by Rod Pyle
Let's talk now with another engineer, Al Chen. We met him in chapter 2, the night Curiosity landed. His soft voice, which only really showed some emotional chroma when he announced touchdown, was the one you heard that night. He's an easygoing fellow in his late thirties who lives in the next community over from JPL in La Crescenta, is married to a woman who also works at JPL, and has three kids, aged seven, four, and just under a year, when we spoke. That's a full house—as a father of one, I can't even imagine. At any rate, Chen seems the type who would handle three kids with grace, kindness, and nary a harsh comment in sight. He picks up the sky crane saga, discussing why the cords between the descent stage (rocket pack) and the rover were a bit of a nightmare. In short: “Having things connected with soft things that are sometimes elasticky is scary. We don't need to have [a] hard structure connecting things to understand how it works, but it's nice.” It's a bit like it was with the parachutes: when you move from dealing with metal parts and plastic circuit boards to cloth parachutes and springy ropes, the engineering numbers get soft—very soft. Testing becomes more critical than ever because the error margins are wide—prediction is almost as much about instinct as engineering. It did help that they had a fair amount of data returned from Mars by previous landings, but none of it modeled anything like this.
So what did he think when he first saw the design for sky crane? “This can't work. There's so many moving parts…” he recalled with a grin. “You definitely begin to feel like, when you take a step back several times in the process, ‘What have we done to ourselves here?’ We bit off so much new stuff on Curiosity that I'm surprised we talked people into letting us do it.” I asked for examples. “Well, let's see: first guided entry on Mars, biggest planetary vehicle, biggest lander, and, of course, what is this new crazy landing system? There's a lot going on there, but when you look at it, does it hold together?”
Then, after the initial shock wore off, he reconsidered. “It was a natural evolution—a lot of the folks [who] worked MER and MSL were Pathfinder folks at the beginning,” Chen observed. “Pathfinder in many ways is the opposite of what MSL is. On Pathfinder, they tried to do almost everything they could mechanically—very little guidance-navigation control. If, on the other hand, you turn around and look at MSL, it had almost gone the other way completely. We're trying to control everything.”
Well, any father of three little kids knows how that works out in most cases. It's nearly impossible…but with Curiosity they came damn close. The guided-entry part of the mission—a new capability—was of course critical, but also doable. Mike Malin's orbital imagery was sufficiently high-resolution that it could identify a lot of potentially rover-murdering situations on the ground. Chen recalled, “We could see almost every rock that was a problem. With guided entry we knew we could track our lift and put the lander in a tight spot, literally on a very tight spot—considering that the walls of the crater were as high as mountains.” In short: “Between sky crane and guided entry, we had to make very few compromises.”
So using guided entry, loading the computer with landing parameters including hazards, and identifying the relatively small area within Gale Crater in which it was supposed to land provided the engineers with a modicum of comfort. But there was still that final seventy feet to deal with.
“An interesting part of sky crane for me is this,” he continued: “Recognizing that once we get up to a mass like Curiosity has, we just can't do these energy-absorption systems we used before.” No airbags. “So we should then flip the paradigm—instead of hitting hard at several meters per second, up to 10 meters per second, on the airbags, we should touchdown softly. If you touchdown softly enough, you don't have to worry much about breaking, it doesn't matter how big it is.”
So bigger can be better after all. “Once you get to a rover this size, there are some advantages. If you touch it down softly enough, you can touch it down within the loads that are expected of it when it's driving around.” It's already designed for driving over and falling off rocks if necessary. “So it's tough, resilient, and strong. If you touch down slowly enough, you already have a landing system—you have your own landing gear right there.” The wheels and the suspension system would be its landing “legs.” “It's designed to conform to the slopes, to deal with rocks; that's everything you also want in a landing system. Something that is very stable and will tackle whatever it has to deal with.”
There is one more huge advantage of using the rover's own wheels as a landing system…“We don't have to drive it off a lander! A lot of people were still biting their fingernails when the MER landed because they were worried that we wouldn't be able to get the rover off the lander.” The airbags, once deflated, always threatened to snag the rover as it departed. If the snag was too profound, the rover would have stayed right there, which would not have made it much of a rover. “With sky crane you don't have to worry about that as much.”
“One technology that really made sky crane possible was good throttleable engines—like the Viking engines—that allowed us to get down and control how fast we were going with precision.” This is why they resurrected the 1970, vintage rockets that landed Viking—they were a good design that, with a bit of improvement and updating, could land MSL right on its wheels. Past rovers—Pathfinder and MER—had used solid-fueled rockets to slow them. These are an evolution of the same skyrockets that you see on the Fourth of July. They burn a solid mass, with a fixed rate of thrust (and, of course, we hope that they do not explode). And once lit, they burn till they are done. If you know in advance exactly how much thrust you need, for exactly how long, they are fine. But with MSL's computer-aided navigation to the surface, responding to the ground-ranging radar, they needed something they could control. The amount of, and the duration of, thrust was critical. At the very least, these rockets needed to be able to throttle-up at the last moment, slowing the rover to almost a hover, when sky crane deployed and the rover was lowered from the rocket stage down to the surface. Every parameter was critical and would need to be handled with a high degree of precision.
“With those rocket engines we had everything we needed to get there. We had the landing gear, we had engines that we could control to slow to low velocities, and we can tell that we are getting to low velocities with the radar. Those were all the key building blocks of sky crane.”
Chen smiled with satisfaction. “There was a rational reason for everything we did. Often we ended up flying systems that are a giant pile of compromises. We made very few compromises on sky crane. We got the engines we wanted that were very well controlled. We got the landing system that we wanted, the very capable high ground clearance rover with great landing gear. For the most part, we got the radar we wanted, it performed very well, told us our altitude and velocity right down to NASA precision.”
So, to paraphrase, the JPL Mars crew had the Six Million Dollar Man (look it up if you are under forty-five): they had the technology, they made it better…and it was ready to assault the red planet in a bold and daring fashion. Now they had to do it.
I've already rattled off a narrative of the evening as seen from the media center at JPL, where I was fortunate enough to be stationed. It was an amazing evening from that vantage point. But Chen was at his console, running the show. Adam Steltzner was pacing like a predatory animal behind the row of consoles, keeping tabs on everything he could see, making decisions. Rob Manning was in the back row, watching his team's machines execute the exacting functions needed to land safely. John Grotzinger and a few other scientists had stationed themselves wherever there was spare room. JPL's director, Charles Elachi, sat in back with Charles Bolden, the NASA administrator.
Talk about pressure. The fact that the week leading up to the landing had been extremely quiet (in engineering terms, anyway) made Chen even more nervous.
“As the operations lead, I was responsible for essentially laying out the plans for the hundreds of people in the week prior to entry and landing. We do all these opera
tional readiness tests and prepare for all sorts of crazy things. Then, the final week leading up to it was quiet. We were prepared for all contingencies. We were working around the clock, looking for problems, and we just didn't see any. It was kind of eerie. We were sitting around and looking at each other a lot, we ran a bunch of scenarios, and we still did all the work that we knew we should do, but in the end, we didn't have to do. Everything went like clockwork.
Fig. 16.2. SKY CRANE DEPLOYING: The main phases of sky crane involved the spacecraft being slowed by the descent stage (rocket pack), then the Curiosity rover being lowered by nylon cords. The descent stage was then cut free to fly away. Technical note: unlike this artist's rendering, the wheels were actually deployed downward by the time the rover was being lowered. Image from NASA/JPL-Caltech.
“We knew that cruise guidance had put us right on target. We didn't have to deal with that whole set of potential problems. I could throw out that part of the operations playbook. We were right down the middle. It's almost like something we didn't practice much—being right down the middle, right on the money. So this was kind of weird.” He left the control center when off-shift, trying to distract himself. “Sitting in the media room, going next door, playing a game, trying to take a nap. In some ways, I wished there was something to do when I was there. At least you wouldn't be too stressed because you [would be] doing something, but instead you are just waiting, and waiting…”
Bobak Ferdowsi expressed similar sentiments and referred to NASA training sessions before the landing day, meant to prepare them for how to behave if the worst case happened: “We had had some briefings a week or so before about what to expect, how to deal with success or failure, some very real scenarios about how to deal with failure, since it's the more difficult of the two to handle.” The agency needs people who can stay cool and focused even if things are not going well. “In your testing you realize that failure is not always immediately obvious, many of the scenarios don't involve realizing that everything broke and it's suddenly over. More often it's silence—we don't know what happened, it could still be alive, it could be dead, one of the radios could be dead and it could be a couple of days before it swaps to the other radio. Most of it is unknowns. So for us it was how to deal with this.” Ferdowsi was also nervous and spent the night before landing in a sleeping bag in his JPL office.
The waiting was over on August 5, 2012. By midafternoon, the parking lots at JPL (which cover an entire hillside) were packed. The press was ushered to the media center, and by early evening most were set up and ready to report. Up the hill a few buildings, in the Space Flight Operations Facility or SFOF, the EDL team was assembled where they had been for many hours, preparing to wait out the atmospheric interface and entry. They would be passive observers as MSL made its way to the surface of Mars, alone and autonomous, fourteen minutes before anyone on Earth would receive a confirmation.
Seven minutes of terror: entry, descent, and landing had begun.
What follows is my account of how the evening played out. It's not necessary to follow this chapter with extreme precision to understand the rest of the mission—but the landing was so cool, and I enjoyed it so much, that I am sharing the nerdy, joyful minutiae. I wish everyone reading this book could have been there to share the moment—it was, for an old guy like me, not far behind the excitement of the Apollo 11 landing. You probably can recall your own comparative historical highlight. But most people experienced it somewhere other than JPL, so we'll relive it together.
As a “true believer,” I headed up to JPL early. I also wanted to get a decent parking space, so I arrived before midday (parking is very limited at the lab's hillside location at the best of times). When I entered the media center, I saw a couple of familiar faces. As it turned out, a few old friends from my Griffith Observatory days were volunteering to help handle the onslaught of media needs for a large portion of the week leading up to landing. John Sepikas (that's Dr. John Sepikas to you) had gone on to become a math and astronomy professor at a local college, replacing the astronomy professor I had studied under and interned for there. The old chap's classes usually started out with about forty students (“Isn't this an astrology class??”—I knew then that they were in for an unpleasant surprise) and ended up with about five of us by the end of the grueling term, no exaggeration. Il Professore was tough. Sepikas has a far gentler touch and, from what I understand, works tirelessly to bring the subject to life for a new generation of students. Jim Somers was another Griffith expat who has completed a long career with Los Angeles County and in his semiretirement spends time working with Sepikas and his students, as well as staring at the sky through various telescopes. I also saw Bob Brooks that day, another Griffith pal, who has worked at JPL since the late 1970s on various programs, including Mars orbiters, and should one day at the very least have a Martian crater named after him. Finally, David Seidel, also of Griffith, has worked at the lab for decades designing and executing its excellent educational outreach programs. None of it got me any special treatment, but it was fun to catch up. Besides, I was there and at this point, many hours before landing, the most compelling thing in the media center (besides the displays of various JPL spacecraft from the past, and, well, the fact that it is JPL) was a series of video displays showing an ever-changing graphic of the current status of MSL. It was really interesting (JPL does provide the very coolest graphic interfaces), looking over the spacecraft's shoulder as it neared Mars. Current velocity, position, and other numeric values counted down on the margins of the display. It was a great way to see where things were and get a sense of the mission's progress.
Fig. 17.1. AS WE SAW IT: This frame is from the live video feed of the landing as sent out by NASA. Lower left readouts showed distance from the surface, velocity, and time until landing. Curiosity was fifty feet from touchdown at the time this was taken. Image from NASA/JPL-Caltech.
Over in the SFOF were the EDL team members and an ever-changing stream of folks involved in MSL but not critical to the landing. There was not a lot of spare room, so it was best to get your looks in early if you wanted to see what was going on in these last few hours before touchdown.
All over Southern California and across the country, public viewing stations and private viewing parties were getting up to speed. The Planetary Society, a nonprofit started a few decades back by some of JPL's top retirees, was hosting a huge viewing event in downtown Pasadena where I had just completed signing a few books I'd sold at the adjoining Mars Society convention. Outside, a throng had gathered around a life-size Curiosity rover display that JPL had lent out for the day. It was great to see so many folks—the place was jammed—who were enthused about the landing. I hoped they would continue to be fascinated and perhaps pressure their elected representatives to vote through more money for the space program, but that's another conversation.
Even Times Square in New York got into the act, showing live coverage on massive video screens. While the crowd of about one thousand never approached New Year's Eve proportions, they were very enthusiastic. Other crowds gathered worldwide. JPL also sent out a continuous stream of tweets, cute first-person things that were signed by Curiosity itself. NASA's web servers were operating at full steam to keep up with viewing demand. The total US audience was estimated to be well over three million. So much for the public being “bored with science,” as some detractors claim.
The hours counted down, and then…it was time. Entry interface (EI), the beginning of the “terror,” was just after 7:00 p.m. local (Pacific) time. At this point, the radio delay between MSL and Earth was about fourteen minutes, so the craft was on its own, thinking autonomously. Everything was automatic and handled by the onboard computer. But that did not mean that the control-room personnel weren't watching with care and concern.
The rest of the times will be in minutes after EI (Please note: these times have been revised by NASA since touchdown; the times used here are correct, according to JPL, at the time of this writing).
ENTRY INTERFACE +/– TIME (IN MINUTES)
–10:00 minutes: The cruise stage, no longer needed, detached from the lander aeroshell.
BOBAK FERDOWSI: The charming young engineer who would forever become known as “Mohawk Guy” spoke to me at length long after the landing. “Of course, we had had many dress rehearsals before that to make sure we knew what we were doing. So we just fell into the routine that we had used for most of cruise—there are procedures in front of you, polls are taken and you respond to the polls, see what the values are for the part of the software that you are looking at. In that way it is sort of calming, because it is a routine that you have been doing this whole time.
“At this point it's all happening fourteen minutes ahead of what we are seeing, so, for instance, at this point we have just jettisoned the cruise stage, etcetera, so things are actually happening right now.”
–08:40: The spacecraft was de-spun from its long journey and stabilized.
–08:10: The entry guidance system was enabled; the first two of the tungsten weights were shed. This meant that the center of gravity of MSL was now offset from the center of pressure, meaning that it could glide off-center on its heat shield, generating lift.
00:00: Entry interface—MSL plunged into the Martian atmosphere.
BOBAK FERDOWSI: “A mix of things had been responsible for launch, cruise, and approach, to make sure it all worked together,” he said, referring to all the parts of the flight right up to the time it entered the Martian atmosphere. “There was some relief when we got to EDL because much of the stuff I was responsible for was over, and I thought ‘Whew—at least that stuff worked!’ You never want to be the one whose thing didn't work. I became very focused on landing at that point.”
