • FD-8 (12/14): Spacewalk 3: Electrical channel 1/4 re-wiring
  • FD-9 (12/15): Equipment transfers
  • FD-10 (12/16): Discovery undocks from space station
  • FD-11 (12/17): Heat shield inspection
  • FD-12 (12/18): Cabin stow
  • FD-13 (12/19): Landing at 4:37 p.m.
Senior NASA managers meeting today at the Johnson Space Center ruled out moving the launch date up to Dec. 6. The flight already had been moved up a week, from Dec. 14 to Dec. 7. Crew trainers and the flight control team argued against another move to Dec. 6 to avoid losing another day of preparation.

Completing development of the new solar array monitoring software is a bit of a wild card in mission planning.

Before the station can be switched from interim to permanent power, one of the two solar array wings providing interim power must be retracted. In addition, the new array panels attached to the left side of the station's main solar power truss in September must begin rotating like a giant paddle wheel to keep face on to the sun.

As the arrays rotate on the end of a massive solar alpha rotary joint, or SARJ, their masts are moved at right angles, by so-called beta-gimbal joints, to change their pitch as required.

Engineers always knew the start of that complex, automated sun tracking would require careful monitoring to make sure station or shuttle thrusters did not deposit contamination on the solar cells or impart excessive loads that might damage or even break the fragile hardware.

As it turned out, the requirements for the software were late and development is not yet complete. Engineers are racing the clock to get a workable, if not certified, version in place by the opening of Discovery's launch window.

With that issue percolating in the background, Program Manager Wayne Hale asked the shuttle team to consider moving launch up to Dec. 6 as a way to add a day to Discovery's launch window. An extra day would provide a bit of insurance getting the shuttle mission off before running up against the end of the year.

When the shuttle's flight control software was developed in the 1970s, NASA managers did not envision the possibility of flying missions during the transition from one year to the next. Internal clocks, instead of rolling over to Jan. 1, 2007, would simply keep counting up, putting them at odds with navigation systems on the ground.

Spaceflight Now Plus
Shuttle program manager Wayne Hale explains why a computer software issue means Discovery must be back on Earth before New Year's Eve. FREE PREVIEW

"It's an interesting problem because if you remember a few years ago, we went through the Y2K change and there was a lot of concern about what computers would do," Hale told reporters earlier this week. "The interesting thing about the shuttle computers and the ground computers that support the shuttle is they were never envisioned to fly through a year-end change over. So the shuttle computers actually keep counting and they believe it's day 366 instead of day 1 of the year.

"That sounds rather trivial, but the fact of the matter is to keep the navigation in synch with the rest of the world, which has changed from day 365 to day 1, you've got to make that change appropriately and it was never designed in."

Space station software, on the other hand, was designed from the ground up to handle year-end rollovers and in the wake of the 2003 Columbia disaster, engineers re-evaluated the shuttle's software to make sure an emergency rescue mission, if needed, could safely fly across a new year.

"We had certified that for contingency use in the sense that if we ever had to fly a launch-on-need rescue mission and it happened to cross a year-end rollover, it would work," Hale said. "So we did quite a bit of testing on the software at that point. But there is a different level of testing that you need to do when you want to execute a procedure like that for a normal, planned, not contingency kind of operation.

"In April, we asked the team to go off and do that work. There had been a series of problems with that work and it turns out while we feel confident that it would work if we had to use it, we did not get the normal amount of testing and a normal amount of runtime on what are some very complicated procedures, both on the ground and with the crew to keep everything in synch across the end of the year.

"So right now, coming out of a review last week, it looks like we will not try to execute the flight over the year end," Hale said. "We're going to review that at the flight readiness review, so I would not call it a hard constraint at this time but rather a recommendation to take forward."

NASA managers today officially ruled out a Dec. 6 launch. A requirement to be back on the ground by Dec. 31 at the latest means Discovery must take off by Dec. 17, which would result in a landing Dec. 29 and still preserve two additional landing days in case of bad weather or other problems.

Florida's nighttime December weather is relatively mild and with 11 launch opportunities from Dec. 7 through Dec. 17, flight controllers are optimistic they'll get Discovery into space before the 2006 window closes.

"With the more benign weather that we have in Florida in December," Hale said, "we think that would probably be adequate to get us off."

Spaceflight Now Plus
Additional coverage for subscribers: