[meteorite-list] Dawn Journal - January 29, 2015

Ron Baalke baalke at zagami.jpl.nasa.gov
Mon Feb 2 13:59:54 EST 2015



http://dawnblog.jpl.nasa.gov/2015/01/29/dawn-journal-january-29/

Dawn Journal 
by Marc Rayman
January 29, 2015
 
Dear Abundawnt Readers,

The dwarf planet Ceres is a giant mystery. Drawn on by the irresistible 
lure of exploring this exotic, alien world, Dawn is closing in on it. 
The probe is much closer to Ceres than the moon is to Earth.

And now it is even closer...

And now it is closer still!

What has been glimpsed as little more than a faint smudge of light amidst 
the stars for more than two centuries is finally coming into focus. The 
first dwarf planet discovered (129 years before Pluto), the largest body 
between the sun and Pluto that a spacecraft has not yet visited, is starting 
to reveal its secrets. Dawn is seeing sights never before beheld, and 
all of humankind is along for the extraordinary experience.

We have had a preview of Dawn's approach phase, and in November we looked 
at the acrobatics the spacecraft performs as it glides gracefully into 
orbit. Now the adventurer is executing those intricate plans, and it is 
flying beautifully, just the way a seasoned space traveler should.

Dawn's unique method of patiently, gradually reshaping its orbit around 
the sun with its ion propulsion system is nearly at its end. Just as two 
cars may drive together at high speed and thus travel at low speed relative 
to each other, Dawn is now close to matching Ceres' heliocentric orbital 
motion. Together, they are traveling around the sun at nearly 39,000 mph 
(almost 64,000 kilometers per hour), or 10.8 miles per second (17.4 kilometers 
per second). But the spaceship is closing in on the world ahead at the 
quite modest relative speed of about 250 mph (400 kilometers per hour), 
much less than is typical for interplanetary spaceflight.

[Image]
Dawn observed Ceres for an hour on Jan. 13, from a distance of 238,000 
miles (383,000 kilometers). A little more than half of the surface was 
revealed as Ceres rotated. This imaging session is known as OpNav 1. Credit: 
NASA/JPL-Caltech/UCLA/MPS/DLR/IDA/PSI

Dawn has begun its approach imaging campaign, and the pictures are wonderfully 
exciting. This month, we will take a more careful look at the plans for 
photographing Ceres. Eager readers may jump directly to the summary table, 
but others may want to emulate the spacecraft by taking a more leisurely 
approach to it, which may aid in understanding some details.

While our faithful Dawn is the star of this bold deep-space adventure 
(along with protoplanet Vesta and dwarf planet Ceres), the real talent 
is behind the scenes, as is often the case with celebrities. The success 
of the mission depends on the dedication and expertise of the members 
of the Dawn flight team, no farther from Earth than the eighth floor of 
JPL’s building 264 (although occasionally your correspondent goes on the 
roof to enjoy the sights of the evening sky). They are carefully guiding 
the distant spacecraft on its approach trajectory and ensuring it accomplishes 
all of its tasks.

To keep Dawn on course to Ceres, navigators need a good fix on where the 
probe and its target are. Both are far, far from Earth, so the job is 
not easy. In addition to the extraordinarily sophisticated but standard 
methods of navigating a remote interplanetary spacecraft, using the radio 
signal to measure its distance and speed, Dawn's controllers use another 
technique now that it is in the vicinity of its destination.

>From the vantage point of Earth, astronomers can determine distant Ceres' 
location remarkably well, and Dawn's navigators achieve impressive accuracy 
in establishing the craft's position. But to enter orbit, still greater 
accuracy is required. Therefore, Dawn photographs Ceres against the background 
of known stars, and the pictures are analyzed to pin down the location 
of the ship relative to the celestial harbor it is approaching. To distinguish 
this method from the one by which Dawn is usually navigated, this supplementary 
technique is generally known as 'optical navigation." Unable to suppress 
their geekiness (or, at least, unmotivated to do so), Dawn team members 
refer to this as OpNav. There are seven dedicated OpNav imaging sessions 
during the four-month approach phase, along with two other imaging sessions. 
(There will also be two more OpNavs in the spiral descent from RC3 to 
survey orbit.)

The positions of the spacecraft and dwarf planet are already determined 
well enough with the conventional navigation methods that controllers 
know which particular stars are near Ceres from Dawn's perspective. It 
is the analysis of precisely where Ceres appears relative to those stars 
that will yield the necessary navigational refinement. Later, when Dawn 
is so close that the colossus occupies most of the camera's view, stars 
will no longer be visible in the pictures. Then the optical navigation 
will be based on determining the location of the spacecraft with respect 
to specific surface features that have been charted in previous images.

To execute an OpNav, Dawn suspends ion thrusting and turns to point its 
camera at Ceres. It usually spends one or two hours taking photos (and 
bonus measurements with its visible and infrared mapping spectrometer). 
Then it turns to point its main antenna to Earth and transmits its findings 
across the solar system to the Deep Space Network.

[Animation]
This animation of Ceres rotating was made by combining images taken by 
the Dawn spacecraft on Jan. 25 over the course of one hour in OpNav 2. 
Dawn was 147,000 miles (237,000 kilometers) from Ceres.
Credit: NASA/JPL-Caltech/UCLA/MPS/DLR/IDA

While it is turning once again to resume ion thrusting, navigators are 
already starting to extract information from the images to calculate where 
the probe is relative to its destination. Experts update the design of 
the trajectory the spacecraft must follow to reach its intended orbital 
position and fine-tune the corresponding ion thrust flight plan. At the 
next communications session, the revised instructions are radioed back 
across the solar system, and then the reliable robot carries them out. 
This process is repeated throughout the approach phase.

Dawn turned to observe Vesta during that approach phase more often than 
it does on approach to Ceres, and the reason is simple. It has lost two 
of its four reaction wheels, devices used to help turn or stabilize the 
craft in the zero-gravity, frictionless conditions of spaceflight. (In 
full disclosure, the units aren't actually lost. We know precisely where 
they are. But given that they stopped functioning, they might as well 
be elsewhere in the universe; they don't do Dawn any good.)

Dawn's sentient colleagues at JPL, along with excellent support from Orbital 
Sciences Corporation, have applied their remarkable creativity, tenacity 
and technical acumen to devise a strategy that allows all the original 
objectives of exploring Ceres to be met regardless of the condition of 
the wheels, even the (currently) healthy ones. Your correspondent refers 
to this as the "zero reaction wheel plan." One of the many methods that 
contributed to this surprising resilience was a substantial reduction 
in the number of turns during all remaining phases of the mission, thus 
conserving the precious hydrazine propellant used by the small jets of 
the reaction control system. Guided by their successful experience at 
Vesta, experts determined that they could accommodate fewer OpNavs during 
the approach to Ceres, thus saving turns. (We will return to the topic 
of hydrazine conservation below.)

The images serve several purposes besides navigation. Of course, they 
provide a tantalizing preview of the intriguing world observed from Earth 
since 1801. Each picture whets our appetite! What will Ceres look like 
as it comes into sharper focus? Will we see evidence of a subsurface ocean? 
What unexpected shapes and structures will we find? What strange new features 
will show up? Just what is that bright spot? Quite simply: we don't know. 
It would be a pretty good idea to send a spacecraft there to find out!

Scientists scrutinize all the photos for moons of Ceres, and OpNavs 3 
- 7 will include many extra images with exposures chosen to help reveal 
moons. In addition, hundreds more pictures will be taken of the space 
around Ceres in the hours before and after OpNav 3 to allow an even more 
thorough search.

On two occasions during the approach, Dawn will take images and spectra 
throughout a complete Ceres rotation of slightly over nine hours, or one 
Cerean day. During that time, Dawn's position will not change significantly, 
so it will be almost as if the spacecraft hovers in place as the dwarf 
planet pirouettes beneath its watchful eye, exhibiting most of the surface. 
These "rotation characterizations" (known by the stirring names RC1 and 
RC2) will provide the first global perspectives.

As Dawn flies into orbit, it arcs around Ceres. In November, we described 
the route into orbit in detail, and one of the figures there is reproduced 
here. Dawn will slip into Ceres' gravitational embrace on the night of 
March 5 (PST). But as the figure shows, its initial elliptical orbit will 
carry it to higher altitudes before it swoops back down. As a result, 
pictures of Ceres will grow for a while, then shrink and then grow again.

[Image]
Dawn's approach trajectory. We are looking down on the north pole of Ceres. 
The sun is off the figure far to the left. The spacecraft flies in from 
the left and then is captured on the way to the apex of its orbit. It 
gets closer to Ceres during the first part of its approach but then recedes 
for a while before coming in still closer at the end. Lighting by the 
sun is not depicted here, but when Dawn is on the right side of the figure, 
it only sees a crescent of Ceres, which is illuminated from the left. 
(The white circles are at one-day intervals.) Credit: NASA/JPL

Because of the changing direction to Ceres, Dawn does not always see a 
fully illuminated disk, just as the moon goes through its familiar phases 
as its position relative to the sun changes. The hemisphere of the moon 
facing the sun is bright and the other is dark. The half facing Earth 
may include part of the lit side and part of the dark side. Sometimes 
we see a full moon, sometimes gibbous, and sometimes a thin crescent.

The table shows what fraction of Ceres is illuminated from Dawn's perspective. 
Seeing a full moon would correspond to 100 percent illumination. A half 
moon would be 50 percent, and a new moon would be zero percent. In OpNav 
6, when Ceres is 18 percent illuminated, it will be a delicate crescent, 
like the moon about four days after it's new.

[Images]
Four views of Ceres as it rotates, as seen with Hubble Space Telescope, 
were the best we had before OpNav 2. All of Dawn’s pictures from now on 
will show finer detail. Credit: NASA, ESA, J. Parker (Southwest Research 
Institute), P. Thomas (Cornell University), and L. McFadden (University 
of Maryland, College Park)

OpNav images of a narrow crescent won't contain enough information to 
warrant the expenditure of hydrazine in all that turning. Moreover, the 
camera's precision optics and sensitive detector, designed for revealing 
the landscapes of Vesta and Ceres, cannot tolerate looking too close to 
the sun, even as far from the brilliant star as it is now. Therefore, 
no pictures will be taken in March and early April when Dawn is far on 
the opposite side of Ceres from the sun. By the end of April, the probe 
will have descended to its first observational orbit (RC3), where it will 
begin its intensive observations.

The closer Dawn is to Ceres, the larger the orb appears to its camera, 
and the table includes the (approximate) diameter the full disk would 
be, measured in the number of camera pixels. To display greater detail, 
each pixel must occupy a smaller portion of the surface. So the "resolution" 
of the picture indicates how sharp Dawn's view is.

We also describe the pictures in comparison to the best that have been 
obtained with Hubble Space Telescope. In Hubble's pictures, each pixel 
covered about 19 miles (30 kilometers). Now, after a journey of more than 
seven years through the solar system, Dawn is finally close enough to 
Ceres that its view surpasses that of the powerful telescope. By the time 
Dawn is in its lowest altitude orbit at the end of this year, its pictures 
will be well over 800 times better than Hubble's and more than 600 times 
better than the OpNav 2 pictures from Jan. 25. This is going to be a fantastic 
year of discovery!

[Table]

Some of the numbers may change slightly as Dawn’s trajectory is refined 
and even as estimates of the strength of Ceres’ gravitational tug improve. 
(Dawn is already feeling that pull, even though it is not yet in orbit.) 
Still, this should help you fill out your social calendar for the next 
few months.

To get views like those Dawn has, you can build your own spaceship and 
fly it deep into the heart of the main asteroid belt to this intriguing 
world of rock and ice. Or you can visit our Ceres image gallery to see 
pictures as soon as they are released. If you chose the first option, 
use your hydrazine wisely!

As we discussed above, to explore Ceres without the use of the reaction 
wheels that were essential to the original design, mission controllers 
have worked very hard to conserve hydrazine. Let's see how productive 
that effort has been. (You should be able to follow the story here without 
careful focus on the numbers. They are here for the more technically oriented 
readers, accountants and our old friends the Numerivores.)

Dawn launched in Sept. 2007 with 101 pounds (45.6 kilograms) of hydrazine. 
The ship escaped from Vesta in Sept. 2012, four weeks after the second 
reaction wheel failed during the climb out of Vesta's gravitational hole. 
(By the way, Dawn is now more than one thousand times farther from Vesta 
than it is from Ceres. It is even farther from Vesta than Earth is from 
the sun!) At the beginning of the long interplanetary flight to Ceres, 
it still had 71.2 pounds (32.3 kilograms) left. As it had expended less 
than one-third of the original supply through the end of the Vesta expedition, 
that might seem like plenty. But it was not. Without the reaction wheels, 
subsequent operations would consume much more hydrazine. Indeed, engineers 
determined that even if they still had the entire amount that had been 
onboard at launch, it would not be enough. The Ceres objectives were at 
serious risk!

The flight team undertook an aggressive campaign to conserve hydrazine. 
They conceived more than 50 new candidate techniques for reducing hydrazine 
consumption in the 30-month journey to Ceres and the 18 months of Ceres 
operations and systematically but quickly assessed every one of them.

The team initially calculated that the long interplanetary flight between 
the departure from Vesta and the beginning of the Ceres approach phase 
would consume 27.6 pounds (12.5 kilograms) of hydrazine even if there 
were no errors, no glitches, no problems and no changes in the plans. 
Following the intensive conservation work, they determined that the spacecraft 
might instead be able to complete all of its assignments for only 9.7 
pounds (4.4 kilograms), an astonishing 65 percent reduction. (Keep track 
of that mass through the end of the next paragraph.) That would translate 
directly into more hydrazine being available for the exploration of Ceres. 
They devised many new methods of conducting the mission at Ceres as well, 
estimating today that it will cost less than 42.5 pounds (19.3 kilograms) 
with the zero reaction wheel plan. (If the two remaining wheels operate 
when called upon in the lowest orbit, they will provide a bonus reduction 
in hydrazine use.)

Dawn's two years and four months of interplanetary cruise concluded on 
Dec. 26, 2014, when the approach phase began. Although the team had computed 
that they might squeeze the consumption down to as low as 9.7 pounds (4.4 
kilograms), it's one thing to predict it and it's another to achieve it. 
Changes to plans become necessary, and not every detail can be foreseen. 
As recounted in October, the trip was not entirely free of problems, as 
a burst of cosmic radiation interrupted the smooth operations. Now that 
the cruise phase is complete, we can measure how well it really went. 
Dawn used 9.7 pounds (4.4 kilograms), exactly as predicted in 2012. Isn't 
flying spacecraft through the forbidding depths of the interplanetary 
void amazing?

This success provides high confidence in our ability to accomplish all 
of the plans at Ceres (even if the remaining reaction wheels are not operable). 
Now that the explorer is so close, it is starting to reap the rewards 
of the daring 3.0-billion-mile (4.9-billion-kilometer) journey to an ancient 
world that has long awaited a terrestrial emissary. As Dawn continues 
its approach phase, our growing anticipation will be fueled by thrilling 
new pictures, each offering a new perspective on this relict from the 
dawn of the solar system. Very soon, patience, diligence and unwavering 
determination will be rewarded with new knowledge and new insight into 
the nature of the cosmos.

Dawn is 121,000 miles (195,000 kilometers) from Ceres, or half the average 
distance between Earth and the moon. It is also 3.63 AU (338 million miles, 
or 544 million kilometers) from Earth, or 1,390 times as far as the moon 
and 3.69 times as far as the sun today. Radio signals, traveling at the 
universal limit of the speed of light, take one hour to make the round 
trip.


More information about the Meteorite-list mailing list