Dawn Journal: We Crave Power!!!

Dear Dawnminants,

Dawn continues its flight through the solar system with all systems functioning well. It is vitally important that the spacecraft is reliably staying on course and on schedule, gently and steadily thrusting with the bluish glow of its ion propulsion system; yet that doesn't lend itself to the sorts of spine-tingling, heart-pounding, hair-raising, planet-shattering logs for which Dawn is famous (at least among immigrants from brown dwarf systems reading these reports in the vicinities of active galactic nuclei). So let's turn out attention to consider a particular aspect of flying a mission with ion propulsion.

We crave power!!

Perhaps that requires a bit more detailed consideration...

Engineers are developing a method to determine how much power the solar arrays can produce. It might seem odd that with the spacecraft having been in interplanetary flight for 10 months, engineers don't already know the answer. (Other facts might seem odd as well, such as the phrase "nihil ad rem" being in this sentence. This log will address only one oddity however.)

When the spacecraft was at Earth's distance from the Sun, shortly after launch, the solar arrays would have been able to supply more than 10 kilowatts, enough to operate about 10 average homes in the US (and nearly as much as your correspondent's cat Regulus generates when Mr. Vacuum Cleaner emerges from his closet). Dawn cannot use that much electrical power, but as it pushes deeper into space, the weaker illumination by the Sun will yield less power. The craft's two solar array wings, each about 2.3 by 8.3 meters (more than 7 by 27 feet), were designed to be large enough to meet the needs of the power-hungry ion propulsion system plus all other spacecraft systems even in orbit about dwarf planet Ceres. To thrust at nearly twice Mars' average distance from the Sun, Dawn carries the most powerful solar arrays ever used on an interplanetary mission.

The only way to measure the power of the arrays is for the spacecraft actually to pull the power from them, and its ability to do that is limited. When thrusting at full throttle and using all systems normally, Dawn consumes 3.2 kilowatts. Even now, traveling farther from the Sun than Mars ever ventures, the solar arrays can provide about 4 kilowatts. If the spacecraft activated all of its nonessential components, it still could not draw this much power. That leaves engineers without an accurate determination of the full potential of the arrays.

Of course, engineers thoroughly tested the electrical power system before launch, including each of the 11,480 solar cells and all other components, and from that they constructed a mathematical simulation of the arrays. But laboratory measurements do not perfectly reproduce conditions in space, so the computational model has some uncertainty. In-flight measurements are needed to improve their simulation of how much power the solar arrays can furnish at different distances from the Sun.

Who cares how much power is available? Well, first and foremost, our readers do! After all, you've gotten this far (and even farther right now) in this log, so you must have some reason for spending otherwise good time reading about the solar arrays. The Dawn project appreciates your interest, and we want to provide the information you apparently seek, even though we have no idea why you suddenly are eager to understand the solar array performance.

As it turns out though, there is another reason for establishing the true capability of the solar arrays. As explained in many (but fewer than 10,001) previous logs, Dawn's unique mission is possible only through the persistent use of its ion propulsion system. Rather than thrusting for minutes, as most spacecraft do, Dawn will thrust for years. As power diminishes in the dim depths of space, Dawn must throttle its ion thruster to lower power (and lower thrust) levels.

Because the throttle level depends on how much power is available, to formulate the details of the craft's trajectory and other plans for the mission, engineers require knowledge of how much power the arrays will provide at any distance from the Sun. After all, it is misleading to think of ion thrusting as an ion propulsion subsystem function; rather, it is a spacecraft system function, requiring most subsystems to operate together. Apart from the inevitable (and quite unpredictable) glitches and anomalies on the spacecraft and appearances of cake in mission control, and contrary to many people's preconceived notions, since well before launch the greatest technical uncertainty in the planning of Dawn's flight has been what the solar array power will be. So far, mission engineers have incorporated a reasonable, but conservative, estimate into the solar array simulation, but to refine the plans, they need to verify or correct the numbers.

Although the arrays produce more power than can be measured now, they would produce less power if they were not pointed directly at the Sun. That could reduce their output low enough to allow the spacecraft to draw as much power as the arrays could generate in that orientation, providing the calibration measurement that is needed. (Engineers would extrapolate to reveal how powerful the arrays would be when Sun-pointed at different distances.) As is usually the case in controlling interplanetary spacecraft, the details make such a test much less simple than it might appear at first blush.

With the normal switching of heaters on and off throughout the spacecraft, the total power consumption fluctuates, and that could add "noise" to the data, making the results harder to interpret and less accurate. If the spacecraft tried to draw more power than the solar arrays could produce, the battery would temporarily make up the difference but, depending upon the circumstances, protective software onboard would intervene to turn some systems off and place the spacecraft in safe mode. While that would not threaten the health of the spacecraft, it would threaten the solar array calibration. (By the way, the battery can store only enough energy to operate the spacecraft for about an hour. The solar arrays keep it charged for its occasional use.)

The solar array calibration working group (a runner up in the highly competitive Least Cool Dawn Team Name Contest) devised a method to calibrate the solar arrays that accounted for all these and many other considerations, including the solar panel thermal equilibration time and the dependence on temperature of the power vs. voltage curve, high voltage down converter phase margin, the solar array voltage set point, power processor unit undervoltage trips, the voltage-temperature control loop for the battery on the low voltage bus, and spacecraft safety even in the event of an unrelated anomaly during the test.

While conceptually simple (rotate the solar arrays by a certain angle and measure how much power the spacecraft can draw), the calibration proved complex enough that a somewhat simplified test was deemed appropriate. The objective was to verify how the spacecraft would operate in the test conditions before committing to the full calibration. The plan was to execute the test on July 21, and if everything went perfectly, the final version would be attempted the next day. Last year, when the planning for this began, it was decided to schedule a backup opportunity late in 2008 in case the first time did not yield the desired data. (In addition, the calibration will be repeated occasionally over the course of the mission to monitor changes in the solar array characteristics, ensuring the power predictions remain accurate.)

Because electrical power is essential to the operation of all subsystems, a test of this nature calls for all subsystem personnel to scrutinize spacecraft telemetry for symptoms of unpredicted and infelicitous behavior. All commands were contained in a single file transmitted to the spacecraft, and immediate intervention would not be physically possible, as radio signals revealing the condition of the spacecraft would take nearly 18 minutes to reach Earth, and commands sent in response would require the same time to travel back to the spacecraft. Nevertheless, the team needed to be prepared to take action in the very unlikely case a problem developed, so two key measures were put in place: all stations in mission control were at the ready, and pizza was provided to help fill the gaps in this early-evening test while radio signals raced across the solar system.

The result: overall the test went well, although there was unexpected spacecraft behavior and unexpected toppings on the pizza. For the former, no response was required by the flight team, as the spacecraft executed all the commands correctly and returned to its normal configuration at the end. The test yielded only a partial set of calibration data however, apparently because some of the reconfigurations of the electrical power system and the ion propulsion system for the purposes of the test led to a few responses that were not anticipated. The spacecraft transmitted a large volume of supporting data, which will take longer to digest than the pizza, and when the satiated engineers have finished, they will determine what modifications to make for a new test. A future log will describe the next test and any corresponding changes in the food delivered to mission control.

Turning their attention on July 22 to a different topic, the team modified software in one of the many computers onboard. In January, with neither permission nor warning, a subatomic particle traveling through the solar system hit a sensitive electronic component on the spacecraft, triggering a quick sequence of events that culminated in the spacecraft entering its safe mode. Since then, programmers have developed a way to prevent space radiation that reaches that particular circuit from having the same effect. With the updated software, now the only consequence would be a notice to controllers that the device was hit, and the spacecraft would not need to enter safe mode or interrupt its activities.

The solar array test and the software change were conducted during a planned 2-day pause in thrusting. On schedule on July 23, Dawn resumed propelling itself with xenon ions. Once again the special lights adorning a wall in mission control were turned on, emitting a blue glow to remind everyone who visits or works there of the probe's patient pursuit of intriguing and unexplored worlds in the asteroid belt.

As Dawn travels through space, Earth and the Sun grow more remote. Although the journey will never bring it near the part of the solar system it used to consider home, we will see in the next log that its path to Vesta and then to Ceres is not as direct as some might expect. As part of the explanation, the log also may reveal something about this misspelling.

Dawn is 324 million kilometers (202 million miles) from Earth, or more than 885 times as far as the moon and 2.14 times as far as the Sun. Radio signals, traveling at the universal limit of the speed of light, take 36 minutes to make the round trip.

Dr. Marc D. Rayman
8:30 pm PDT July 27, 2008