March 18, 2001

Mission Update:

Thank you for visiting the Deep Space 1 mission status information site, the most popular log anywhere in the 4-dimensional space-time continuum for information on this daring mission through the solar system. This message was logged in at 6:30 pm Pacific Time on Sunday, March 18.

DS1 is now the proud owner of a gift from its controllers on distant Earth -- a new load of software. The new computer programs were transmitted across the solar system as the craft continues its trek through the cosmos.

The previous version of software was loaded in June 2000. In combination with new methods for flying the spacecraft, it led to the rejuvenation of the probe following the failure of its star tracker shortly after the end of the primary mission in 1999. That rescue proved fully successful, and the software has operated marvelously for the past nine months. So why did the operations team undertake the extraordinarily difficult task of replacing the software yet again? With a very small group responsible for this awesome mission, it certainly was neither for lack of excitement nor inadequacy of challenges!

Later this year, DS1 will attempt a risky and ambitious meeting with comet Borrelly. The spacecraft was not designed for this formidable task, and it will encounter daunting obstacles, some of which may well prove insurmountable. They will be described in future logs, but most of the changes in the software are designed to improve the chances that the spacecraft can acquire photographs and infrared measurements during its very brief and high speed meeting with the comet. The new capabilities contained in the software are not enough to guarantee that all goals of the flyby will be met, but they will give DS1 a chance when it greets this enigmatic visitor to the inner solar system.

Last month's mission log, still being analyzed and debated by several amiable space-faring species in the halo of the Milky Way galaxy, described the preparations for sending the software to the probe.

Starting on March 5, controllers began beaming more than 4 megabytes of software, divided into 267 files, to Deep Space 1. With a main antenna only about 30 centimeters (one foot) in diameter to receive the signals from more than twice the distance between Earth and the Sun, DS1 demonstrated admirable patience as it collected and stored the files in its memory. If you have ever waited for a large file to download through a slow modem, you know that it can tie up a connection for a long time. Because of the almost inconceivable gulf separating Earth from the spacecraft, even with the huge antennas and powerful transmitters at NASA's Deep Space Network stations in California and near Canberra, Australia, it took until March 8 to transmit all the files. Additional time was needed to send instructions to the craft to reassemble the files into the full computer code and to prepare for running the software. To begin using the software, the central computer needed to be rebooted.

Software had been replaced four times before during DS1's journey through space, three times during the primary mission to allow it to conduct more experiments with its advanced technologies and once during the extended mission to compensate for the loss of the star tracker. Nevertheless, because the smooth operation of the computer is so crucial to keeping the probe safe and healthy, the team approached the job of rebooting it with great diligence. It is quite unusual for a spacecraft to undergo a complete replacement of software, and DS1, having only a single central computer to rely on, could not afford a mistake. Whether the patient is undergoing emergency surgery, as DS1 did last year, or elective surgery, as it did this month, all participants need to be mindful of the serious risk posed by the operation! Throughout the process of loading the software, rebooting, and recovering from the reboot, JPL's Space Flight Operations Facility, from which DS1 is controlled, and other support areas were maintained under a "level 2" control to assure that all ground systems were maintained at maximum readiness.

On March 12 the DS1 team determined that the software had arrived on board with no errors and the spacecraft and ground systems were ready for the reboot. In addition, all the commands that were planned for use had been tested in the Deep Space 1 simulator at JPL. The next morning, some of the key engineers arrived at JPL well before sunrise (of course, the distant Sun never sets for lonely Deep Space 1) for one final review of the spacecraft condition. Following the confirmation of a clean bill of health and a final authorization by your loyal correspondent who was on the scene, the DS1 "ace" (the person who operates the computer that sends a command from JPL to the Deep Space Network for transmission to the spacecraft) was given the "go" to dispatch a command designated RESETX01 ("Romeo Echo Sierra Echo Tango X-ray Zero One" for the ultra-cautious, if not compulsive, controllers). RESETX01 activated a set of instructions on board that culminated in causing the computer to reboot.

Even though the reboot is intentional, the spacecraft treats it as if it were the result of a problem on board. In fact, when a serious problem is detected by the craft's own protective software, it sometimes uses a reboot, plus the subsequent automatic sequence of steps that that triggers, to correct the problem or, at least, to keep itself safe until human experts can intervene. This is known generically to those who fly spacecraft as a safing, and on DS1 this causes the spacecraft to configure itself for what is known as "Sun_standby_SSA level 1." The reboot forces the computer to reload its software (this time using the new version), leaving the probe without computer control for a few minutes. Among other actions, when the new software starts running it repoints the spacecraft to the only easily recognizable location from any isolated point in the solar system: the Sun. After the ace sent RESETX01, the bits forming that command raced across the solar system, DS1 dutifully executed all the steps of the reboot and associated safing, then began transmitting new signals indicating it was in the correct configuration, and its signals sped back to a waiting antenna in California, where they were detected and translated into data which were relayed to the DS1 control room staffed by a group of expectant engineers who quickly verified that the software was behaving as expected. Later, bolstered by the positive signals from DS1 and Chinese take-out food (including predictions, of course, of nothing but good fortunes!), the team instructed DS1 to save the software to a permanent (or "nonvolatile") memory location, thus overwriting the version that had served so well since the rescue in June 2000.

If not for the loss of the star tracker in November 1999, returning the spacecraft to normal operations after a safing would be relatively easy (to the extent that anything is easy when flying a low-cost and aging spacecraft hundreds of millions of kilometers from Earth). But now it is an extremely complex and difficult process, as the operations team has to help the spacecraft use its camera to lock on to a reference star. The principles underlying this remarkable trick to allow the spacecraft to orient itself in outer space without the equipment that was designed for this purpose were described extensively in mission logs from March through August 2000. (These logs can be seen at the archives page or now may be obtained in a special deluxe collector's edition of DS1 mission logs available in most elliptical galaxies.) And now that the craft's supply of hydrazine (the conventional rocket propellant it uses for orienting itself) is so low, the process is even more complicated.

In brief, with a method posted on January 16, 2000, the team helped DS1 to direct its main antenna to Earth, using the craft's radio signal as an indication of where the spacecraft was pointed. This aims the antenna close enough to our planet to allow reliable communications. By knowing that the antenna is pointed approximately at Earth, we can calculate roughly where the camera is pointed. Engineers instruct the camera to take some long exposures and radio the resulting pictures to Earth. Then analyzing the locations and brightness of the few faint stars in those images allows a determination of exactly where the camera is pointed and thus the precise orientation of the entire spacecraft. The details on the exact pointing are sent back up to the craft along with directions to a preselected lock star. While this procedure is taking place, the craft relies on directives from Earth to maintain its orientation. It does have gyros (which use light traveling through loops of optical fiber to sense spacecraft rotations), but they are not highly stable devices, so they are only reliable for short times. In fact, their instabilities induce a tiny drift in the craft. This very gradual motion is more than 10 times slower than the hour hand on a clock, but in this exacting work, that's too fast to be ignored. Nevertheless, by the afternoon of Friday, March 16, a very tired team had managed to get DS1 to lock to a star in Taurus, thus restoring the stable knowledge and control the craft needs for normal operations. If the tasks had not been quite so time consuming and, to be honest, exhausting, your ever-faithful correspondent might have devoted more time this weekend to providing better explanations of this complex undertaking or more of the details, including the propagation of the quaternion, the consequences of an amplifier failure at DSS-14 and a power failure at the DSN in Canberra, the decision to update the uplink subcarrier frequency to compensate for the narrow subcarrier tracking loop in the spacecraft's transponder, and why we didn't order Middle Eastern food on reboot day as planned.

After an arduous week, the spacecraft is taking the weekend off, calmly thrusting at impulse power. This coming week, Deep Space 1 will throttle up to full power. Over the coming months, as it continues its ion powered flight through space, on occasion it will conduct special tests of its new software functions.

DS1 is now about 225 million kilometers, or 140 million miles, from comet Borrelly.

Deep Space 1 is 2.1 times as far from Earth as the Sun is and more than 820 times as far as the moon. At this distance of 315 million kilometers, or 196 million miles, radio signals, traveling at the universal limit of the speed of light, take 35 minutes to make the round trip.

Thanks again for logging in!