The space station's electrical system was designed to operate in an interim mode during the initial stages of construction, providing power to critical systems from a set of solar arrays - P6 - mounted atop the Z1 truss extending upward from the Unity module.

P6 provides electrical power to six DDCU transformers inside the lab module and two others in the Z1 truss. An interim cooling system keeps the electrical components from overheating.

Before the station's power system can be reconfigured, the left wing of the P6 array - known as the 4B wing - must be retracted to permit the newly installed P4 panels to rotate as needed to track the sun. The right wing of P6 - the 2B wing - will be retracted during the next shuttle flight in March. If all goes well, P6 will be moved, bolted to the P5 spacer truss and its 2B/4B solar blankets re-extended during a shuttle flight next fall.

"Until now, P6 has been parked in an interim location in the middle of the truss on top of Unity," Hill said in an email. "Not only has it provided electricity, but it used a smaller, temporary cooling system for the U.S. segment until the MBSUs and the more robust permanent cooling system were installed and activated.

"This set up has worked very well, but it also means we have many electrical and cooling lines connected to the early or temporary systems, rather than the permanent or assembly-complete architecture. In order to finally connect all of the powered equipment to the their permanent power and cooling sources, the vast majority of the U.S. equipment must be powered down, the connections physically switched by spacewalking astronauts, then powered back up.

"This is a long, choreographed activity in order to ensure at least one of every critical component remains powered throughout the reconfigurations," he said. "The important element in all of this is that the additional power provided by the new solar arrays and the additional cooling the permanent cooling system provide are necessary before we can install the Japanese and European research laboratories which are also coming in the next year."

The P6 port wing retraction is scheduled for the day between the first and second spacewalks. The night before, flight controllers will begin moving critical station operations from hardware powered by the 4B solar array (through channels 1 and 4) over to hardware powered by the 2B solar array (through channels 2 and 3). The goal is to get the total load on the 4B array to around 6.5 kilowatts.

On retraction day, flight controllers will execute a procedure known as "seamless power channel handover," which will move the 6.5-kilowatt load remaining on the P6-4B wing to the newly installed P4-4A wing on the left end of the main solar array truss. Once the handover is complete, the P6-2B wing will be providing power to station channel 2 and 3 while P4-4A will be powering channel 1 and 4.

At that point, commands will be sent to retract the left side P6-4B solar array wing. The wing must be retracted at least 40 percent to provide the clearance needed for the new P4 array to rotate as needed to track the sun. But from a structural loads standpoint, the array must be retracted to the point where no more than one of the 31 bays making up the central mast remains extended.

"Flight director's always worry about the worst case," Curry said in an interview. "These wings have been out there for six years now, taking thermal cycles and those kinds of things. The mechanism is relatively complicated. There's a chance that we might have a problem with that."

Working in a step-by-step fashion, the astronauts first will send commands to retract the mast three 40-inch bays. If that goes well, retraction will resume and continue to the point where only one bay remains extended. All of that will happen during a single 45-minute-long daylight pass.

"We're going to retract to three bays, stop, survey, make sure everything looks good then start back up again," Curry said. "Then we're going to go from three bays all the way down to one bay, which is about 40 inches. I think there's a good chance that we could end up staying at that location for a long time. I've gotten the loads people to tell me that from an attitude control perspective, from a shuttle undocking perspective, those kind of things, I could go ahead and do the rest of the mission and still keep the array not fully retracted, keep the array at one bay."

The reason that might be needed is because the solar blankets, which were packed up like venetian blinds for launch, might not fold up smoothly during retraction."

"What I'm telling you is there's a decent chance that won't happen during the 116 mission, or that day, and I don't particularly care because it doesn't stop me from doing the rest of the flight," Curry said.

"Why would I think we might stop at one bay? The analogy I always give people, when we buy a map at the store it's nice and compressed. But when you start using your map for six years and you decide it's time to fold it back up again, it doesn't fold as tightly as when it was bought at the store. That's what's going to happen here. There's definitely going to be some waves on those folds even if it comes back perfectly.

"So I think there's going to be some concern with compressing that last 40 inches because you could potentially be breaking glass, damaging those cells. So if I'm benign there, I'll let the engineering guys assess that for weeks, let them get comfortable and figure out whether it's a good idea to compress that final 40 inches and latch down.

"My goal for 116, to allow me to execute the rest of the flight, is to get to one bay. If everything's great, we'll go ahead and do the final retraction. (If not) I don't want anybody to think that that means something's bad, it just means we're being conservative."

If the array doesn't retract for any reason, gets jammed or suffers some sort of mechanical failure, Curbeam and Fuglesang can attempt repairs - the array can be retracted manually if the drive motor fails or the spacewalkers could attempt to re-spool a tensioning cable if it popped off its rollers.

But there is little else the spacewalkers can do. If a problem crops up that can't be fixed, they are prepared to dump the costly array overboard.

"If you have your three-sigma bad day, those are you're two big things," Curbeam said in an interview with CBS News. "Hopefully it's just a motor problem or a gear problem where you can just retract it manually, which is fine. It's going to take a long time to get that done, but we can do it.

"But if that's not working for you, and the right answer comes out we've got to get rid of this thing so we can have fairly normal ops, well, we get rid of it. In the big scheme of things, if you have to fabricate and fly another array, that's a lot better than some of the bad things that can happen when you have a piece of structure that's been compromised on the station, i.e., it can come off, it can hit something. Not a good idea."

The P6 arrays cannot be moved to their eventual P5 attachment point with either wing extended. Aside from extremely tight clearance issues, a structural analysis indicates the fragile support mast likely would fail if subjected to the sort of loads it would experience being moved about on the end of the station's robot arm.

"That would be reassessed in realtime," Curry said. "But when we previously looked at it, it had two issues. Kinematically, it was really hard to figure out how to move the wing using the arm and move it out forward and just not hit anything. And then secondarily, and this is the one that killed it, they determined that because these wings are so flimsy you would end up permanently damaging them on the way to moving them outboard. Otherwise... the preference would be not to retract."

In any case, jettisoning the arrays is a strictly worst-case scenario.

"Let's say we start retracting and one of the battens breaks or the mast starts to ball up on itself or the array just gets all crumpled up on itself to the point where it can't be fixed," Curry said. "That's the only way we would ever consider that.

"I haven't spent a lot of time worrying about this because I can't imagine the program telling us to dump those arrays on this mission. If the array got stuck, I think we would spend a lot of time, two EVAs worth, trying to figure out what's wrong with the arrays and if there's any way to fix it. That means another flight has to be added to the manifest anyway."

Assuming P6-4B successfully retracts, flight controllers will configure its fully charged batteries to operate in "parachute mode." While the three nickel-hydrogen batteries cannot easily be recharged with the port array retracted, they can provide up to eight hours of emergency power to channel 1/4 if any problems crop up with the P4 arrays.

Then, within an hour of solar array retraction, computer commands will be executed to power up the left-side solar-alpha rotary joint, or SARJ, to begin rotating the P4 solar arrays as needed to track the sun. Because of the angle between the sun and the plane of the station's orbit, the array must start active tracking right away to generate enough power in the absence of the P6-4B wing.

"Once we get retracted down to one bay, I have a 'go' for loads and attitude controls so I can start rotating P4," Curry said. "This is going to be cool because we're going to be retracting P6 and rotating the SARJ ... basically at the same time. That will put the SARJ in the configuration where it starts tracking the sun like it's supposed to do. First time we will have done that. There are a lot of people really worried about the loads of this and trying to determine how much momentum we put into the system because of the fact that the wings are out."

When the SARJ is activated, control computers will carry out calculations to determine where the sun is relative to the arrays. The massive joint then will be driven at a higher-than-normal rate to get in the proper position. When the SARJ drive motor engages the teeth of a large gear and begins turning, the arrays will be deflected as momentum is transferred to the fragile mast.

"It will put momentum into the system," Curry said. "Because the way it works, as soon as you do that SARJ activation it's calculating an algorithm and figuring out where the sun is and it moves itself to wherever the sun is. So it could go forward or backward depending on where the sun is and when we start it up. ... It will dump momentum into the system, that's for sure."

Continue to Part 3 -->