Engineering and science information from NASA's Deep Space 1 was transmitted from the spacecraft to ground controllers today, and the mission team continues to analyze the data.

JPL Press Release, July 30, 1999

The spacecraft's infrared camera has returned four sets of valuable data which will help scientists determine the asteroid's composition. Called spectra, a form of data which break light into component colors much like a prism does (usually displayed as graphs), they cover different parts of the asteroid and were taken both before and after closest approach.

In addition to two black-and-white images taken about 70 minutes prior to closest approach, two black-and-white images of limited quality, taken 20 seconds apart at approximately 15 minutes after the encounter, have been returned.

Diagnosis of an apparent target tracking problem continues. If the asteroid were much dimmer than expected or if the camera was much less sensitive than expected, this would explain why the autonomous navigation system did not lock on to allow the camera to point at the asteroid during closest encounter. This possibility is being analyzed.

The spacecraft's xenon ion engine was fired at 9 a.m. Pacific time and will continue thrusting almost continuously for the next three months in preparation for flybys of two comets during a possible extended mission after the September 18, 1999, conclusion of Deep Space 1's primary mission.

DS1 MISSION LOG: July 18, 1999 (from Marc Rayman, JPL)

With its mission of testing high risk, important new technologies having exceeded expectations, Deep Space 1 is now making final preparations for a very challenging encounter with an asteroid with the simple yet somehow noble name 1992 KD. This asteroid was discovered 7 years ago by JPL astronomers Eleanor Helin and Ken Lawrence as part of an extensive and productive effort to locate and study asteroids. 1992 KD is so small that it can only barely be seen through large telescopes, so very little is known about it. It is believed to be about 1 or 2 km (perhaps a mile or so) in diameter. It apparently rotates very slowly, taking nearly 10 Earth days to complete one turn. Although DS1 was designed to test technologies, it will attempt to make scientific measurements when it flies past 1992 KD. Black and white pictures may reveal the asteroid's size and shape and show craters, hills, valleys, and other topography. Infrared measurements may help scientists determine the minerals that make up the surface. By searching for changes in the solar wind, the stream of charged particles flowing from the Sun, in the vicinity of the asteroid, it may be possible to determine if it has a magnetic field. Perhaps the solar wind or sunlight even cause surface material to be slowly eroded from the asteroid and flung into space, in which case the spacecraft may directly measure the resulting free atoms.

To get close enough to make all these measurements, AutoNav will attempt to bring DS1 closer to 1992 KD than any spacecraft has ever come to a solar system body without actually landing on it. Speeding by at 15.5 km/s, or nearly 35,000 miles/hour, the spacecraft will pass by more than 50 times faster than a commercial jet and more than twice as fast as the space shuttle. But it will come a mere 15 kilometers from the center of the asteroid, or less than 9 miles from the surface. This is a great challenge to AutoNav and to the operations team, but if it works it should be exciting indeed.

The small operations team has been developing the complex set of instructions that will govern the spacecraft, including the ones that give AutoNav the opportunity to design and execute maneuvers to correct the spacecraft's trajectory, the commands to the new technology science instruments to collect data, directions to the attitude control system on how to turn the spacecraft as it nears the asteroid, and instructions on how to transfer, manipulate, and store the large volume of data to be collected. Refining this unusually complex choreography, in which all the spacecraft systems including AutoNav need to work together, is the focus of the team's work right now. A group of instructions is known as a sequence, and each day, the sequences covering the final 6 hours before the closest approach to 1992 KD are run through the Deep Space 1 test facility at JPL. This is a simulation of the spacecraft, created using some hardware similar to what is on the real spacecraft and some computer programs that emulate the behavior of other parts of the spacecraft. This allows the team to test, modify, and retest sequences, a cycle repeated now nearly every day. The test facility is certainly not identical to the spacecraft, so a successful test does not guarantee success on the spacecraft, but it does allow many of the bugs to be worked out.

A rehearsal of the encounter using the actual spacecraft was conducted on July 13. DS1 flew by an imaginary asteroid affectionately named Spoof2. For AutoNav to find this asteroid, special computer files were sent to the spacecraft that contained the orbit of Spoof2. Whenever AutoNav took a picture for navigation information, a special computer program intercepted the picture, added Spoof2 to it, and then sent it on its way for AutoNav to analyze. Under AutoNav's guidance then, DS1 executed two course corrections with its small thrusters and did correctly fly past the elusive Spoof2. The complex test went well. After analyzing data on how the spacecraft handled the sequences, the operations team is making a number of changes, and revised sequences continue to be tested in the test facility. When DS1 takes the final test on the evening of July 28, it will be on its own. Its closest approach to the asteroid will occur at approximately 9:46 pm PDT, and it will be several hours after that before it can begin reporting its results to Earth. Returning all the data will then require several days.

In the meantime, AutoNav continues to collect images of distant asteroids and stars to refine its estimate of where it is in the solar system using a method described in many earlier logs. But it is still too far away from 1992 KD to see that tiny asteroid, so for now it is simply using the best estimate of where the asteroid is. AutoNav will not be able to detect 1992 KD until about one day before it arrives. AutoNav has done a remarkable job calculating its position in the solar system, now routinely getting within 900 km or under 600 miles of the position as determined by conventional radio tracking. Compared to its distance from the Sun this is impressive indeed. This accuracy is comparable to being anywhere in the continental United States and determining your position to within about 70 feet. When DS1 is traveling through most of the expansive emptiness of the solar system, this knowledge is more than adequate. But when it gets in the vicinity of 1992 KD, it needs to do even better. The principal limitation now in AutoNav's ability is the result of imperfections in the camera. Although AutoNav has sophisticated techniques to analyze the camera's pictures, it cannot fully compensate for shortcomings in the camera. AutoNav's designers and testers are paying close attention now to see how well it can do.

Deep Space 1 is still nearly 14 million kilometers, or over 8.6 million miles, from 1992 KD. The spacecraft is now 20% farther away from Earth than the Sun is and almost 470 times as far as the moon. At this distance of over 179 million kilometers, or more than 111 million miles, radio signals, traveling at the universal limit of the speed of light, take almost 20 minutes to make the round trip.

DS1 MISSION LOG: August 1, 1999

Deep Space 1 successfully accomplished an extraordinarily challenging encounter with asteroid Braille late Wednesday evening. As the hundreds of trillions of faithful listeners to these recordings know, reaching this tiny asteroid was a bonus. DS1 had already been remarkably successful in its primary mission of testing 12 advanced technologies that will allow many future missions to explore the solar system. The visit to Braille was icing on the cake.

Prior to this encounter, very little was known about asteroid Braille. In fact, it is so small and distant that astronomers had had great difficulty even in pinning down its location. So beginning on Sunday July 25, after flying for 9 months toward the asteroid, and less than 4 days before getting there, AutoNav began taking pictures to try to find it. As expected, the little asteroid did not show up in the first pictures. On Monday, a faint image began to appear, so vague that AutoNav was not able to detect it, but more sophisticated processing of the images on Earth revealed the asteroid. It turned out to be over 400 kilometers, or more than 250 miles, from where it was expected it to be. Because time was getting short and the small asteroid still did not show up in AutoNav's analysis of the images, the DS1 navigation team estimated from the pictures where the asteroid was. Then they used the same computer programs that AutoNav has on board to design a course correction that was radioed to DS1. Then with less than a day and a half to go, using that information just as if it had designed the course correction itself, AutoNav turned the spacecraft and ordered the thrusters to be fired. The activity went well and put the spacecraft on course for the new estimated location of Braille.

As DS1 continued to speed toward Braille, it continued to take navigation pictures. But the dim asteroid refused to show up in AutoNav's analysis of the images. Finally, early Wednesday morning, only 17 hours before arriving, AutoNav was able to lock on to Braille. But before it could design its own course correction, a problem arose on the spacecraft. A small bug in the extremely complex software manifested itself. AutoNav has a record of its locations at different times, and it uses this whenever it computes its orbit. As it updates the information, it drops the oldest information. Because it has been trying to zero in on Braille and has been making course corrections more frequently than usual, it recently accumulated a larger set of information than normal. In essence, it tried to process too much data on Wednesday morning. Anyone who has used a computer knows that sometimes the programs run into trouble, and the computer needs to be restarted to clear it. Well a more sophisticated response occurred when DS1's computer detected this problem and allowed protective software to stop all spacecraft activities, reboot the computer, turn off nonessential devices, and place DS1 in a predefined safe configuration known to the operations team as Sun standby SSA. The spacecraft turned to point at the Sun (the only easily identifiable landmark from anywhere in the solar system), used a back-up antenna, and awaited help from Earth. A shocked operations team discovered around 5:30 Wednesday morning that the spacecraft had entered standby, and the encounter with Braille was going to occur Wednesday evening.

Working astonishingly quickly, the team sent commands to the spacecraft to return it to its normal operational state. The spacecraft was instructed to turn to point its main antenna at Earth. The combination camera and spectrometer was turned back on. The advanced ion and electron spectrometer, which measures charged particles in space, has high voltage power supplies that use up to 8000 volts. It has to be turned on slowly and delicately, but it was brought back to its correct settings. The suite of sensors designed to monitor the electric and magnetic fields of the ion propulsion system had been reprogrammed to try to detect the asteroid. But when they were turned off, they reverted to their old program, so the team reloaded the newer software.

But what about further course corrections? When the computer rebooted, it deleted the results of AutoNav's detection of the asteroid. So now with only a few hours to go, it was not on an accurate enough course to reach the asteroid, and while the operations team was restoring DS1 to its normal configuration, AutoNav could not design a new course correction. But from 116 million miles away, the ever-resourceful team was able to recover from the spacecraft's memory 3 of the pictures AutoNav had taken and analyzed before the problem arose. They were transmitted to Earth, analyzed, and used to compute the final course correction. The information was transmitted back to the spacecraft in time for AutoNav to take over. Working against the clock the entire time, the operations team sent the last command to the spacecraft with only 4 minutes to spare before it had to turn its antenna away from Earth to execute the course correction. In fact, while all this was going on, DS1 continued rushing toward Braille at a speed of over 9.6 miles every second. From the time the spacecraft entered standby until the last command was sent by the operations team, the spacecraft moved well over half a million kilometers or more than 300,000 miles closer to Braille. That's more than the distance between Earth and the moon.

Once DS1 turned, the operations team could only watch anxiously. As planned, the spacecraft returned only very limited data during the flyby, as pointing its camera at the asteroid meant that it could not point its antenna at Earth. After such a complex and challenging day, everyone wondered whether DS1 was really ready to encounter the asteroid.

AutoNav carried out its final course correction flawlessly just 6 hours before reaching Braille. Following that it resumed taking distant images to assure that it could point the camera as it got in still closer. It tried and succeeded four times, completing this process 1 hour and 10 minutes before its closest approach to the asteroid. Still at that time, it was 65,000 kilometers or over 40,000 miles away. Then 28 minutes before encounter, AutoNav switched to a different mode and used a different part of the camera that was designed for use closer to celestial bodies. But the asteroid did not show up as expected, so AutoNav received no new information to analyze. The mystery of why Braille could not be detected is still being studied. But it is likely because the asteroid was much darker than astronomers had predicted or because of an unknown behavior of this part of the camera with a dim object. In any case, when DS1 got close to the asteroid, although AutoNav knew approximately where the asteroid was, its information was not sufficient to point the camera accurately enough to capture close-in images. It was adequate however to acquire infrared data, and DS1 also collected all the data desired with the ion and electron spectrometer and with the other sensors. Images at the moment of closest approach were never planned, so these other data could be collected. After closest approach to the asteroid, the spacecraft turned over to look back, as AutoNav switched back to its normal mode. It correctly pointed the camera and captured images about 15 minutes after its closest approach to Braille.

This was a remarkable day for the team that has been overseeing a little spacecraft that has performed such a big job. DS1 accomplished the closest flyby ever of an asteroid, obtained important new information on the performance of AutoNav and the camera, acquired all the data desired from the instruments to measure charged particles and magnetic and electric fields, and collected infrared and black and white images of the asteroid. Of course it would have been fun to have a detailed picture of asteroid Braille, but all in all it was a spectacular finale to an incredible mission!

It took the spacecraft about a day to transmit all of its the data now, and the results will be announced this week. This information log will be updated next weekend with the findings.

DS1 MISSION LOG: August 8, 1999

After its remarkably successful mission of testing what were high-risk advanced technologies critical for future space science missions, Deep Space 1 blazed a new trail with its extraordinary encounter with asteroid Braille on July 28. DS1 managed to swoop to within about 26 kilometers, or just 16 miles, of the surface. This is far closer than any spacecraft has ever come to an asteroid, and it's only about twice as high above the asteroid as a commercial jet flies above the Earth. Yet it was going at 15.5 kilometers/second or 35,000 miles/hr, well over 50 times as fast as a commercial jet or over twice as fast as the space shuttle. Getting to this close encounter with this asteroid was like kicking a soccer ball on Earth and scoring a goal on the moon.

DS1 collected all the data that was wanted to test the autonomous navigation system, to measure ions and electrons (or charged particles) in the vicinity of the asteroid, and to search for magnetic and electric fields near the asteroid. The spacecraft also acquired pictures and infrared spectra, except it did not get close-in images and it got only some of the spectra. To have gotten the kind of images DS1 did capture would have required a telescope 200 times more powerful than the Hubble Space Telescope if they had been taken from Earth orbit.

Last week's log told the chilling tale of the spacecraft's entry into standby followed by the dramatic and truly heroic recovery by the operations team. Following that, AutoNav tracked the asteroid down to 70 minutes before closest approach. At 28 minutes before the closest approach, it correctly switched to a different operating mode and as part of that used a different portion of the camera for its navigational sightings. The asteroid however did not register in the camera, so AutoNav had no new information with which to update its estimate of the actual location of the asteroid. All it had was its last estimate made at 70 minutes, or 65,000 kilometers (40,000 miles), away. And that was not accurate enough to keep the asteroid in the camera's view down to the time that the close-in images were to be collected.

Essentially, this is similar to the situation you would face if you were trying to drive down a dark country road and you had merely a glimpse of the surroundings at the beginning of your drive. You couldn't expect to get to your destination just on that basis. AutoNav however had accurately estimated the actual time that it would get to the asteroid, and so it ordered the various sensors to collect data at the correct times. Finally AutoNav switched back to its normal mode and just after the spacecraft zipped by the asteroid, it slowly turned to look back at it. The camera pointing was dead on and it got pictures and infrared spectra 15 minutes later.

Apparently the strangely shaped Braille, illuminated from the side by the Sun, caused it to be surprisingly dim. And the camera simply could not register that faint light as DS1 approached the asteroid. But on the other side, as DS1 receded, the asteroid did show up in the camera.

The pictures reveal that Braille is a very irregular, elongated object, only about 2.2 kilometers in one axis and 1 kilometer in another, or about 1.4 miles by 0.6 miles. This is smaller than many mountains in the United States. The real scientific prize however is in the infrared spectra. The spectrum of Braille tells a fascinating story. It is nearly identical to the spectrum of Vesta, one of the largest asteroids. That means their surfaces are made of the same material. And scientists already knew by comparing Vesta's spectrum with that of many minerals in laboratories that it is made of basalt. Now basalt is a rock that is formed when lava cools, and one question scientists have grappled with is: How did Vesta ever get hot enough to form lava? Vesta is too small to have the inventory of radioactive materials that a large planet like Earth has, in which the decay of those elements produces enough heat to keep the interior hot. There are several possible explanations, but one is that collisions with other asteroids caused enough heating to make the lava. The Hubble Space Telescope has revealed an enormous crater on Vesta that suggests a tremendous impact has occurred there. An exciting possibility is that such a collision sprayed many fragments into the solar system, and Braille is one of them.

Now the spectra of Vesta and Braille also match those of some meteorites. But it is not known how chips from Vesta, liberated in collisions with other asteroids, could reach Earth to fall as meteorites, as Vesta is in the main asteroid belt, between Mars and Jupiter. But now we have Braille as an example of an asteroid closer to Earth yet resembling Vesta, giving astronomers new clues to the trail followed by fragments of Vesta as they make their way to Earth. As these new data from Braille are analyzed more, they should add to the intriguing puzzle of how our solar system has evolved. In addition, as asteroids like Braille threaten Earth with catastrophic impacts in the future, a better understanding of their composition and structure will aid in determining how to protect our planet.