Curiosity Read online

  Before the scientists finalized their selection of this rock, the rover shot the veins in it with ChemCam. The veins seemed to be something that had been wet in the past, and had the general makeup of gypsum here on Earth. Richard Cook, the MSL project manager, said in a NASA interview, “Drilling into a rock to collect a sample will be this mission's most challenging activity since the landing. It has never been done on Mars. The drill hardware interacts energetically with Martian material we don't control. We won't be surprised if some steps in the process don't go exactly as planned the first time.”

  Once Curiosity got close to John Klein, the DRT instrument was used to clean off a portion of the rock and Ken Edgett's MAHLI magnifying camera could be brought into play again. It was one more way to gain an understanding of what kind of rock this might be and how rewarding it could be as a test target.

  “MAHLI has several roles when deciding about drilling,” Edgett said. “One of them is characterizing the rock itself, as in, is this the stuff we want to drill? In this case at Yellowknife Bay, after they brushed the rocks, they cleared away the dust, the MAHLI could look at it and we could see that it was very fine-grained. So fine-grained that you really couldn't see the individual grains in MAHLI images, which told us the grains were smaller than fifty microns—how much smaller we really couldn't say. We don't have a microscope, but that was good enough to say ‘We have a rock where there are no grains larger than fifty microns, so it is very well sorted and very fine.’” This is referring to the evenness and smallness of the little bits of sand that make up the rock, not how pretty it is…“That leads you down the path of something like a mudstone because of the grain size.”

  Between this diagnosis of mudstone and the hydrated minerals in the veins, Klein was looking like an ideal place for the first drilling attempt.

  After taking some down time over the Christmas vacation, the crew was back and ready to drill in January. After a bit more reconnoitering, they performed pre-drilling tests on John Klein before boring into it.

  The first step was to check pressure loading on the arm. The arm operators pressed the drill mechanism up against a few different parts of the rock, then measured the pressure feedback on the robotic arm to see if it was equivalent to expectations. It was within spec. This also allowed them to make sure that the area was stable enough to be drilled—you don't want your target tilting or skittering away or rolling while you are pounding and grinding on it with the drill bit. In Klein's case, the rock was a flat tablet of mudstone, so unexpected motions were not a concern. But in future drill attempts, the motions of more complexly shaped rocks certainly would be, so it was considered a good idea to establish proper procedures at the start.

  The arm operators pushed the drill gently against Klein and left it there overnight to check out how much the thermal contraction that I mentioned earlier would affect things. It was possible that once you took the structure of the rover and arm into consideration, the whole thing could shrink as much as a tenth of an inch. Fortunately, for this kind of test, without a drill stuck in the rock, the most likely outcome was for things to pull away (contract), not push (expand)—the arm would contract up and back, simply pulling the drill head out of contact. When the rover and arm did expand in the morning hours, it would be returning roughly to the position it began in the day before. Dan Limonadi explained: “We don't plan on leaving the drill in a rock overnight once we start drilling, but in case that happens, it is important to know what to expect in terms of stress on the hardware…this test is done at lower preload values than we plan to use during drilling, to let us learn about the temperature effects without putting the hardware at risk.”

  Fig. 28.1. PLANNING: Drill placement was planned to the last detail on an image from MAHLI. Here you can see the drill hole's planned position (the circle at center) and the contact points of the stabilizing posts, or prongs (the two circles to the lower left and upper right). Above the lower left indicator are five tiny pebbles that were marked off-limits, as “Pebbles to Avoid.” Image from NASA/JPL-Caltech/MSSS.

  There were a few other considerations when picking the target sample. Though it is unlikely, there are some minerals in Gale Crater that might gunk-up the drill. Remember that the way the drill gathers a sample is not by coring, but by grinding some rock dust, walking it up the screw cutouts on the drill bit, and collecting it in chambers for sorting and sieving. This is all predicated on a dry sample—if it is wet, or acted wet or syrupy, it might clog the sample system, possibly permanently. Even dry minerals can act like wet ones under certain circumstances.

  The good news was that the vast majority of Mars is a dry desert and Gale appeared to be arid. Also, the rock in question did not appear to be of a type that, once ground, might behave poorly. The less good news: many of the rocks they wanted to look at, and possibly drill, were clays. Though these are outwardly dry, they can release water when drilled, and that would be risky. So there were many variables, and one always seemed to be tugging at the other.

  A note: While this was the first time anyone had drilled on Mars, other celestial bodies had been drilled in the past. The US Apollo-program astronauts took many core samples on the moon, some six feet deep. The Soviet Union's Lunakhod lunar rover did the same thing, and the Soviets even took some shallow cores on Venus. But this was a first for Mars and also for the US planetary-exploration program. A lot was riding on this first drilling event. Planners hoped to do at least nineteen more drilled samples after this one, and hopefully well more than that. Without any experience drilling on Mars, patience and caution were the watchwords.

  So, over the next few days, they conducted more tests—slowly and carefully. They applied more force to the rock and arm. They percussed the drill just a bit against the rock to ensure that the data looked right, and that loads were within expectations. They made a small, tentative predrill, which brought up just enough powder—onto the surface, not into the CHIMRA—to ensure that the consistency of the powder was what they wanted. This also ran the drill in both twisting and hammering motions, ensuring that it functioned as hoped when in contact with rock.

  Brian Cooper, the rover driver, recalls some of the stress he was experiencing: “We did take baby steps. Drilling [is] not something we do quickly because we are not able to do it very often. There's a whole crew of people whose entire focus is on the drilling campaign. We check for slip on the rover, slipping by the wheels, every day. We are very sensitive to that, and we've never seen any noticeable slip due to thermal cycling [heating and cooling of the rover and arm]. But we would always keep that in mind, especially when you have this drill in a rock and you have a thermal cycle, you're getting a lot of stress on it. So drilling is planned as a single-day operation, but you might be stuck in a hole due to a fault, or something happens to the rover…you have to assume that might happen; you have to program it so that you wouldn't [have this] happen [and to ensure that] the arm would be safe in that situation.”

  It was ten sols between the beginning of these tests and the actual drilling—and they had already been at the drill site for over a month. That's an abundance of caution. I suspect there was at least one engineer who wanted just one more test of something, but the vote was in: time to go ahead.

  The night before the big event, I had a chance to speak to Vandi Tompkins again, who was intimately involved with the drilling preparations, and she summarized the previous few days. “We'll be drilling this weekend! We have simulated the actual duration of the activity, and then there's all the time you spent coming up with the simulation. We have simulated and reviewed the entire operation.” She seemed very comfortable with the thoroughness of the preparations. “For drilling not only do we have to place the arm exactly, but we have to do a whole set of things. You take a MAHLI image…then you place the APXS instrument so they have the initial assessment, because drilling is investing a lot of resources, so we tackle those other measurements first, make sure it's worth it. And then you place the arm in po
sition just to preload it, you put pressure against it, so when you're drilling it won't jerk and slide away. All those things take time—anywhere from ten minutes to a couple of hours. Of course, the drilling itself takes time, too. The arm is very busy, not just for the MAHLI Mosaic [collection of images], but once you are sample processing, that can take hours of moving the arm around, getting it in the right place, when compared to MER, the arm motions are whole lot more complicated and [take] much longer.”

  It was February 8, or sol 180, and Curiosity was ready to drill at last; it would commence the historic operation tomorrow. But first, another long, cold night.

  February 9, 2013. Sol 181. Whatever you wanted to call it, this was the big day. The drill was in position, and the sequence had begun. Midspeed on the drilling, level 4 of 6 on percussion. The drill entered John Klein like a dentist working on a rotten tooth.

  Remember that this is a semiautonomous operation, and there was still a substantial delay between Earth and Mars as they tracked the progress of the drill. And with all those parameters to worry about…

  “We have accelerometers and an inertial measurement unit which monitor the pitch of the vehicle,” said Limonadi. “We also have a force sensor in the arm, and we have the engineering cameras that can take pictures of the terrain and then take a picture again five minutes later to see if the rover moved. So we are effectively using onboard sequences when we do contact science. We turn on sensors on board and if any of those three things, the accelerometers, the arm force sensor, or the visual geometry, change; if any those things indicate that something's changed, that would stop the operation.” He smiled tiredly. This may just be a bit more stressful than his search-and-rescue work for the county. “We just haven't done this before with the drill. It's a little bit like a mini EDL for us. My job as a project manager is to not break the sample system.” He's pretty serious about the last part.

  They didn't break the drill. In fact, it ground through the rock much quicker than anticipated, the stone was so soft. The resulting hole was a bit smaller in diameter than a dime and about 2.5 inches deep. As expected, the “tailings,” the powder from the ground and beaten rock, traveled up into the drill's collection chambers. That powder needed to be sorted and sieved, which was how the grains—dust, almost—would be reduced down to the 150-micron size the instruments needed. That meant delivering the sample to the scoop, and from there, to CHIMRA, where the sorting mechanism and vibrator were.

  Fig. 29.1. NO CUMBERLAND GAP: No gap here, the drill snugs right up to Cumberland, its second drill target. Drilling at John Klein, just a few feet away, looked about the same. Note the two metal posts on either side of the drill bit that make contact first to stabilize the turret and arm during drilling. Image from NASA/JPL-Caltech/MSSS.

  But the geologists were in for a surprise when they pulled the drill out of the hole. Since every part of the mission is recorded visually when possible, the sample itself was imaged.

  Ashwin Vasavada warmed to the memory: “Seeing the sample was the best moment for me, coupled with the science team's realization that it wasn't oxidized.” He grinned ear to ear as he remembered the day. “Mars before was red and now it's gray!” The contrast was even clear right there on the sampling scoop. “There was still some of the sample from Rocknest, some sand clinging to the back of the scoop. So you can see this bright red material we had sampled at Rocknest and then, right next to it, the gray material from John Klein. We said to ourselves, ‘It's different!’ That's why we have a drill to get inside these rocks where the oxidation wasn't there. That's the whole reason we did it, and it worked.”

  John Grotzinger remembered it as a high point, but one tempered with the patience he counsels; this is a story that is still unfolding: “When the picture came down and we got a look at it, I sent a message to the entire team and I said, “Look guys, Mars is gray!” To those of us who work on Earth [geologists], when you see a rock that you know has iron in it, as measured by the APXS, and now you've created a powder that's gray, there's a good chance that the iron is possibly not in a completely oxidized state. That's important because if there's fluid and if there has been organics, this is the kind of place that might be as good or bad as we're going to get for where the organics might be.”

  He continued: “We reached the total depth in six minutes. That's like a knife through hot butter when you are talking about rocks.” You have probably never tried to drill into a rock, but even with this high-tech drill, it should have taken longer…“so we knew it was ‘gray Mars’ and we knew that it was soft. If you put both of those together, everything about it says mudstone.”

  To most of us, mudstone would be just that: hardened mud. But to a geologist, it implies so much more. As Grotzinger said, “where there is mudstone, there could well have been a lake and a habitable environment.” Bingo: the bills for the MSL mission had just been paid with this single successful drilling. Gray rocks and soil equal a whole different Mars way back when.

  “When I saw that and saw those data sets, I thought ‘as far as this mission goes, we are golden. We've hit the jackpot.’” Grotzinger summarized. “We really couldn't ask for much more. We can turn the rover around and drive to the mountain to find the hydrated minerals, but everything that we had just found—right here—looks spectacularly good. The one exception is that we don't know if any of the minerals are hydrated. So we are not going to know until we get that powder into CheMin.”

  So off the powder went for analysis. Not only was the sample hydrated (that is, containing water molecules), but the surrounding region, as explored over the next couple of weeks, was as well. They had, quite literally, hit pay dirt. Gray pay dirt.

  But I am jumping ahead a bit. Within three days, when bits of sample powder had been delivered to both SAM and CheMin, Curiosity's mission was back in prime-time status with the public, really for the first time (other than the “Martian Mystery” debacle) since landing. Then, right in the middle of the analyses, a problem; Curiosity went into “safe mode.” It's very much what it sounds like, a bit like when Windows boots into safe mode after it detects a problem.

  The problem was noticed on the afternoon and evening of February 27, when the rover failed to report in, uplink data, and go into its overnight sleep mode. The engineers switched from computer bank A to bank B again. NASA sent out a pun-drenched tweet on Curiosity's feed on @MarsCuriosity: “Don't flip out: I just flipped over to my B-side computer while the team looks into an A-side memory issue.” It could have been advice to themselves—computer glitches are always a cause of deep concern.

  It was those pesky high-energy particles, radiation from space, that constantly pour onto the Martian surface. Even with the radiation-hardened computers, occasionally a stray particle would smack into a 0 or a 1 in the binary-coded memory and cause corruption, and that's all it took to make Curiosity take a break to figure things out.

  By March 4, they had things back on track. “We are making good progress in the recovery,” Richard Cook said in a NASA press release. “One path of progress is evaluating the A-side with intent to recover it as a backup. Also, we need to go through a series of steps with the B-side, such as informing the computer about the state of the rover—the position of the arm, the position of the mast, that kind of information.”

  As the troubleshooting continued for the next two weeks, on March 16, another hiccup occurred: now the B-side computer was having memory fits. It too went into safety mode, leaving no confirmed backup computer. This said, the A-side had been fixed and was almost completely certified to go back into operation, but it was an uncomfortably close brush. If both went out at the same time, things would have been more challenging, or interesting, to say the least.

  While all this was going on, there were still other science duties to attend to. The science team wanted a close look at that wonderful little drilled hole that had provided so much excitement. They turned to MAHLI.

  “With the drilling, of course, we can only
see down to the bottom that is visible,” Edgett explained. “It turns out that when you look down the hole, you see a bunch of dust down there, which is the tailings or cuttings from the drill.” These deposits filled the borehole to about halfway of its total depth, but they could still see enough. “From the Mastcam and the ChemCam, you could see in [the] walls this sort of a lightning bolt, a light-toned, zigzagging thing. Also, knowing that this depth was about three centimeters, they were able to accurately target their laser with ChemCam, going down the wall.” They would be able to test at least the half of the hole that was visible to MAHLI. “We were able to see what made up these white vein minerals in the hole, and how much was something else. They also wanted to look down the hole to see if there were any layers. If there were fine, thin layers of sediment, that is there we would have seen them. So we ended up looking at it with the MAHLI, obliquely from four directions, so you can map the entire circumference around the wall.”

  Fig. 29.2. FIRST DRILLING ON MARS: The drill hole at John Klein. Subsequent to drilling, the shaft was shot multiple times with ChemCam (note the row of laser-burned pits), and the entire area was photographed with the MAHLI instrument. The drill hole is just slightly smaller than a dime. Image from NASA/JPL-Caltech/MSSS.

  While the engineers were wrestling with the computer issues, there was another headlight coming down the tunnel. This was sample aging: the fresher the material is when it is analyzed, the better. The balky computer was slowing progress on sample analysis, and in April, just a few weeks away, solar conjunction would be upon them, when Mars would pass behind the sun as seen from Earth. While limited commands and monitoring are achievable during this period, “best practices” dictate shutting down and taking a break. Any transmissions could be corrupted, and it's just not worth taking the chance unless there are no other options. They needed to get things back on track as soon as possible.