Red Rover

by Roger Wiens

  Once again our backs were to the wall. There was nothing we could do to fix the situation. I was praying desperately that something would happen. Little did I know how soon relief would come!

  That week turned out to be quite eventful in Washington. News that the twin MERs might be shut down touched off a wave of protest among the science community. Apparently Dr. Stern had gone too far. Late the next day, there was a rumor afloat that he would step down. I couldn’t believe it. But an official announcement came the next day, and the day after that I got a call announcing that ChemCam’s full funding would be restored. The tide started to turn.

  In the meantime, we got to watch one more spacecraft head to Mars. The Phoenix mission had beat out SCIM with the promise of using already-built hardware from the abandoned 2001 Mars mission. Phoenix’s destination was the frozen northern plains of the Red Planet, where vast quantities of water ice had been recently predicted to repose under a thin layer of soil. Orbiting neutron and gamma-ray detectors had seen the water signal from orbit just a few years earlier, but no mission had ever actually touched ice on another planet. Proving water was there would revolutionize our understanding of Mars, which was thought in the 1990s to be a desiccated planet.

  Phoenix was launched in August 2007 and landed the following May. ChemCam’s instrument scientist, Diana Blaney, also played a leading role in Phoenix, so I got regular reports on how the mission was going. The mission was relatively low in cost and was necessarily of short duration, as the short boreal summer of endless sunlight would turn to fall and then winter. The temperatures would plummet to below –200°F, and the solar panels would no longer provide the power needed for the lander.

  The first few days after the landing were very exciting. The retro-rockets, which slowed the craft to a soft touchdown, uncovered large patches of smooth white material—ice—just underneath the surface. The robotic arm was also able to scratch away the soil and reveal whitish material below it. In this way the massive ice sheets underlying the soil on the northern plains of Mars were confirmed. A number of other discoveries were made regarding the composition of the Mars soil. A mystery dating back to the Viking lander days was solved, as perchlorate chemicals found in the soil by Phoenix could explain the strange results of a life-detection experiment performed during the 1970s mission. When the soil was wetted in that experiment it had given off oxygen, suggestive of living organisms. However, perchlorate chemicals, which are toxic and were now found by Phoenix, yield the same result. Overall, Phoenix looked like it was off to a great start.

  However, as the summer wore on, the relatively small Phoenix team faced more hurdles with its software and hardware, issues that it could have solved if it had had fewer cost constraints—if it could have been better prepared. Some NASA officials, particularly at JPL, were worried that MSL might be headed for the same fate.

  chapter

  sixteen

  ROVER MOTORS

  THE IDEA FROM THE BEGINNING WAS A GOOD ONE: BUILD A program that develops technology incrementally from one mission to another. The MSL rover was intended to be just that—a follow-on to the MER twins with new-technology features to increase the capabilities.

  The MERs were just a start. The size and weight of golf carts, they used what little energy they could get from the Sun to drive a few feet each day, utilizing their relatively small instruments to take pictures and make occasional measurements. They had to touch down in an extremely large, flat ellipse approximately 20 miles wide by 80 miles long because the landing sequence was quite inaccurate. This area had to be as devoid of boulders and geological formations as possible; one didn’t know where the package would end up, and the rougher the surface, the riskier the landing became. So the engineers were looking for a huge “parking lot” to target, in the lingo of NASA. With the combination of a large parking lot and the ability to traverse a few miles at most, these rovers would clearly not have access to the most interesting Martian features.

  The mission planners had to make two specific advances to guarantee more interesting terrain—and potentially more interesting science: the ability to set down in a much smaller ellipse, and considerably longer traversibility, so that the vehicle could drive out of the parking lot to get to the really appealing geology. With enough study, NASA could try a guided entry, which would bring the ellipse down to around 12 miles.

  To guarantee a drive long enough to get out of the parking lot in a reasonable amount of time, the new rover would use nuclear power in the form of a radioisotope thermoelectric generator (RTG). This device produces a small amount of heat, which is converted to electricity. Such power packs had been used ever since the 1960s on spacecraft for which solar power was considered insufficient, including the first Mars landers in the 1970s. The decision to use an RTG on MSL harked back to pre-MER days, when it was thought that the solar panels would become hopelessly covered by dust after a few months. It turned out that dust devils and wind gusts periodically cleared off the MER panels, allowing the twin rovers to last many times longer than originally expected. However, an RTG would still provide much more power than solar panels would.

  A couple of other features were intended to make MSL more versatile. NASA wanted the rover to be able to land anywhere between –60 and +60 degrees latitude and to operate at almost any temperature encountered in this latitude band. On Earth, that would be the equivalent of equipping an expedition to operate anywhere from the Sahara Desert to the ice cap of Greenland. Previous Mars landers had been limited to areas very near the equator, excluding nearly 80 percent of the potentially interesting landing sites.

  Couched in these new capabilities were a number of technologies that had to be developed. Some of the changes involved increasing the size of various assemblies, particularly the entry, descent, and landing hardware. Every vehicle destined for the surface of a planet having an atmosphere needs an entry capsule and a parachute to slow it down from interplanetary cruise to subsonic speeds. For the new rover, both the chute and the entry capsule would be the biggest yet—even larger than the three-person Apollo capsules used for the human missions to the Moon. At one point in the development of the new capsule, the MSL engineers became worried that they couldn’t just increase the size of the MER capsule and use the same heat-shield technology, so they ended up having to switch approaches in midstream—at a significant cost hit.

  Another area where size caused a radical design change was the final touchdown. Mars has an atmosphere only 1 percent as thick as Earth’s, so the descent speed is too fast to survive parachuting to the ground even with a very large chute. Just before hitting the ground, the previous rovers had deployed airbags, which allowed them, boxed up, to hit and bounce until the energy was dissipated. But there was a limit to how much weight an airbag could take. The engineers determined that the MERs, at around 400 pounds each, were somewhere close to that limit. MSL was five times heavier. One can’t just expect something the size and weight of a car to bounce around on airbags, even under the half-scale gravity of Mars.

  For all these reasons, the new rover had to use a completely different landing technology. Retro-rockets had been used on some other Mars spacecraft, notably Viking and Phoenix, so the engineers came up with a retro-rocket package and radar guidance system that would fit on top of the rover. But for MSL they didn’t want to saddle the rover with these as permanent fixtures, so they devised a way to get rid of the retro-rocket package in midair. As the rover came close to the ground, it would be lowered from the hovering package on cables until it sat on the ground. Once it was firmly on its wheels, the cables would be cut, and the retro-rockets could be dumped by flying away in one direction or another. Sky Crane was the name originally given to this invention. Watching an animation of it was really scary! Would it actually work? In spite of the questions asked by those of us on the payload, its development seemed to be proceeding smoothly.

  The RTG could also warm the rest of the rover with its waste heat. This was important, e
specially if the rover might go all the way to 60 degrees latitude. The only rover parts that would not be heated in this way were the appendages—the mast, the arm, and the wheels, where most of the rover’s electrical motors are located.

  Each wheel has one motor to power it, and four of the six wheels have motors to steer the rover. The arm has motors to power the shoulder, elbow, and wrist, as well as some of its devices, such as the drill and brush. The mast has two motors: one for turning side to side and one for pointing up or down. Besides these, there are the antenna motor and a few motors for onetime deployments of various things, including the wheels and mast, all of which start out in a stowed position.

  How cold would these motors have to operate? It turns out that there is a hard limit to the coldest the Mars atmosphere can get. At around –140°C (–220°F) the atmosphere, consisting of 95 percent carbon dioxide, starts to freeze into dry ice, keeping the temperature from plummeting further. The bottom line is that the motors would have to work down to the freezing point of CO2.

  Cold operation of the motors was a new technology. The MERs had relatively small motors that were heated when needed. The numerous and much larger motors on MSL together constituted a lot of mass, which would require a lot of electrical energy if heating were needed, hence the desire to operate these units at any temperature. Right at the start of the MSL development, JPL contracted with the company that provided the MER motors to build and start testing motors down to the temperature of dry ice on Mars. After more than half a year of building and testing, we were told that the first test had been a failure; a new contract would be required for the company to try again. However, the subsequent rounds of testing didn’t go any better. The motors always degraded and failed. So in 2007 the project opted for Plan B—to use conventional lubricants and equip the motors with heaters. This meant that the rover would have to wait until later in the Martian day to operate. It would use the Sun in addition to its electrical heaters to warm its extremities.

  Unfortunately, even under Plan B the failures continued. The motors were supposed to be delivered to JPL in the early summer of 2008 at the latest, but that didn’t happen. Various mechanical assemblies at JPL now fell behind schedule, because the motors that were integral to these assemblies had not been delivered. As summer wore on, JPL sent more people to visit the manufacturer. By the end of summer, the situation seemed rather dire.

  For a successful Mars mission to take place, a lot of tests have to be performed successfully after the last parts are delivered and the whole spacecraft is assembled. Some of these must happen in different configurations—rover only, rover bundled up inside the spacecraft that takes it to Mars, and so on. Then all the hardware must be shipped to the launch facility in Florida, where a number of exercises are repeated for the last time before the whole thing is stacked up and the countdown begins. Failure of any system anywhere along the way can result in days or weeks of delays. A reasonable amount of time to expect between assembly completion and launch for a project this complex is at minimum a year.

  Discussion about possibly delaying the mission had already started when Dr. Stern was still at the helm of NASA’s Space Exploration. At that point the reason had been to reduce the yearly cost of MSL and spread it over a longer duration. However, if the launch was delayed, the overall cost would increase significantly, so the idea was shelved.

  A delay for a Mars mission is not an arbitrary thing. As mentioned earlier, the Earth, with its orbit closer to the Sun, passes Mars about every 27 months, at which point the two planets come within 30 million to 60 million miles of each other. A launch during a planetary convergence allows the spacecraft to arrive at Mars in a little over half a year. At other times, Mars can be well over 100 million miles away from Earth, so it takes many times the energy and months longer to travel between the two planets. If MSL was late, it would have to be delayed for more than two years, not just a few extra weeks to give everyone time to finish. Maintaining the systems and know-how over this extra length of time would cost a bundle. This was a strong incentive to keep the launch schedule on track if at all possible.

  As the time became more squeezed, JPL managers went back to NASA to ask for a large amount of money to add hundreds more people to the project. NASA hesitated, but it was clear that a long launch delay would be even more costly. The real question was whether the project would actually be ready on time even with the extra staff. JPL had been through tight squeezes before, including on the MER project, when two spacecraft had to be readied. So everyone decided to go for broke, but the project leaders agreed that if worst came to worst, they would be sure to consult NASA at the first possible moment that a slip appeared inevitable.

  Various other aspects were moving forward. One important detail for the public was what to name the rover. NASA has a tradition of calling its missions by a generic or technical name during development, but finding a charismatic name the public can relate to when the mission really becomes public, typically sometime within the year before launch. For the previous rovers, the technical name had been Mars Exploration Rovers (MERs), with code names A and B for the twin units. However, a naming contest had resulted in the much more exciting monikers Spirit and Opportunity. So, with the launch expected within about a year, a big contest was held for the new rover. The selection was done by Disney to make sure that the title would resonate with the masses. Out of nine thousand entries, Clara Ma, a sixth-grader from Kansas, provided the winning entry with the name Curiosity. Yes, we would definitely be following our Curiosity as we explored Mars!

  Meanwhile, the motor testing continued. There were by now about as many JPL staff members crawling around the plant as there were employees—surely a frustrating situation. Then, in the last week of November 2008, another motor failed. The units were not going to be delivered anytime soon. Within days, NASA announced the launch slip.

  For the ChemCam team the news was bittersweet. We had assembled the instrument and sent it through all but the last series of tests. However, it had some issues of its own. Our backs had been to the wall with the schedule, and we knew we could use some time to check things out more carefully. Our team was very stressed and tired. After many months with only a couple of days off, I was ready for a little more calm. But two years of delay would be a long time to wait. Many things could happen between now and then. Team members nearing retirement might not be with us when the mission finally happened. Anyway, there was nothing to do but accept the delay.

  I went for a walk outside one evening during the week the delay was announced. The snow crunched under my shoes and I could see my breath as I looked at the night sky. The stars were shining brightly. Orion, the winter hunter constellation, had risen, and Sirius, the brightest star in the heavens, was twinkling above the horizon. Just above the snowy mountains ringing our town to the west were three bright celestial objects. Venus, Jupiter, and the crescent Moon were all passing right next to each other from the Earth’s vantage point. It was a glorious sight—the closest triple conjunction in more than forty years. Only one other planet shines brightly like Venus and Jupiter, and that is Mars, but only when the Earth overtakes it. I knew that would happen, just like after the SCIM decision, in just over a year. For the first time in over a decade, no spacecraft would be on its way to our neighboring planet.

  I made my way back to our house. Christmas was approaching and for once I would really be able to relax.

  chapter

  seventeen

  FINISHING CHEMCAM

  OUR TEAM ENJOYED SOME TIME AWAY AFTER THE LAUNCH delay—it was good to have the pressure off. But we weren’t done yet. Around the time the postponement was announced, we found several problems with ChemCam. In going through the final tests before delivery, we discovered that the instrument wouldn’t communicate when its data-handling electronics were below about –15°C (5°F). Even though the rover should normally keep this part of the instrument warm, under certain conditions it might end up colder than that on
Mars. The really frustrating thing was that we had tested the unit to much colder temperatures just two weeks earlier and it was fine then. We also noticed another troubling problem. Although the detectors were recording signals well when there was plenty of light, such as when the rock we were zapping was close up, as the sample was moved farther away the counts suddenly went completely to zero instead of decreasing smoothly with distance.

  We fixed the cold-temperature communication problem relatively quickly. But the detectors had us really scratching our heads.

  At first we didn’t have a clue what was going on. Could this be a feature of our LIBS technique? No—although the technique was relatively immature, we knew we should be seeing the small signals. The data were clearly being lost somewhere, but where—in the CCD detectors, or in the electronics that converted the current to digital counts, or in the computer’s memory? We ruled out the memory quite soon. At least the instrument wasn’t losing its mind! So we had two possibilities left. I thought it was quite unlikely that the CCDs could be faulty. We had bought them from a good company and we were using not one, but three of them. It seemed improbable that all three were bad. It was more conceivable that either our converter electronics were bad or that we were running the detectors improperly. Either of these seemed possible, as our team still did not have much experience operating these devices.