that two more sensors would fail. however, if you have some sort of common cause failure, perhaps this lot of sensors wasn't, you know, manufactured correctly. then the possibility of two more sensors failing was a lot more likely. because we didn't know what was wrong, we couldn't really say whether we were in a common cause situation or not. so one of the other things that happened is on this return to flight mission, again, we launched successfully a second time. none of these eco sensor problems. however, something that disturbed us all greatly is we had a large piece of foam come off the tank. this was after two and a half years of working to make sure we would not have any debris come off that tank. it was really the only large piece of manually applied foam

that was still left on that tank, and i can remember very well a lot of discussions that we had about whether we needed to do something differently. we did do something a little bit differently, but we didn't totally change that system. that could have been, again, a very serious situation if it had hit the shuttle, which it did not. but because of that, the next launch was actually delayed another year as we reworked that particular part of the tank to deal with that other problem. so even after looking at that problem for two and a half years, we had not solved everything we needed to on the tank, and we had to go back and do that. so now we come down to this second flight. we called it the second return

to flight. again, a year later. we're counting down to launch, and we have an eco sensor failure. and a few days later we tried again. again, we were invoking that launch commit criteria corollary rule, where if everything else looks good, you go to launch. we had another sensor failure, a different one on that launch. so we just had two scrubs of that mission now, there were many people that were arguing, even with one failure, and even though we didn't know what that failure was, that we still had enough margin we could go launch. again, because we had other sensors. but we decided not to do that. we instituted a different kind of inspection of these sensors. so we were looking at them with a different type of non-destructive evaluation to understand. people thought they had some idea of what was happening with the sensors. they were looking at that, trying to pick the best sensors. we were looking at many other

possibilities. there's an electronics box that the signals go through, and there were some people that felt the issue was actually an electronics box. you would see people come into meetings with t-shirts saying it's not the box. you know, other people arguing about what it was. we looked at the cabling between the sensors all the way from the sensors through the tank, into the orbiter where the box was. there didn't seem to be any commonality as to what cables were being used in missions that had problems and missions that didn't. they were trying to look everywhere along that line. but we were not able to find it. eventually we did, as i mentioned, go to a different lot of sensors that had been inspected. we were able to put some different sensors in, and we launched that mission. about two or three months later we were counting down to the next mission. this was sts-115.

so we're now in 2006. let me just tell you, this launch had already been delayed three times. once for a malfunction in one of the fuel cells. once after a lightning strike on the pad. once you have that, now you've got to go through a bunch of testing of the whole orbiter because a lightning strike could potentially have caused some electrical damage somewhere in the shuttle, and you've got to do a lot of testing to make sure that hadn't happened. then once due to weather from a tropical storm. so again, the day we were doing this launch countdown, in the news, i'm quoting from the news, the agency is under pressure to launch atlantis this week or face delaying its mission to the iss until october. again, let me put this in context. after columbia we came out with

a new vision for space exploration, which said we are going to shut down the shuttle program in 2010. we want to be finished with the assembly of the international space station by then. then we're going to start up an exploration program where we'll develop a heavy lift rocket, a new spacecraft, and look toward going back to the moon. they needed to shut down the shuttle program in order to use that money to really ramp up the development of the exploration program. so what was of course on everybody's mind is, we've got to get these certain number of shuttle flights in by 2010 in order to have the international space station finished, and then we can operate it and retire the space shuttle program. again, that's the environment we were all working in. so we're counting down. my boss, the director of flight crew ops, was in florida as part of the mission management team. i was in mcc. he and i were talking back and forth by phone, and one of the eco sensors failed as they loaded the cryo that morning. so, you know, of course they went off.

there's some tests they can do during the launch countdown. you don't have to stop the launch countdown. this usually happens several hours before the actual launch is scheduled to occur because it happens during the tanking. and so he and i were talking back and forth by phone. he was asking me a lot of questions because i had been the one that had sat through most of the technical and operational briefings and most of the control boards where we had discussed this issue. and so he was looking for my opinion because he felt i was probably more up on the details than he was in terms of everything that we had talked about. let me just say that, again, we had talked through the situation many times of if on a first launch we have just one sensor that's acting up, should we just

go ahead and launch? as a community through the many months that we had discussed this, we had never agreed to do that. some combination of engineering or safety and other people had just not been able to sign up to that. so we had never changed that rule. we still had on the rule that you needed to have four. however, on a second attempt, you could go with only three, if it was the same sensor that was acting up. but on this day, again, we talked about, well, should we launch today even with one sensor not operating? so i was listening very hard, and as the engineering people came up and briefed, trying to understand, this is there any new piece of information? obviously, if there's something in the testing you were able to do as they were proceeding during this launch count that gave some new information, that, for example, could give you some information about whether it was a common cause failure or not. any piece of new information that might allow you to think differently about that rule, that's what i was listening for. and i was also listening to what other people said. i happened to be sitting next to

the flight controller in mission control whose job is to monitor the main engines during flight. in certain situations he would have to be the one to make a call up to the crew to do an emergency shutdown of an engine if he saw something happening that he was concerned about. so you can imagine that, you know, he was really in the hot seat for any sort of engine issues. when the flight director in houston mission control pulled his team, that flight controller said he was a no go. however, the flight director reported down to florida that he personally was go, and he was speaking for his whole team. now, he's the authority to be go, even if one of his team members is no go. because everybody personally has to decide what they think that risk is and determine whether they're go or no go for that. but i thought it was quite

interesting that the flight controller himself was no go. that fact wasn't actually reported to the mission management team. i knew it because i was sitting there. the people on the loop, all they heard was mission control was go. so as we talked about it and they came down to the point where each mission management team had to say go or no go, again, i was talking to my boss and said my vote is no go because we've talked many times about should we change the flight rule. we've never agreed as a community. during a launch countdown is not the time to change that unless there's some new information, and i said i have not heard anything.

i've been listening to all the loops. i have not heard anything. and we all have to be willing to take risks because there's never a time we launch the shuttle where there's no risk to that crew. so you have to be willing to take a risk. however, what i felt is we were taking a risk we didn't fully understand. we had no idea why these sensors weren't working. and therefore, we didn't know if there was a common cause failure. we didn't know if there was potentially something else wrong with that system that could even show up in a way that was unanticipated. that was my reason. so i was pretty confident that as we went around the room engineering was going to vote no and safety was going to vote no. they went around the room. my boss voted no go. when they finished the poll, it turned out the crew was the only one that voted no go, which really surprised me. so because i was the one that had sort of influenced that

decision, the nasa administrator actually called me up on the phone. i could tell he thought the risk equation was we could go that day. however, he was definitely willing to listen to me, and i gave him my reasons and told him i felt we really didn't fully understand the risk we were taking and that as a community we had not been ability to get to that point of launching with one of these sensor fails. so in the end, about less than an hour before launch, we'd been talking about this for four hours at that point. they postponed the launch and we didn't launch that day. so let me tell you a little bit

of the rest of the story. so the next day we did another launch countdown. all the sensors were operating and we launched. so there was still obviously a team that was off trying to understand what was going on with these eco sensors. they postulated a new theory on the sensors. they came up with a new inspection. for the next four shuttle flights, we didn't see any issues with the sensors. so many people thought the second new inspection was doing the trick. we were finding the best sensors, and therefore we were not having issues with this problem. well, five flights down the line, and now we're december of 2007. we come to the launch countdown of sts-122. during that countdown, two eco sensors fail. and then another one failed as they were actually draining the cryogenic fuel out of the tank. so three of the four eco sensors failed on that day. we tried a launch attempt again three days later, and one of the sensors failed. and we did not launch. the shuttle program manager finally got in front of his team and said, okay, we're not going to prepare for another launch until we understand why these sensors are failing. we are not going to delay another shuttle launch on the day of the launch because of an eco sensor fail. i think that was really the first time that statement had really been made. so we ended up doing a highly

instrumented tanking test. that ended up focusing some of the work on a connector that passes electrical signals from the sensors through the wall of the external tank to the shuttle to the sensor box. there was movement inside the connecter that caused some metal pins to stop touching the metal inside the sockets. and it kind of makes sense that you would then see it when you would load the tank with cryogenic fuel because that would change essentially how much room you have in the connector between the pins and the sockets they were in. it turns out once that was postulated there were other people outside the shuttle program that heard about it. they said, hey, about a dozen years ago the centaur rocket had a similar problem.

we ended up fixing it by permanently soldering the pins and the sockets. it was a difficult soldering task, precise, but since they had done it they hadn't had any further problems. we got some of the same technicians that supported this other program, titan program, and ended up doing a very around the clock design and fix problem. about two months later is when we went back to launch the mission we delayed. didn't see a censor problem. never saw another sensor problem throughout the rest of the shuttle program which was another 13 flights. so i have often thought back to that day, certainly never went into mission control thinking i was going to stop a shuttle launch.

but i thought a lot about that means and particularly in the context of today of walking the walk. you know, i don't feel that anybody was particularly right or particularly wrong on that day. i do think as a team we should have tried harder to determine the root cause earlier. that we didn't get to that point early enough although thankfully it didn't lead to any sort of tragedy in this case. i think a lot of it was were each of us following the core values? were we incorporating engineering integrity and following the data as much as possible and fostering a culture where we could speak up. did we have a culture where people would listen? that was obviously critical too. and a culture of asking questions. so that is just one of the examples of some of the difficult decisions that i think

managers in human space flight program have to face and when you think about it, it was almost on a weekly basis where we had to make decisions about changes we would or would not make to the shuttle based on concerns that people had or problems that they had seen and every week you were in there wondering if i vote we don't need to make this decision, this change, two years down the line will there be something really bad that happens on the shuttle. so it wasn't just during launch countdowns you face difficult decision but during that time where there is added extra time pressure and pressure of everyone wanting to launch that day that you have to set up that culture in advance that people feel like, i'm not going to act any different on this day than i would on any day previously. fluz so what i would like to do now is -- i guess there was one other thing i wanted to talk about, which is obviously now

i'm the deputy center director at jsc. i would say our biggest leadership challenge lately has been the transition that we're going through in the human space flight program. obviously shutting down the shuttle program. we did have a number of years to prepare for that but it was still a difficult thing to do. part of our transition plan was we had this constellation program. in terms of keeping critical skills and transitioning to a new program that people would work on and moving human space flight forward, we thought we had the makings of a pretty good plan. then with about a year left in the shuttle program, a new administration canceled the constellation exploration program so that made our leadership challenge at jsc more difficult when our transition plan now changed and two out of the three programs that we were running were essentially

shutting down at the same time. rather than going into a big talk about that, i will just talk about one of the leadership books that i've read recently that i've tried to keep a couple things in mind from. we had a group of employees at jsc who were early career employees who decided to put together their own leadership program because they felt we didn't have one that was really addressed at early career employees. and part of that, it was a year-long program but part was a book club where they read books during the year and they would ask senior manager at the center to have a book club discussion. they came to me and asked if i would lead the discussion on book "seven habits of highly effective people." a lot of you might be familiar with that book. i heard about that book but had never actually read it. it was actually a really good thing for me to read that book

and then to think about a lot of the actions and situations that i had been in over the last year or two and particularly where i thought i could be more effective and i came up with two for me. i think anybody that reads that book will probably find ones that particularly speak to them in the situations that they find them in. i didn't have to get far in the book. habit one spoke to me. that was be pro-active and realizing that you choose how to respond to any situation. if you find yourself sitting back thinking why are they doing this to me? why are they canceling the constellation program? or you find yourself criticizing or complaining about what is going on in your world, that you're missing out on the single most important thing that you can do to be effective which is you choose how to respond to that. what you want to choose to do is understand where your sphere of influence is and where you can take action so i tried to think of that as we head to shut down

two programs and understand how we can motivate our folks at jsc to continue to work in human space flight program. the second one that really spoke to me in the situations that i find myself in is one that says seek first to understand, then to be understood. in is that sa first to understand and then to be understood. then a lot of situations i find myself in in leadership positions, i felt my role in a meeting was to advocate for the organization i represented. if i was in there as the flight crew representative, i was advoit voicating for a position we worked out. my main job is to advocate. if i am in a meeting representing johnson space center, i am advocating for johnson space center whether for resources or something like that.

it's hard to be effective is you are just advocating over and over again. everyone else is taking their position and how are we ever going to find the solution if people don't sit back and first ask themselves why does that other person in that meeting think the way they do. why are they recommending that? i bet in the end they have the same goal i do and they have come to a different conclusion about what needs to be done and what we should be fixing or changing and my first job is to understand why that smart person over there thinks the way they do. then see if that influences or can help us come to a solution which often incorporates what i have come to believe what i felt they need to do in that situation. so perhaps that will inspire you to maybe look at that book and think about how to be more

personally effective as you go about your day to daywork. what i would like to do now is show that video i brought that was my fourth space shuttle flight. part of the assembly at the space station program. at the end it has more current video at the international space station today. if you can show that, this was april 2002. our commander is front right there. by the way, a graduate of the air force academy. there were already three astronauts and cosmonauts on board the space station by the time they got there. i am the flight engineer and i'm in the middle in the back row supporting the commander and pilot in the dynamic phases of

launch. undocumenting the station in landing. that was one of my jobs on that mission. taking off with seven million pounds of thrust which is pretty exciting on a florida afternoon there. as you know, it operates for the first two minutes of launch. that's the solid fuel used up from the rest of the shuttle and we go the rest of the way on the main imagines. once we get into orbit, we start preparing for the rendez vous at the international space station and do a number of burns over the first two days. you need to be hanging on when you do the burns because they do in part force on the shuttle. then there was a team of three of us, the commander and pilot in me working through the rendez

vous procedures as they go through these burns and bring in navigation information and data and try to understand where we are. this is a picture of us taken from the crew on the international space station. that big piece in the middle was what we were bringing up and this is the view we had as we were coming up to dock and the silver ball in the center is where we were going to go dock. this is the view on the aft flight deck, the station into view. it looks like we are meeting it. the top part is the station. that's the actual initial contact between the two. about two hours later, we open the hatches and the commanders of the space station and the russian cosmonaut and the rest

of the crew on to the space station. if just the three of them for four months and they were happy to see other people. then we started transferring supplies, that is something you do on all the missions and carry things between your knees in addition in your hands. it makes it more esfficient. the next day we started our major task which was to lift this piece of trust structure up out of the shuttle, using the station robot arm into the shuttle. move it into position to attach it to the zenith side of the international space station. i was working along with dan who was one of the members on the robot arm. we don't have windows. when we were working on the spacetation arm at this time and you use cameras that are not always the best.

this shows the middle east view of earth out beyond the trust. this is a little further on. you are looking at the trust being moved into position. this is on the zenith part of the lab. you can see that clog that forms the initial mechanical attachment. over the next few days, we will form a better structural attachment. dan and i, dan was at the controls for the actual birthing on top of the space station. over the next week, we did four space walks. there were four crew members on board who were space walking crew. going out and each day one of them was on the arm and the other was free floating just with a tether. my job essentially was to move around the person on the end of the station arm as they were doing the work they needed to do

to hook up this trust. it's not just an empty trust. it's all kinds of electrical and computer and power equipment. they had to make many, many connections. it had to make a lot of those structural data. you can see a person on the end of the arm holding a v-shaped piece of equipment. she got to spend a half hour looking out on to the earth and the space from one side to the other and said it was the most fun half hour of the space walk. this is what it looked like once we repressurize.

good booarbecued beef. steve smith showing how tension is doing really important things in the absence of gravity. after a week on orbit, our week is done and made sure and prepared to undock. this is a part of the flight that is pretty dynamic. a navy pilot was the one at the controls for the undock. that's the target that you use to make sure you are backing away in a straight line for about 400 feet. we wait for sunrise and do a

full fly around taking pictures to help document the outside of the station that is used to understand fully the configuration of the station for future engineering and training exercises. here's what we looked like as we are leaving. this is when we are halfway around. you can see at the time they said we only had one solar ray. you can see the station arm hanging off. that's what we added to the space station. you can see the shuttle tail. we prepare for entry ourselves and get into our launch and