Martin Tomasko
People have been sending instruments to space for a long time now, and if you’re in the high-vacuum space environment, people have learned how to make instruments work in that environment. They’ve got thirty years of experience.
If you’re going to a new planet that nobody’s been to before, nobody’s real sure exactly what the density and the temperature and the pressure are. They have models, they know within some range of families what they are, but you have to design the heat shield and the parachutes and all the explosive bolts, the whole sequence, so that the instrument really works in this new environment. You’re beginning to get the sense that this is the first time anybody’s ever been in this environment. There’s a lot of hand-wringing going on.
You design the heat shield as well as you can, and then you hear somebody’s got a model where the density’s a little different, and, geez, you never tested the heat shield under those conditions: I wonder if it will work. Somebody gets a blow torch in the lab and they blast on the heat shield for a while, and, “Yeah, it sort of works at that temperature, but it’s going to be a little bit tight.”
And the parachutes, well, they have to be deployed now in a supersonic regime, and the cables have to be strong enough so the parachutes don’t snap off, and the parachutes have to open. “Yeah, we think that will work, but actually the parachutes are a little bit arguable. We’re not really sure.” And the probe has to rotate, and there are constraints on the rotation rate and the stability of the platform. The Europeans always sign up for the requirements, whatever the requirements are, and you’ve got a little piece of paper saying they’re going to meet those requirements. But, you know, that’s really up to them to meet.
That game was really quite interesting, because you’ve got lots of people from NASA pounding on you and saying, “This is the amount of money you’re going to get. You’re not going to get any more. So make sure you don’t spend anymore, don’t overrun the budget, we’re not going to give you any more.”
Then you’ve got the Europeans saying, ‘The launch date is this date and you’ve got to deliver by this date. If you don’t, we’re going to fly without you. In fact, the first thing you have to deliver is a lead weight with the same bolt hole patterns as your instrument, and if things go wrong and you don’t deliver your instrument in time we’re going to fly that weight. It will have the same center of gravity and the same total mass, and that’s what’s going to fly.”
But nobody is actually pressing on the Principal Investigator to say, “Make sure your instrument actually works and makes some useful measurements and that you’re actually going to learn something about this object.” All the pressure is: Don’t exceed the budget and don’t blow the schedule, but you’re the only guy there trying to stand up and say, “Yeah, but I want my instrument to work."
The Huygens probe gets its data back by transmitting the data to the orbiting spacecraft as it flies by. The orbiter records the data, and it turns around later after the mission is over and points its big antennae at the Earth to blast the data down to Earth in no time at all.
The Europeans had this idea that it would be a really good thing if they could do a test on the radio communication between the Huygens probe and the orbiter. They said, “Suppose we pretend the Earth is the Huygens probe, and we transmit signals from the Earth with just the right Doppler shift and the right Doppler frequencies and the right way to the orbiter, and see if the orbiter receiver picks them up?”
So they did the test. They very carefully transmitted the signals just like what Huygens would transmit, and the orbiter didn’t receive a thing. They did the test over and they did it over and they did it over again, and after six months they said, “You know, it just doesn’t work."
There’s this thing known as the Doppler effect. The probe is going to be in Titan’s atmosphere, in parachute, hardly moving at all, coming down at six meters per second, while the Cassini orbiter is whistling by like a son-of-a-gun at 20 thousand kilometers per second. There’s a tremendous Doppler shift, and the radio frequency at which Huygens transmits is outside the band of the receiver with this Doppler shift going on. They’re not listening on the same frequency. This isn’t going to work.
So we said, “Well, okay, what are we going to do to fix it?
If it were all programmable hardware you could send a command signal to change the frequencies it would all be fine. But it’s all hard-wired. There’s no computer that changes the frequency. The frequency that has been programmed in, and wired in, is the frequency you’re going to get. So what are you going to do?
One possibility is that you change the geometry of the entry, so that instead of the orbiter whistling right by the edge of Titan, it goes by at a glancing angle, and so its component of velocity toward the probe is much smaller and the Doppler shift is much less. Of course that means that the whole orbiter tour won’t work anymore—which they just spend the last two years designing and arguing about it—because we’re going off in a different direction. We’re not in the right place at the right time going the right speed. So what are we going to do about that?
Actually, the trajectory designers at JPL are very, very good, and very, very clever. They decided that if they put in two more small orbits and delay the probe delivery to the third of those orbits, they can eventually—within a month—rejoin the original tour at the same place, going in the same direction and at the same speed, and save all of the rest of the four years of the tour.
We said, “It might actually work!” Our motto in this project was, “You know, it actually could still work!"
Even then, the frequency is almost right. But it’s just a little bit off. We could drop out data. So what should we do?
The guy who was in charge of the European Probe Design, it turns out he just hated batteries. He had bad experiences with batteries in other missions: “I don’t trust them. I don’t like them. They’re not going to work. They’re going to be stored for seven years during the cruise—I want a big margin of batteries.” So they flew six batteries. They could’ve probably done it with three or four. But they flew six. They had battery power up the wahzoo.
He said, “Suppose we heat this radio transmitter for four hours before we get to Titan. We turn on all the heaters full-blast. We’ve got enough battery power. We still will have one battery to spare. We’re going to devote one battery to just heater power, and it’s going to heat this probe up by 20 to 30 degrees."
Now, if you just change the temperature a little bit, the frequency doesn’t move. But if you heat it up 20 to 30 degrees Fahrenheit, the frequency actually shifts. It shifts in the right direction. The combination of the new trajectory with the smaller Doppler shift, and four hours of pre-heating the probe, is actually enough that we should get signal back. They verified this with a test and confirmed that this would work.
But it was only because the whole dang cruise was seven years long that they had enough time to do the test, discover this problem, and then to come up with a solution, and then to do the test verifying that the solution would actually work. Otherwise, it would’ve been dropped into Titan’s atmosphere and we wouldn’t have heard a peep out of it, and nobody would have a clue why. It was a really close, close call.
It was an adventure, and a real story of some triumph that both the Europeans and the Americans can take credit for: The Europeans for discovering the problem, understanding the radio, and proposing a solution, and the Americans for coming up with this very elegant tour that saves all of the planned four-year tour and just puts in these two quick loops and gets back on track. It was really beautiful.
Peter Smith
Titan was scary. I was lucky enough to be in the control room over in Germany, at Darmstadt, where they have the European Operations Center. Of course everybody knew exactly what time the signals were supposed to arrive. As that time came and went, the signals weren’t arriving. Poor Marty was pacing.
On the probe there are two antennae, A and B. They were sending redundant signals, and the idea is, if you lose one, you don’t lose the whole mission. After checking the command sequence, they finally realized that somebody forget to turn on the receiver to receive the signals from channel A. They’re realizing this a few minutes after the signals were supposed to come: “I don’t see the command to turn on the receivers.”
Everybody in the room wanted to know, “What about B?”
“Well, that’s turned on, but we’re not getting anything.”
I think six minutes went by, and then the signals started to come. In those six minutes, we died a thousand deaths. All the reporters were there, people from all over Europe and the United States, everyone waiting to see data coming down.
Finally, of course, data did come down on side B. We’re looking at the pictures, and the strange thing was, the surface images looked like the coast of Italy. There was a lake—little rivers coming down off a hill into what looked like a lake. That’s unbelievable. What a thrill. You don’t know what to expect from Titan but you don’t expect rivers flowing down hills into lakes.
Martin Tomasko
Well, with the pressure of going to reviews, and the pressure of a different team of guys actually doing everything a different way, somebody left out a command. The command to select the ultra-stable oscillator was included, but the command to actually turn the receiver on was omitted. So now we’ve got two receivers, but only one is working. The other receiver just isn’t turned on.
We’ve got all our redundant data, because it worked on that channel, but on the images, we’ve got 350 instead of 700. Even a little bit worse for us was the fact that the probe says, “Hey, both channels are working peachy. I’m just going to divvy them up between here and here and here and here.” That means we’re missing every other image. My images aren’t a consistent set the way I was expecting, but have holes in them—half the images are missing. That makes my life, trying to put those images together to make that nice mosaic, a little bit difficult.
In fact, it makes it especially difficult because, at the same time, the probe dynamics in this new atmosphere was not what was expected. The probe is bobbing and weaving much faster than what was expected, and at higher rates. The probe rotated in the intended direction for first ten minutes, and for the next two and a half hours it rotated the wrong way. It wiggled and shook and did all of these things.
Now you’ve got this set of 350 images that you’ve got to put together. It’s like a puzzle. You don’t have a picture on the front of the box; you don’t know what it’s supposed to look like. Half the pieces are missing. Every time the probe tips and turns, the footprint of the image on the ground changes, so the pieces change. The shape of the pieces change, half the pieces are missing, and the clock is going around backwards so you don’t know where you are. The housekeeping data to tell you where the image is pointed isn’t there.
So to prevent people from being too proud, the dynamics were really screwy, and somebody just left out the command to turn on the other receiver. But we got most of the data that we needed. The whole story is, the Lord gives and the Lord takes away. It’s only by the grace of God that you get what you want. We planned on getting about five times what we needed, and we got about 95% of what we needed. So we almost have the whole mission done exactly right.