2009-05-22

Spirit Sol 137

This not-living-on-Mars-time thing is weird. RP-1s used to show up at 15:00 rover-time, RP-2s at 22:00 rover-time. I'm here as RP-2 today (Ashitey is RP-1 -- filling in for Chris, who has a visitor), and it's only 09:16 rover-time. It works out because the rover hasn't moved since yestersol (that's what restricted sols are for), and we used the time to get imagery we can use for planning thisol. But it feels so unnatural.

Ashitey has done about a third of the sequencing by the time I arrive, and he works on the second third while I tackle the, uh, third third.

Ashitey also conducts an intense series of cell phone conversations with Eric Baumgartner. Since winter is approaching, bringing shorter days and colder temperatures, our poor solar-powered rovers have to start making do with less energy. To help with this, Eric and others have worked out that we don't have to have the IMU on while using the IDD. The IMU -- Inertial Measurement Unit -- keeps track of changes in the rover's attitude (that is, its roll, pitch, and yaw). For example, if the rover is heading straight down a gentle hill, the IMU might report that we were pitched nose-down at five degrees. Well, there was concern that the rover's attitude might change when it extended the IDD -- doing this changes the vehicle's center of mass slightly, and could cause us to pitch forward a little, for instance. So to keep track of these changes, we've had a rule that we must turn on the IMU when doing any IDD motion. But the IMU is, relatively speaking, a power hog, and measurements have shown that the original concern was overstated. So the rules have changed: under most circumstances, we can leave the IMU off and save some power. So Ashitey was going to leave the IMU off during thisol's IDD sequences.

That, at least, was the plan. To set up for that, we already set some parameters on the rover that tell it not to worry that the IMU is off when the IDD is moving, and we got no flak about setting those parameters because the sequence that set them had already been run on Opportunity. But it turns out today that Opportunity hasn't actually used the IDD since setting the parameters this way -- so, in other words, the effect of the new settings hasn't been tested on the flight vehicles. They've sort of been tested on the testbed vehicles -- only one of them ever had an IMU, and that IMU was removed some time ago. But it's not quite the same thing, and after some discussion, Art wimps out. Because of our current weird timing, we'd lose two or more sols if something goes wrong with this. "We don't want to be the guinea pig on this one," he says.

"Well, how about leaving the IMU off for the overnight sequence only?" Ashitey proposes. It's a reasonable suggestion: this way, the sol-137 daytime IDD sequences would still succeed; only the overnight sequence could possibly fail.

Art muses, frowns, shakes his head. "No, we'd still lose that observation, which we'd have to redo," he points out. "Plus, we'd lose sol 138. It's not worth it."

The IMU commands go back in. Looks like Opportunity will have to be the risk-taker this time.

Meanwhile, I'm having some fun writing an IDD sequence for the first time since -- well, I can't remember. It's been long enough that I've almost forgotten how to do it, or so it seems at first. Turns out, though, that it's like riding a bike: you have to keep pedaling or you'll fall off and hurt yourself. Or something like that. I've kind of missed it.

The IDD work we're doing is all to explore a trench we dug the other sol. One of the targets the scientists want us to poke at is a spot called "Mill Basin," located on the trench's floor. When we show the animation at the Activity Plan Approval Meeting, Art doesn't like the instrument placement on that target. Not our fault, it just puts the instruments a little too close to the trench wall for his taste.

So we work with the scientists to choose another target that's a little closer to the middle of the trench floor, and when we place the instruments there, we have a new problem. When we switch from the MB to the APXS, we normally place the APXS 4mm closer to the soil than the last position of the MB. We have the MB touch the soil, but the APXS is about 8mm shorter, so we split the difference, leaving the APXS slightly off the soil (we don't want the APXS to touch the soil, as that can mess up the instrument permanently) but close enough to still get good data. (Originally, we didn't overdrive at all, and the 8mm standoff significantly reduced the quality of the APXS data we gathered in the early part of the mission.) It so happens with this new target that the usual 4mm overdrive is just a little too much -- it causes the IDD's elbow joint to extend just a tiny amount beyond its limit. Ashitey and I work out that 3mm is under the limit -- just barely -- and 4mm is over.

Seems like the solution is obvious -- just back off to 3mm. But we don't want to compromise the quality of the APXS data. Another concern is that the 3mm overdrive still leaves us awfully close to the elbow joint's limit, and we're not sure whether we need to leave more margin. The good news is that we probably won't actually come close to the limit -- the simulation is assuming the trench is about 1cm deeper than it probably is, so the position we actually contact will probably be 1cm higher, and we won't come close to the joint limit at all. But we're not sure, and our job is to make sure this stuff works even in the worst case.

Argh. Did I say I've missed this?

So I call Rudi Reider, thisol's APXS PUL, and ask him about the quality difference between 3mm and 4mm. (Rudi, you'll remember, is the German fellow who wears both belts and suspenders.) "You do what you can do technically, because we can work with almost anything," he says. One problem down, one to go.

Art asks me to check into the joint-limit problem with an IDD expert, such as Eric Baumgartner or Bob Bonitz ("not that we don't trust you ..."). I don't see Eric, but to my surprise, Bob's in his office. I describe the problem to him. Bob complains good-naturedly that they cut off his charge number for dealing with this stuff, then thinks, mutters, swears, flips through the flight software source code, and says it will be OK at the 3mm standoff position. The limit in question isn't the joint's physical limit, it's a limit built into the software, so we don't have to worry about the motors stalling as they try to push against the hardstops, or anything like that. As long as we're not commanding the joint past the soft limit, and we're not, the sequence will be fine.

"What would you do?" I ask him.

"Probably just exactly what you did," he says. High praise!

I go back downstairs to finish up the sequence, and Bob shows up a few minutes later to let me know the picture is actually a little better than we'd thought -- the software allows for a little bit of slop, which translates into a little extra margin for this move.

"Thanks, Angry Bob!"

He corrects me: "Frustrated Bob."

In addition to our visit from Frustrated Bob, we're visited by Rob Manning. Rob doesn't work on MER any more, but he stops by once in a while to see how things are going.

"I heard you set the distance record for Spirit," he says, shaking my hand. (That was a while back, but what the hell.) "Congratulations!"

We get to talking about our recent backward drives, and Rob, a veteran of Pathfinder's rover team, points out something I never noticed. The MER rovers use the same rocker/bogie suspension as Sojourner, of course, but -- the part I never noticed -- the suspension was front-to-back reversed on MER. No big deal, as it's constructed to work either way, but it's funny that I'd never heard this or observed it myself until now.

"So Sojourner was driving backward all the time," I say.

"Sojourner would say we're always driving backward," he says.

Rob and Art gab a while about MER and MSL. One thing they agree on is that Cassini will start stealing our thunder (what thunder we have left) when they get to Saturn. "We've gotta make it to those hills before SOI[1]," says Rob. "Well, we're running ahead of schedule," says Art.

"Where will we go after the hills?" Rob asks.

Art says another good target would be the wall of Gusev Crater itself. "We've got two years to get there."

"Two Earth years?"

"Yeah, about -- assuming the rovers can survive that long. If the rovers can survive this winter solstice, that's how long they've got until the next winter solstice on Mars." But he shakes his head. "I don't even want to think about the rovers lasting that long."

I have a sudden vision of myself as a doddering old man, still nursing Spirit around Mars, a few meters at a time .... Well, there are worse ways to spend your life.

Art also relays a bit of gossip. From what he says, NASA HQ does not want another solar-powered rover on Mars. Future rovers will use RTGs[2], and HQ is willing to fight the public battle that's sure to ensue.

That's a good thing, as it will maximize the value of the rovers. We could have gotten a lot more science done with the power improvements (and mass savings) RTGs provide. But, like so many decisions, that one might change with the presidential election. "We'll know in five months whether we have to redesign the things," Art says.

Art still intends to stay on MER rather than jump ship to MSL. When he says so, Rob says, "Well, that's a shame."

"You're saying you want me to work on the project?"

"I think you'd bring a lot to MSL."

Art feigns wounded shock. "I thought you loved me for who I am!"

Rob's lips quirk as he answers tactfully. "I ... appreciate you for who you are."




[1] Saturn Orbit Insertion.

[2] Radioisotope Thermoelectric Generators. You hook a chunk of a radioactive rock, such as plutonium, to a device called a thermocouple, which converts the heat from the rock's radioactive decay into electricity that powers your spacecraft. They're about a million times less dangerous than your car, but some people freak out whenever they hear about a spacecraft making energy from a radioactive rock, because they see it on the launch pad and get visions of mushroom clouds wiping out half of Florida. (Yes, some people do think that would be a bad thing.)

My own personal view is that the best way to sell the use of RTGs to these people is to stress that we're getting that scary old plutonium safely off of this planet and putting it where it can't hurt them any more. Maybe that tack wouldn't work in the real world as well as it does in my imagination, though.

1 comment:

Anonymous said...

The main danger with RTGs is with a launch vehicle accident--vaporized plutonium dioxide can be inhaled, after which it gets into bone marrow and then does nasty things. It's damn near harmless otherwise--Pu-238 is primarily an alpha emitter, which skin will stop, and it's too heavy to contaminate groundwater with. Good thing our launch sites are decently far away from concentrations of people.

Other than in launch accidents our RTGs are pretty awesome, though. A SNAP-27 survived a cislunar re-entry when Apollo 13's LEM came down. They put it into a deepwater trench in the Pacific and it's still sitting there, without having released its fuel.

I still haven't figured out why they switched back to PbTe thermocouples on the MMRTG though. I've yet to see any metric make those look better than the GPHS RTGs.