Season 1, Episode 9: InSight's Insights
InSight Mission Controller: Atmospheric entry on my mark: three, two, one, mark.
[00:11] Narrator: On November 26, 2018, the Insight mission arrived at Mars after a 6-month journey of over 300 million miles.
Tom Hoffman is the project manager for the mission.
Tom Hoffman: On landing day for InSight, I was in the Mission Control Center here at the Jet Propulsion Laboratory. Being the project manager, I was sitting in the back. All the management from both JPL and NASA were sitting back there with me. And it was a little bit nerve-wracking because they're all looking at me for what's going on, and I don't know any better than what the call is in the room at any given time.
Mission control is a complete misnomer in that particular situation. Because the reality is, our planned time for entry, descent, and landing was a little over six minutes. One-way light-time at that point was eight minutes. So whatever happened on Mars had already happened. And we were observing the past.
I wanted to actually have scenes of kittens put up on their monitors to keep everybody calm, but that was voted down.
InSight Mission Controller: InSight should now be experiencing the peak heating rate. Portions of the heat shield may reach nearly 3,000 degrees Fahrenheit as it protects the lander from the heating environment.
[01:18] Narrator: As InSight barreled down through the planet’s atmosphere, it had to perform every aspect of the landing maneuvers perfectly, in order to arrive safely on the Martian surface.
InSight Mission Controller: InSight is now traveling at 1,000 meters per second. Once InSight slows to about 400 meters per second, it will deploy its 12-meter diameter supersonic parachute.
[01:37] Narrator: Sue Smrekar, deputy project scientist for InSight, was also in mission control that day, hoping this mission would succeed where others had failed.
Sue Smrekar: Yep. Seen many disasters in my day, and did not want to see one more unfold in front of me. You don't want to have to play out that tragedy in front of the camera, if you can help it. Not that that's the most important factor, but it adds a little extra element of anxiety.
InSight Mission Controller: Once the radar locks on the ground, and InSight is about 1 kilometer above the surface, the lander will separate from the backshell and begin terminal descent using its 12 descent engines.
[02:14] Narrator: Bruce Banerdt is the lead scientist on the mission.
Bruce Banerdt: I was in the back row of the, quote, control room. We're all just watching, and listening, and hoping. I was nervous, obviously, but I was pretty hopeful that everything would go well. It's always the nagging fear back there that things just blink out and you'd never hear from it again.
But we were getting the telemetry, and every time something good happens, that's one less thing, or 150 fewer things that can go wrong. And so, you keep on ticking off, okay, well, that went well, there's fewer things that can go wrong now.
InSight Mission Controller: Lander separation commanded.
Altitude 600 meters.
Altitude 400 meters.
50 meters, constant velocity.
17 meters, standing by for touchdown.
Touchdown confirmed! InSight is on the surface of Mars! (applause)
[3:35] Narrator: Welcome to “On a Mission,” a podcast of NASA’s Jet Propulsion Laboratory. I’m Leslie Mullen, back with a bonus episode for our first season. We’re going to find out what the InSight mission’s been up to since it landed on Mars.
InSight’s landing on the Monday after Thanksgiving gave the scientists a lot to be thankful for: the mission they’d dreamed about and sweated over for so many years was finally on Mars. Tom says they even set a new record for the fastest successful landing.
Tom Hoffman: We came in at 5 minutes, 53 seconds. So it was about 30 seconds faster than we had planned. We spent a lot of time trying to figure out exactly why that was, and the short answer is the upper atmosphere of Mars was a little bit different than we had predicted. We came in a little bit quicker, because it was a little less dense.
As much as we've studied Mars and done things on Mars, we've only landed just a couple handfuls of times. Understanding the atmosphere is still something we're trying to grapple with.
[4:34] Narrator: Two suitcase-sized spacecraft, named MarCO, had flown behind InSight all the way to Mars, and they played a vital role on landing day. InSight and the other landers and rovers on the Martian surface generally can’t send data back to Earth directly. Instead they send signals up to satellites that orbit Mars, such as Mars Odyssey, which then relay that information back to us. For InSight’s landing, the Mars satellites weren’t positioned to send us that data right away. So the MarCOs acted as temporary relay stations as they flew past the planet. Bruce says he was on the edge of his seat, waiting for the MarCOs to send the first photo that InSight took of its landing site on Mars.
Bruce Banerdt: Within a couple of minutes we started getting back that first picture from our ICC, our instrument context camera, and that was super exciting, you know, we've already jumped up and down and high fived and hugged and stuff like that, and then hustled back over to the console for the image processing lead Hallie Gengl. Justin Maki and I were hanging over her shoulder and she was plugged into the feed from MarCO, which was still communicating up to seven or eight minutes after landing, and then we started to get the image down. It starts to fill in on one edge of the image and kind of moves across, as they get a little bit more and a little bit more data. And we got sort of like the first 10th of the image, and it was just reddish, muddy looking, it just looked like static, and we're going yeah, well, maybe this isn't going to be such a great picture. And then the next stripe came down, and suddenly, you can see the ghosts of the rock at the bottom of the image.
And as soon as we saw that, then we knew, yes, this was a real image, we have a picture from Mars. And then the rest of it started filling in faster and faster, and we could see the horizon.
And it was funny, because, Justin and I, like I said, we're kind of hanging over Halley's shoulder, and as far as I knew, everybody was off doing their thing in the room. But later there's actually a photo that ended up in a lot of papers across the world, and there's like a dozen people crowded around behind us. And I had no idea anybody else was there, it was just that little bubble with the three of us sitting there looking at it, trying to figure things out. But I think half the room was huddled behind us, and I had no idea.
[6:44] Narrator: Some would say the hardest part was over. But the mission still had many hurdles to overcome. For one thing, the solar panels had to open.
Bruce Banerdt: We had to wait for about six or eight hours before we had confirmation that the solar panels were out. And that was the last nerve-wracking thing. We're on Mars safely, but our batteries are only going to last for 24 hours or so. We went into a radio blackout once the MarCOs went out of range, we had to wait for Odyssey to come back around for its communication passing over the lander to see whether the solar panels had deployed.
And so, we had a little bit of a break where we had our press conference. And I almost missed the press conference because my family was viewing the landing with the science team, and then once we’d finished up at the mission control and I went and I grabbed them, bringing them over to the press conference, and the guards are going, “Oh, wait, they don't have a press badge, I can't let them in.” I said, “Well, if they're not going in, I'm not going in.” They're just doing their job, but I wasn't in the mood to deal with bureaucracy at that point. And so, I kept on looking around for somebody with the authority to let these dangerous family members into the press room. And I finally found one of the lab managers, Richard Cook. I said, “Richard, can I bring my family?” And he says, "Well, yeah, of course." So I actually walked in about three minutes, I think, before we had our press conference, which was a big party.
[NASA InSight Mars Landing press conference]
Tom Hoffman: Bruce Banerdt is our principal investigator. We have been working together for the last seven years to make this a reality. Bruce has been working for decades, and I’m so excited for him that he’s finally going to be getting his science back that he’s been working so long and so hard far for. So Bruce?
Bruce Banerdt: Thanks Tom.
Bruce Banerdt: Well, I can’t tell you what a privilege it is to be up here today. People keep talking about my science, and my mission. But this is really something that we’re doing as a science team for the world.
[8:47] Narrator: After the solar panels opened, InSight could sit back and admire the view. But the first photo from the ICC camera on the underside of the lander was heavily clotted with dirt.
Bruce Banerdt: I think just the high pressure from the blast of the landing rockets was enough to force some dirt under the dust cover. And so it was still almost as dirty when we opened the cover as before, which was a surprise and a disappointment. But fortunately, over the last months, almost all that dirt's come off the camera. And so it's now giving us a really clear picture.
[9:17] Narrator: InSight’s landing rockets also pushed a rock about 3 feet, or 1 meter. That’s the farthest we’ve seen a rock roll while landing a spacecraft on another planet. Team members named the golfball-sized rock, “Rolling Stones rock,” after the rock band.
(“Around and Around,” by The Rolling Stones)
[9:51] Narrator: Sue says the blasts of the landing rockets did more than just dirty a camera and kick a rock.
Sue Smrekar: The rockets dug some fairly deep holes under the lander. They are about 20 centimeters, so about eight inches, but nothing actually out of the ordinary. And it gives us a little window into the surface.
What we saw was a layer an inch or so thick, it's commonly called duricrust, and it's basically a region near the surface where there's been kind of a cementing of the grains together. You see this out in the desert, too. Often it's because as you have a day-night cycle, the atmosphere is cycling in and out of the soil, right? And it helps introduce minute amounts of water, things that can chemically cement things together. So that's a common thing on Mars.
We came down in exactly the terrain that had been predicted. Matt Golombek can put another notch in his belt, that he got a safe lander down.
Matt Golombek: And of course, one of the first things you do when you start looking at the surface, what did we get right, and what did we get wrong?
[10:58] Narrator: That’s the Mars landing site dude himself, Matt Golombek.
Matt Golombek: For me, as the landing site dude, it’s not just the knowledge that you landed safely, I need to see the picture as fast as possible to see whether or not the surface looks some semblance of what we expected it to look like.
Everybody else was jumping up and down when they got the telemetry that said that it had landed safely. But I was waiting for the first picture, which took a little bit later. And that’s when I jumped up and down, because it was pretty much what we expected.
[11:34] Narrator: InSight needed a flat landing site that was relatively free of large rocks. The landing ellipse, the area where InSight could possibly come down, given all the uncertainties of parachuting through the atmosphere, was a huge 130 kilometers across -- about the size of Los Angeles.
Matt Golombek: Exactly where you wind up in that giant ellipse matters. There’s some big craters in that ellipse, and not all that far away, that we could not have survived in. If we had landed on the inner slope of one of those big craters, that would’ve been death for the lander. When you select the site, you say, well, the percent area covered by those dangerous places is less than one percent out of the total. And then you say, ok, we have a 99 percent chance of success. And that’s kind of as good as you can do. So there’s always going to be things in the ellipse that are worse than others.
So then the question is, how lucky were we, right?
It turns out we came down in a rougher part of the ellipse and a rockier part of the ellipse than most of the average of the ellipse. But the individual location where we came down was this smooth terrain, which is like most of the rest of the ellipse.
There was less than a degree slope at all on the surface. In the area that we were the most interested in, there was no rock bigger than, you know, a pebble. I mean, they were teeny, it almost didn’t matter.
So were we lucky that we hit smooth terrain? No! That was 90-something percent of the ellipse. Were we lucky to have hit smooth terrain in that particular portion? Ok! But you know, as they say, if you’re good, then you’re more likely to be lucky!
[13:30] Narrator: Using a robotic arm, the InSight lander had to place instruments on the surface of Mars: a seismometer to listen for Mars quakes, a wind shield to cover the seismometer, and a heat flow probe to take the planet’s temperature. Part of Matt’s job was to figure out the best places to put them. Because InSight landed in such a good spot, with no big rocks or slopes to complicate things, that was relatively easy.
Matt Golombek: It turned out we could put the instruments in the places that the instrument people most wanted them to be, which was away from the lander, and away from each other, to have minimal noise.
[14:05] Narrator: Moving the instruments off the lander was a slow and careful process. Here’s Tom Hoffman again.
Tom Hoffman: Before we get to that step, we have to test that. So we have a testbed here at JPL that we set up to look pretty much exactly like Mars. So we practiced to make sure that we actually fully understood what we were doing before we try it on Mars, because you don't really get a second attempt.
So it ended up taking us about 87 sols. We call days on Mars sols. It seems like taking three elements and putting them onto the surface of Mars shouldn't take 87 days. We wanted to be sure that we're doing things safely and carefully, so it did take us a fair amount of time to get that done.
[14:41] Narrator: The robot arm’s grapple hand was designed to pick up each instrument by a small lollipop-shaped structure sticking up out of their tops.
Tom Hoffman: We call those the grapple hooks. Basically, they're the point where the robotic arm's grapple grabs. That's what we have to lift it with. And it's not a very large structure. It's pretty thin. It's thinner than a pencil but it is made out of titanium, so it's pretty strong.
But after we had launched, we had brought a flight spare seismometer, so something that's built exactly like the flight unit but we didn't fly, here to JPL, and in the process of doing some very simple testing, we snapped off this grapple hook.
And so that was a big scare for us, because we're thinking, if that happens on Mars, our mission isn't going to work. And that would be a total disaster having this $10 part cause the whole mission not to work.
[15:35] Narrator: After months of investigations, they were able to figure out what caused the thin stem of the grapple hook to break in the test bed.
Tom Hoffman: We actually had a little bit of a nick in it from some of the tooling, and that created a stress point, and the stress point is what broke. And we're able to show by looking at the procedure and talking to the person who built all of them that that one was a little bit different than the other ones. And then, we went through the process of building a whole bunch more of them using both processes and tried to break them, and it turns out that, I think, the one that broke was just really, really bad. Because even the other ones where we tried to make them bad and break them, we couldn't break them. I mean, we kind of got lucky that we had one bad one, and that was the one that didn't fly.
It ended up not being that big of a deal, but we did spend a couple of months thinking really hard about it between the time we launched and the time we landed, and trying to prove to ourselves that everything was okay, and thinking, well, even if it isn't okay, there's not much we can do other than try anyway.
There's any number of other problems that we could have had too. Each of the devices are secured to the deck with basically little bolts that have to activate and release. It's like a phase-change material, basically a thermal expansion, we heat it and it expands and then breaks it, and then you can lift up the device. Even though we've done lots of testing of making sure that they release every time, you just never know. Something could get cockeyed in there. We did a lot of testing on Earth, but Earth is not Mars. You always hold your breath whenever you do something for the first time for sure. But in the end, they all worked great.
[17:03] Narrator: One of the most delicate operations was placing the seismometer – the sensitive instrument that would measure even the slightest vibrations caused by quakes on Mars.
Tom Hoffman: When we put the seismometer down, one of the things that we needed to do was activate this mechanism which allowed the tether to be released from holding on to the seismometer.
Both the seismometer and the heat flow physical properties package are tied back to the lander with these tethers. And the tethers are what provide the power to each of those instruments, and then, in turn, send commands and then also receive the data coming back from the instrument. So “tether” is a little bit of a misnomer. They're actually fairly thick, heavy, and hard to move, almost printed circuit boards, so they’re pretty stiff. They're not at all flexible like your iPhone tether. They're much, much bigger, a lot more signals and kind of a little bit of a pain, frankly.
Because even a little bit of perturbation of the tether -- shaking in the wind or just even expanding temperature wise, just like a nanometer -- creates a huge, huge noise source on our seismometer.
So we created this whole system to make the tether so that it wouldn't be pushing against the seismometer. We have a little loop in there that’s basically a noise shunt. Instead of pushing directly on the seismometer, any little vibrations or thermal noise or whatever is just pushing against a spring. And so one of the things we had to do is actually pull the tether away from the seismometer and towards the lander, which was not particularly easy.
It took us a bunch of tries to get it just pulled back enough, because the problem that we had is if we pulled it too much, there's no way to push it back. We kinda had to sneak up on pulling it a little bit. In fact, the very first time we tried to pull it, we didn't move it at all. The second time we tried to pull it, we moved it like barely. And then, the third time we moved it just about enough. In the end, that ended up taking us about two weeks.
[18:57] Narrator: All told, it took about two months to set up all the elements of the seismometer. Bruce says it took another month to make sure it was working correctly.
Bruce Banerdt: We were busy that whole time. There's just a handful of things you have to do, but each thing takes preparation, and then you have to check it out afterwards, and we're always trying to make sure everything is safe. We don't want to be swinging the robotic arm around and run into the antenna or punch a hole in the thermal protection of the seismometer. So, we checked everything in the test facility at JPL. Each one of those was round-the-clock preparation and work by the team to get it done. And so, that was an exhausting three months to get the seismometer working.
It was worth every minute because the seismometer has been performing amazingly well. I mean, we're getting seismic data at a precision that is, in some parts of our frequency bands, it’s 1,000 times better than anything's been done on the Earth, because of the background noise on the Earth.
No matter where you go on the Earth, you have a certain amount of vibration just from the storms in the ocean, and the waves beating against the shoreline. You can go to the center of Kansas, and you're still getting a pretty big signal from ocean turbulence. But we don't have any oceans on Mars, so we're actually seeing vibrations that have never been seen on the Earth, because you can never get that quiet on the Earth. And so, this is new territory for seismology in a technical sense, as well as in a geographical sense.
[20:29] Narrator: As noted in episode 2, Apollo astronauts had put seismometers on the Moon, which also doesn’t have oceans.
Bruce Banerdt: The Moon seismometers were very similar, but that was using 60s and 70s technology, so they actually got some noise that was as low as ours, but over a much more narrow frequency band.
Seismic signals have information at all different frequencies and all the different frequency bands have different kinds of information. Higher frequencies tend to die out with distance. Just like certain sounds that you hear from a long distance, like if you hear thunder from a long distance away, it's just a rumble. (thunder). If you hear thunder up close, you hear the crack, you hear the high frequencies as well. (lightning)
So the high frequencies in seismology tell you about things that are close in. But for more distant quakes several hundred miles to several thousand miles away, you're seeing lower and lower frequencies as the dominant contributor. And so, you want a seismometer that measures all the different frequencies, all the way down to some frequencies that maybe takes almost an hour for an oscillation to finish, which is the frequencies at which the whole planet vibrates or rings.
[21:45] Narrator: In episode 2, Bruce had explained how quakes cause Earth to ring like a clear bell (ding) but the Moon to crash like a gong. (gong) He’d guessed that Mars would have more of a bell ring, like Earth.
Bruce Banerdt: There's some ring, but there's still some of that scattery kind of hissy stuff going on. The signals that we’re looking at are kind of halfway in between. So, on the Earth if you get a fault or get a crack, and then just leave it alone for a few hundred million years, there's water that seeps through it, minerals crystallize in the cracks, and you get annealing, or healing, of these cracks. And so, after some amount of time, another wave goes through there, it doesn't even see that crack anymore. It just passes through. (ding)
On the Moon, it's very dry, and so those cracks maintain themselves for hundreds of millions or even billions of years. And so there's lots of things that reflect and scatter the seismic waves. And so when you scatter a wave, instead of going straight from point A to point B, it bounces around and sort of takes a drunken walk, and finally maybe ends up back at point B, but instead of going 100 miles in a direct route, it may have taken 2, 300 miles to get to where you are, and so it comes in really late. (gong)
And so we've got Moon quakes signals that last for an hour or more, whereas the same size quake on the Earth would last for maybe 15 or 20 seconds.
And so, we're seeing some of the same type of scattering processes, we think, on Mars as we saw in the Moon. And especially for the close events, so we think that the outer portions of the Martian crust, maybe the upper 10 or 20 miles, might be more fractured and less annealed than the Earth, which tells us that possibly the Martian crust is drier than the Earth's crust, although that's still speculative at this point.
The few events we've had from farther away don't seem to be quite so scattered. A few events from over 1,000 kilometers away, those waves are dipping down into the upper mantle. So we think that deeper into the crust or into the mantle, things don't seem to be so fractured up.
Mars is between the Earth and the Moon in a lot of different ways, and it looks like this is just one more way that it really lies on that line.
[24:12] Narrator: Even though the picture of Mars is just starting to emerge, Bruce is happy to be seeing even small Mars quakes.
Bruce Banerdt: Again, we didn't start measuring at a level that we could detect them until about Sol 100, and then we went for another month with seeing nothing, which was a little concerning.
So we started watching and we're waiting for our first Mars quake, and we're waiting, and we're waiting, and after about a month we're kind of looking at our watch, and saying, "mmm."
And when you're going for weeks without seeing anything, you start to get a little bit fearful that well, do we really know what we're doing here? Are we really going to see something?
(Sol 128 Insight rumble)
So our first Mars quake that we saw was on Sol 128. And it turns out, it's very unusual, we haven't seen another one quite like that one since, most of the energy was at high frequencies and since then most of our quakes have been much lower frequency. Since then, we've been getting about one or two seismic events per week. Most of them very small, on Earth, they would probably be magnitude one to two.
We're actually only able to see very small Mars quake signals for part of the day, because during the day there's a lot of wind activity, and so the small quakes get drowned out by that noise.
We've found out very early on that the pattern of atmospheric noise is very repeatable every day. So, about six o'clock at night, it gets very quiet. And then around midnight, the wind comes up, and from midnight to about six or seven in the morning, the wind is blowing very steadily. And then the Sun comes up and you start getting turbulence and the noise just goes up quite a bit. It just keeps on going until about sundown again. And we can see that each and every day. So now we know the window to spend most of our time watching for Mars quakes.
One of the nice things about Earth seismology is you get dozens of nice earthquakes every day. We get all excited when an earthquake happens, but most of the Earth is unpopulated, there's oceans, there's wilderness, and when earthquakes occur out there, nobody but seismologists knows about it. And so, seismologists can spend a few months collecting data, and you'll have a few hundred signals to work from and lots of information there.
On Mars, the activity is probably something like a factor of 1,000 less. So, instead of having several hundred quakes over the course of a month, we have a handful, maybe a half a dozen. And so we have to wait a lot longer for the number of quakes. We have this giant thousand-piece jigsaw puzzle, and every week we get one piece.
So, we have to be really, really patient. And we have all these pieces that don't fit together yet. And we just have to assume that as we get more pieces, we'll be able to start making the connections.
It's funny, I started thinking about what it must have been like to be a seismologist back in the early 1900s, when people were first starting to piece together what was going on the Earth. For those first 10 or 20 years of the 20th century, you’d look at a seismogram and just scratch your head. What's going on underneath there? And we take it for granted you open up a textbook there's a crust, there's a mantle, there's a core, and they show you nice clean seismograms with all these wiggles, and this corresponds to this path, but somebody had to figure that out from scratch. And we're having to do that on Mars now, and it's not as easy as it looks.
[27:47] Narrator: The weather station on InSight helps rule out when a breeze or an air pressure event shakes the seismometer. By amplifying the vibration of InSight shaking in the wind, you can get a sense of the weather on Mars.
(InSight wind “sound”)
Bruce Banerdt: Plus, as kind of a bonus, we're actually able to see dust devils in the weather data. We have the pressure data and it's varying slowly during the day, and suddenly, it'll drop about several Pascals, and we interpret that as a dust devil passing very close to us. And it's like a hurricane, (wind) when you have super low pressures in the eye of a hurricane. You have very low pressures in the center of a dust devil because it's essentially a little vacuum cleaner, where the air from the surface is being sucked up higher into the atmosphere. And we can actually see that in the seismic data as well because as the dust devil is sucking up the air, it's actually like a vacuum cleaner pulling on the ground as well. And so we can actually see the tilt of the ground as the dust devil goes by.
And so we've seen hundreds of those things now. And by doing that, we actually get some information about the surface of Mars because the amount that it tilts is related to how stiff the surface is. If it's very stiff, it doesn't tilt very much, if it's relatively bendable, it tilts a lot. And so we're actually starting to measure the elastic properties of the upper few meters of the Martian surface by looking at these dust devil signals.
[29:16] Narrator: If a Martian dust devil flew right on top of InSight, what are the odds it would mess up the lander?
Bruce Banerdt: We were actually worried about that. We have our solar panels out there and they're relatively fragile. They're on pretty skinny little spars. And actually, if you look at the biggest dust devils that have been seen on Mars and you combine them with the highest winds that you see, that would actually be enough to rip the panels right off of our spacecraft. (metal breaks, wind fades)
But statistically, that's extremely unlikely… and so we beefed up our solar panels as much as we could within the limitations of mass and volume that we had to work with, and we believe that we can survive any likely dust devil or even some pretty unlikely ones. And so far, none of them have bothered us at all. But that's always one of the worries in the back of my mind that some really big dust devil comes by, and off our solar panels go to Oz or someplace.
(From the movie, “The Wizard of Oz”)
“Toto, I have a feeling we’re not in Kansas anymore.”
[30:19] Narrator: Another instrument on InSight, called RISE, is measuring how much Mars wobbles as it orbits the Sun.
Bruce Banerdt: The wobble of the Martian pole is tied very closely to the activity of the core. The amount that it actually wobbles back and forth depends on how much mass there is in the core sloshing around. We do think at least the outer part of the core is liquid. So that's working together with the seismology to look at the deepest structure of the planet.
[30:45] Narrator: All the instruments on InSight aim to help figure out what Mars is like deep inside. Quakes are the result of underground activity.
Bruce Banerdt: On Mars, we think that probably the basic process is the shrinkage of the planet as it cools. But it's more complex because there's volcanism going on, there may be a convection in the mantle that's pushing up some areas and pulling down some other areas and maybe even dragging some things laterally.
On the Earth, we have plate tectonics that are moving the plates around, either moving them past each other, at places like the San Andreas fault zone, or moving them apart from each other at spreading ridges or together at mid-ocean trenches.
In the end, even on the Earth, the forces due to plate tectonics are from the cooling of the planet. Plate tectonics is driven by heat bringing up hot mantle material to the mid-ocean ridges where it cools as it spreads and then the cold slabs sink back down into the Earth, cooling off the interior. And so, this is just the way that the Earth gets rid of its heat.
Every planet is a heat engine. What we're really trying to do is understand the various different heat engines that are going on. Earth has one that we're pretty familiar with, it's got lots of heat, it's a high-performance engine. The Moon might be a car in the junkyard, it's not really running very well anymore. And Mars might be a little bit more of a Volkswagen bug or something like that.
(car revs, stalls)
[32:14] Narrator: InSight’s instrument to measure the heat of Mars’s engine got stopped in its tracks right out of the starting gate. Sue says the heat flow probe, also known as the mole, ran into trouble the first day it tried hammering itself into the ground.
Sue Smrekar: It started going great guns for about 10, 15 centimeters, and then slowed noticeably. The thing that we noticed right away is that the support structure that is holding the mole in place until it gets under the ground, moved. Which was a surprise. It wasn't just hammering itself vertically down into the ground. It had tipped, tilted, and was pushing against the side of the support structure.
[32:53] Narrator: Scientists didn’t see this happening in real time. After the command to hammer had been sent to the mole, the results didn’t come back until much later.
Sue Smrekar: Yeah, because we have to send the data up to an orbiter (ding), the orbiter has to send it back to a Deep Space Network station (ding) somewhere on the Earth, and then it has to come to JPL (ding). So all of that takes ... In the best possible case it would be a couple of hours. Almost never do you get it that rapidly. Usually it's the next day that you get the data. That's in fact why everything takes so long, because it's basically two days to uplink commands, execute them, downlink them, and then analyze them on the Earth to build new commands, and send them up.
[33:32] Narrator: Looking over the data from that first hour of drilling, it looked like the mole had run into trouble about 15 minutes in.
Sue Smrekar: We learned that it went down about 35 centimeters, and then it just stopped progressing. And we tried a second hammering session as well, because we weren't clear about what kind of resistance it was actually meeting. We've of course tried to anticipate every possible scenario and test them out in the lab, and among the scenarios we've tested is, "What if it encounters a small rock?" And what we've seen is that if it's a small rock, if it keeps hammering, it can effectively push it to the side and keep going. We've also looked at big rocks, and what happens then. And in that case, if the rock has a slope on its top, the mole can actually kind of skitter around the side. If it hits a big, flat rock, then we have no recourse. We were hopeful that perhaps it was a small rock that we could push aside, or break, as we've seen in the lab. So we decided to try a second hammering session. But that did not progress any further. So we basically went into problem-solving mode, to try to figure out what could be causing this.
[34:37] Narrator: One idea was the tether that connected the mole to the lander had gotten bound up inside the support structure that the mole was housed in.
Sue Smrekar: There are these friction springs, and basically what their job is, is to hold the mole vertical while it's dropping out of the support structure, and to give it resistance. Because, it's a hammer. If you've hit a hammer, you know a hammer has a recoil, right? And so, you might think that the hardest time is when you get deep, and you're pushing against more weight of soil, but actually, the hardest part is getting fully into the ground. And that's because of this recoil. We rely on friction from the soil that it's moving through, on the sides, to stop it from bouncing back upwards. And so, these friction springs assist with damping that recoil while the mole is still in the support structure. And so, one idea was that maybe the mole was tilting and it had somehow snagged the tether inside these friction springs.
[35:40] Narrator: If the tether wasn’t holding the mole back, then perhaps the Martian soil was to blame.
Sue Smrekar: Another scenario was that the soil hadn't collapsed in around the mole, and it made a hole and didn't have any resistance to that recoil. So it was just able to bounce freely backwards. We thought that the soil would just flow in, based on our understanding of the properties of the Martian soil.
So basically, we got busy in the lab, and brought out all the spare hardware that we had, and tried out a whole bunch of things, to try to assess what we thought was most likely. And people pretty much ruled out the idea of the tether getting snagged in these friction springs.
Eventually we came to the conclusion that we really can't help the mole at all unless we lift the support structure and get a peek underneath to see what is going on. So in our test bed, they practiced picking up the support structure with the robotic arm, and lifting it. We did it in a series of steps because the risk was that we would pull on the tether and actually lift the mole out of the ground. That was our fear, right?
So we did it in little, small steps. You know, you send the command, take the picture, send the data back, analyze it. Slow process. And so, it was clear that we were not pulling the mole out of the ground, which is good. And so, through a series of steps we ended up placing the support structure off to the side of where the mole was stuck into the ground.
[37:01] Narrator: With the support structure off to the side, InSight’s camera could get a good look at what trouble the mole had gotten itself into.
Sue Smrekar: So, it revealed the pit. The mole is roughly an inch in diameter, and there's a hole that's several mole diameters wide. We've taken a ton of images at different times of day, so the light's at a different angle and so forth. So we've seen to the bottom of the pit, and it's about five inches. It's not super deep. Clearly the soil around the mole has gotten compressed, because, you know, where has that soil gone? That soil is compacted, either at the bottom of the pit, in front of the mole, or off to the sides.
In the end, what we decided to do was try to fill in that pit, and try to increase the friction on the sides of the mole, to damp out the recoil and see if that is the solution to getting it moving again. So, again, it took a lot of testing and analysis to make sure we could carry out a push on the soil without hurting either the mole or the arm. There's a scoop on the end of the arm, and it has the knuckle, the flat, and the tip of the scoop.
So, the first thing we did was push that flat scoop against the side of the pit, to see if just kind of pushing on it would collapse it. Unfortunately, it did not collapse it. And so, the next attempt was to take the knuckle of the scoop, and push that against the side. And again, not a lot of motion of the soil.
[38:27] Narrator: A straightforward strategy would be to scoop up soil and dump it in the pit.
Sue Smrekar: Well, in theory it could kind of scrape the soil. It would probably be easier to do that than to push the soil. But it's very challenging, because when we placed the support structure on the ground, our goal was to put it as far away from the lander as we could, because we wanted to keep the thermal effect of the lander as removed as possible. So we really stretched the arm out as much as possible, initially. So to be able to reach beyond the pit and scoop in, that's super challenging just from the standpoint of the arm. And also, there's this duricrust. We don't actually think it's that hard. You ought to be able to crumble it in your fingers. We're certainly hopeful that we can use the scoop to crunch it, and kind of break it into pieces, and allow it to start falling into the pit and getting some more of that looser material underneath to start filling in the hole, too.
[39:24] Narrator: The frustration of trying to get the mole going is magnified by how easy it could be to fix.
Sue Smrekar: If we could just go stick a finger on the back of the mole, you know, it would probably be enough resistance to get it going again. It really doesn't take a huge amount of resistance to help it move forward, because the recoil is much smaller than the forward motion. So some other possible scenarios are trying to put the scoop on the edge of the mole, or on the back, just pushing where the tether is. Pushing on the top, where the tether is, is the last-ditch effort, because the tether was never designed to be pushed from the back. And those connections are ... They're not delicate, but the interface between the tether and the mole has kind of always been a challenge, because you want to keep it sealed so that the dirt doesn't get in there, but you don't want to pinch the tether. And you don't want it to have too much force from the hammering pulling it forward. So that's kind of our Hail Mary, "Let's just push on the back and see what happens, and see if we can get it to move forward."
[40:31] Narrator: That simple act is tough to do with a robot arm.
Sue Smrekar: That's a tricky thing, the arm was never intended to be extremely precise in its measurements. And it was never designed to push against anything, you know? And so, basically, you command it to be against the ground, and it will kind of keep that force for a little while, but it doesn't maintain a push. It's just basically holding itself in that location. Which is different than pushing, right? And so, we think we can time it so there'll be some resistance while the mole is hammering, but not for many hammer strokes. So, you can imagine that's a very painstaking process. Because then, if it does move forward, then you have to move the arm forward a little bit more. And using the arm takes power, and as the season progresses, we have less solar power available. At some point, we could get to the situation where using the arm is impacting our ability to take seismic data.
We don't want to rush, and do something that's going to damage the mole and just ruin any chance of getting under the ground. But at the same time, we can't keep doing it forever.
[41:32] Narrator: To save the mole, they must bury the mole. Tom says the engineers have relished coming to the mole’s rescue.
Tom Hoffman: All the engineers were really excited. Because engineers generally either want to build stuff or fix stuff. And we'd already built something and it wasn't working. Now, we had a chance to fix it.
We're going to try to see if we can get some dirt into the hole, I call it the mole hole or the pit of doom. We can get around almost anything, including like concrete, as long as it's not flat. So hopefully, we're not hitting any flat concrete. It's a fun, fascinating problem to work on. Because if you think about it, you're trying to do something nobody's ever done before; nobody's ever done anything other than scratch the surface of Mars, literally. And we're trying to hammer into it. And we have no idea what we're hammering into. We're trying to figure out, okay, having no idea what we're hammering into, how are we going to get around whatever it is we don't know about?
And everybody who's ever dug a hole in their backyard and encountered rocks has infinite numbers of things that they can tell you about in terms of what you should be doing. But we don't have a shovel. We can't look in the hole that well with our one camera. There's only limited things we can do with our arm, which we never intend to do any of the things we're starting to do now in terms of scratching, or punching, or hacking, or chopping the soil.
[42:44] Narrator: Tom says InSight’s situation is the opposite of what they’d expected.
Tom Hoffman: The initial concern actually was, oh, it's a filled-in impact crater. It's going to be just really loose sand. For the seismometer, the concern was it's going to absorb a lot of the seismic energy. Because as it goes through the sand, it's not nearly as good as being on concrete. As it turns out, we're probably closer to concrete than sand. On the flipside, we thought, oh, this is going to be great for the mole. This is going to go through this like nobody's business because it's just sand. It's going to be like a knife through butter. Even the geologists looking at where we landed, knowing what they know about the surface of Mars, we now know we don't know that much about the surface of Mars. And certainly, we don't know that much about what's underneath the surface of Mars.
That's the interesting part about exploration when you're on even Earth or another planet. When you're trying to do something for the first time somewhere else, inevitably, something is going to surprise you, something is going to be different.
[43:35] Narrator: Even if they can’t get the mole to dig down into Mars, InSight is learning a lot about the planet’s interior, and the team expects to do so for at least another year. Tom says what we learn about Mars is just the beginning.
Tom Hoffman: Even just thinking about how we dug on to Mars and it was different than we expected, something as simple as that, you can think about all the different ways other planets could be so different than what we expect. Because we look around on Earth, and there's lots of different places. There's icy places. There's volcanoes. There's underwater trenches. That doesn't mean that those are all the same everywhere. And in fact, if anything, I think we've learned just from our little digging experiment that more likely than not, things are a lot more different than they are similar on other planets.
[44:24] Narrator: There’s so much out there we still want to explore. But for Bruce, who’s been working for over thirty years to send a seismometer to Mars, nothing compares to InSight.
Bruce Banerdt: Even though you're working towards this, and you're hoping for it, on some level, you think it probably is not going to work out. You got to try, but especially after years and years, you have to come to terms with the possibility that it just might not happen. And then it happens, and here we are on Mars.
I mean, the things that I've been dreaming about as possibly happening, here they are. And, of course, once you're there, it's exciting and it's amazing, and you just walk outside and look at the sky and find Mars, and think about your spacecraft up there.
[45:04] Narrator: In one way, many of us are up there on Mars, with InSight. The lander has a microchip inscribed with two million names of people from all around the world. Go to mars.nasa.gov for more details. You can also visit that web site to learn more about InSight, get Mars weather updates, and see all the raw images from the mission.
The next season of this podcast will be about a whole new topic. Come join us soon, “On a Mission,” a podcast of NASA’s Jet Propulsion Laboratory.
[run time: 45:47]