next up previous contents
Next: Venus Up: Chandra News Previous: HELP DESK

Chandra and the Leonids

The following article was adapted by Dan Shropshire and Roger Brissenden from a pair of articles from the Chandra Chronicles:

http://chandra.harvard.edu/chronicle/0401/leonids.html

http://chandra.harvard.edu/chronicle/0401/leonids_part2.html

-Editor

While skywatchers eagerly anticipated the 2001 Leonid meteor shower, members of the Chandra X-ray Center (CXC) team were busy in their own right. However, instead of gazing into the night sky for ``shooting stars," they spent most of November 18th monitoring and maneuvering Chandra to protect NASA's premier X-ray facility from harm.

``The chances of Chandra getting hit by one of the meteors in this shower is really very low," said Roger Brissenden, manager of the CXC and of Chandra's Operation Control Center (OCC), which is responsible for operating and communicating with the satellite. ``Regardless, we will still take every precaution to make sure that Chandra is safe" (Fig. 19).


  
Figure 19: Roger Brissenden, Manager of the CXC
\begin{figure}\centering
\resizebox{\textwidth}{!}{\includegraphics{leonids_fig1.ps}}\end{figure}

LINK TO POSTSCRIPT FILE for Figure 19

The Leonid meteor shower - so-called because the meteors appear to come from the direction of the constellation Leo - is due to trails of debris left by Comet Tempel-Tuttle on its 33-year orbit of the Sun. When the Earth passes through these streams of rocks, dust grains, and gas every November, observers on Earth can be treated to a wonderful show.

Because of the relative difference in velocity and direction between the debris from Tempel-Tuttle and the Earth, the Leonids are among the fastest meteors around, entering the Earth's atmosphere at 44 miles per second. Every so often, the Earth passes through an especially dense clump of debris from Tempel-Tuttle, and a truly spectacular meteor storm occurs, such as the Leonid storm in 1966 that produced 150,000 meteors per hour. Just such an intense show was predicted by several models to occur this year.

Data on predicted meteor storm intensity were produced by Marshall Space Flight Center's Space Environments Team. Resident expert Bill Cooke ran several meteoroid density models using Chandra ephemeris data provided by the flight operations team. Each model produced a meteoroid flux vs time for a given mass distribution. These data are then analyzed by Chandra operations engineers and compared to a pre-defined threshold. If the threshold is violated then action is taken to protect the observatory. This action usually involves changing spacecraft attitude to minimize the exposed cross sectional area to the incoming meteoroid stream. This action also includes the powering down of sensitive electronic units. This year's predictions of meteoroid flux pushed the Chandra team to be extra cautious during the 2001 Leonid shower.


  
Figure 20: This schematic shows Chandra's orbit in relation to the Earth and the Sun. The light green portion indicates when the Chandra spacecraft will be in the predict ed path of the meteors.
\begin{figure}\centering
\resizebox{\textwidth}{!}{\includegraphics{leonids_fig2.ps}}\end{figure}

LINK TO POSTSCRIPT FILE for Figure 20

``We have done many things to safeguard Chandra against this year's Leonid meteor shower," said Dan Shropshire, who leads Chandra's Flight Operations Engineering Team. ``Chandra will be pointed in the exact opposite direction as the incoming meteors, for example. We have made sure that the solar arrays are angled to protect the sensitive back of the arrays and minimize the surface area presented to the meteor direction."


  
Figure 21: Chandra's orbit in comparison to the predicted path of the 2001 meteor shower and the location of previous Leonids. (From Bill Cooke.)
\begin{figure}\centering
\resizebox{\textwidth}{!}{\includegraphics{leonids_fig3.ps}}\end{figure}

LINK TO POSTSCRIPT FILE for Figure 21

``We've gone through a number of other meteor showers before with Chandra and we've done all our homework for this event," said Brissenden. ``Still, I think we'll all breathe a little sigh of relief when this weekend is over."


  
Figure 22: This plot shows one prediction of the Leonid meteor shower's intensity versus time. This model predicts the Leonids will peak around 20:00 Universal Time on Nov. 18. (From Bill Cooke.)
\begin{figure}\centering
\resizebox{\textwidth}{!}{\includegraphics{leonids_fig3a.ps}}\end{figure}

LINK TO POSTSCRIPT FILE for Figure 22

On Sunday, November 18, 2001 at 6:45 a.m. two cars lined up at the security gate to Chandra's Operation Control Center (OCC) parking lot. The drivers waved at each other. They and other members of the Flight Operations and Science Operations teams arrived early on this Sunday morning to keep watch during Chandra's encounter with the 2001 Leonid event.

Actually, these are the ``late" arrivals. Some team members had arrived at 5:00 a.m. and still others had been on overnight shifts. Once inside the main OCC control room, staff swapped stories about the meteors seen at viewing sites worldwide earlier in the morning. In addition to scientific curiosity and the beauty of the spectacle, many Chandra staff were motivated to get up extra early to see the Leonids with their own eyes, giving them an idea of what the spacecraft would be encountering in space a few hours later. (Figure 23)


  
Figure 23: Meteors from the Leonid Shower as seen during the early morning hours of Nov. 18 from Valley Forge National Park, PA, USA (Photo: John Welsh)
\begin{figure}\centering
\resizebox{\textwidth}{!}{\includegraphics{leonids_fig4.ps}}\end{figure}

LINK TO POSTSCRIPT FILE for Figure 23

Janet Houser, of the OCC Facility Systems Team, up at 3:00 a.m. and at her desk by 5:00, didn't have time to stop to see the meteors. It was her job to see that all computers, software, and communications systems were functioning during this crucial pass. (Figure 24)


  
Figure 24: Janet Houser, of the OCC Facility Systems Team monitoring critical Chandra funct ions. (Photo: CXC)
\begin{figure}\centering
\resizebox{\textwidth}{!}{\includegraphics{leonids_fig5.ps}}\end{figure}

LINK TO POSTSCRIPT FILE for Figure 24
A few days before the Leonids, a sequence of planned commands was sent to Chandra from the OCC via the Deep Space Network - the telephone line, so to speak, between the Earth and this orbiting spacecraft. One of these precautionary steps included turning off the high voltage to the High Resolution Camera (HRC) on Thursday, three days before the Leonid storm. Observations with the Advanced CCD Imaging Spectrometer (ACIS) continued until Saturday, when it was safely stowed in a secure position aboard the spacecraft.

As the predicted peak of the Leonids approached late Saturday night (EST), Chandra maneuvered to the first of several anti-radiant positions. The term ``radiant" refers to the direction from which the Leonid meteors would be coming. Pointing Chandra in the exact opposite direction allows maximum protection for the most vulnerable parts of the spacecraft.

To allow for continuous monitoring of the spacecraft as it passed through the predicted peaks of the Leonids, the team arranged an unusually long live contact between the control room and two Deep Space Network stations (Canberra, Australia and Goldstone, California) for approximately fourteen hours.


  
Figure 25: Members of the Chandra team discussing maneuvers of the spacecraft during the Le onids; from L-R: Bill Simmons, Dan Shropshire, Leon McKendrick, Ken Gage. (Phot o: CXC)
\begin{figure}\centering
\resizebox{\textwidth}{!}{\includegraphics{leonids_fig6.ps}}\end{figure}

LINK TO POSTSCRIPT FILE for Figure 25

At 7:10 a.m. EST, November 18, Chandra encountered the first of the two predicted debris peaks in its orbit, the Brown/Cooke Peak, named for the two scientists who modeled the forecast. Their calculations predicted that the concentration of debris would be relatively spread out and of less concern than the other peak that Chandra would pass through several hours later. Shortly after the Chandra spacecraft exited the Brown/Cooke Peak, Flight Director and CXC Manager, Roger Brissenden, announced ``there is no indication of any anomaly in the monitoring data" and concluded that Chandra was not affected by any of the debris. (Figures 20, 21, 25 and 26)

Chandra's orbit for the 2001 Leonid event posed some difficulties that called for an innovative approach to get Chandra into the anti-radiant relative to the second area of predicted intensity. The Asher/McNaught Peak was located on the part of Chandra's orbit just after the exit from Earth occultation. During occultation, the position of the Earth blocked the target guide stars, making it impossible to maintain the correct orientation through the usual interaction of aspect camera and gyros.

Additionally, Chandra's aspect camera, the instrument that detects and locks on to guide stars to provide information to orient the satellite, is typically not operated within 20 degrees of the Earth. This is because the Earth reflects sufficient sunlight to saturate the camera's detectors. If the guide stars are lost or cannot be acquired, Chandra could go into a ``bright star hold" or ``normal sun mode" and require a time-consuming series of steps to recover operations. This was not a chance the Chandra team wanted to take during a meteor event.

The answer that the planning team developed was to place Chandra in its second anti-radiant position before it entered occultation of the Earth. (The trick was to keep it there when the guide stars were blocked by the Earth, but more on that later!) The maneuver was executed (again by pre-loaded command) at 7:46 a.m., about 45 minutes before Chandra reached perigee, the point in its orbit closest to Earth. At this place in the orbit, Chandra was still ``in comm" with the Deep Space Network (DSN) station in Canberra, Australia. As Canberra disappeared over the horizon, the Goldstone DSN station in California was beginning to rise.


  
Figure 26: A screen from the Satellite Tool Kit that was used by the Flight Ops team throug hout the Leonid passage. This image shows Chandra's exact position and orientati on between the two DSN stations, Canberra, Australia (lower left), and Goldston e, California (upper right) (Photo: CXC)
\begin{figure}\centering
\resizebox{\textwidth}{!}{\includegraphics{leonids_fig7.ps}}\end{figure}

LINK TO POSTSCRIPT FILE for Figure 26

While Canberra continued as the official ``comm" (point of communication), Goldstone acquired and began tracking Chandra. About half an hour later, the official handover was made and Goldstone took over the communications for the rest of the Leonid passage.


  
Figure 27: The Flight Director, Flight Team Manager, Flight Engineers, and PCAD (Pointing C ontrol and Attitude Determination) Engineers closely monitor Chandra's status du ring Earth Occultation. (Photo: CXC)
\begin{figure}\centering
\resizebox{\textwidth}{!}{\includegraphics{leonids_fig8.ps}}\end{figure}

LINK TO POSTSCRIPT FILE for Figure 27

At this point, the crucial command to place Chandra in ``gyro hold" was successfully executed. ``Gyro hold" is a standard mode (the same one used for maneuvers) that stops using the information coming from the star camera and relies solely on the gyroscope data to hold position. Thus, in this key part of the orbit where the guide stars could either be obscured by reflected Earth light or blocked by the Earth's disk, Chandra maintained position based on the last information that the gyros had from the guide stars. That information held Chandra in its second anti-radiant position.

Transit from a position 20 degrees on one side of Earth to 20 degrees on the other side of Earth took about 3 hours. During this time, the assembled team closely monitored Chandra's attitude for ``gyro drift" - any indication that the inertial position is changing. If gyro drift had occurred, Chandra could have had difficulty reacquiring the guide stars needed to orient it away from the Asher/McNaught peak when Earth occultation ended. To cover all bases, contingency plans were made for recovery from a ``bright star hold". During this time, Dan Shropshire, head of the Flight Operations engineering team, and engineer Bill Simmons, provided continual updates from real-time engineering data. All indicators showed that Chandra's orientation was holding steady. (Figure 27) At the same time, the team had planned an important experiment. Because Chandra was in gyro hold mode, the aspect camera was freed up for a calibration test. Chandra has been operating on the conservative assumption that guide star acquisition must take place 20 degrees away from the bright limb on either side of the Earth's disk. In the weeks prior to the Leonid shower, the Chandra team saw a chance to test this assumption in real time.


  
Figure 28: Rob Cameron of the Science Operations Team at his console during the aspect came ra calibration test.
\begin{figure}\centering
\resizebox{\textwidth}{!}{\includegraphics{leonids_fig9.ps}}\end{figure}

LINK TO POSTSCRIPT FILE for Figure 28

Rob Cameron, a member of the Science Operations team, was sitting on another console tracking the now independent operations of the aspect camera. (Figure 28) There was much speculation on when the last guide star would be lost as Chandra headed for Earth Occultation. As a prize for the best estimate, Bill Simmons offered up the lone remaining bagel in the ``break room." Amazingly, the aspect camera was able to maintain its lock on the guide stars to within 1 degree of occultation, probably because night was rising on that side of Earth. (There was no official word on who ate the bagel.)

  
Figure 29: Confirmation that Chandra's aspect camera has acquired 8 guide stars. The row of small windows in the upper right of the screen contains actual images of the 8 stars taken by the aspect camera. (Photo: CXC)
\begin{figure}\centering
\resizebox{\textwidth}{!}{\includegraphics{leonids_fig10.ps}}\end{figure}

LINK TO POSTSCRIPT FILE for Figure 29

The real test was the reacquisition of the guide stars after Earth transit. Rob Cameron had 8 small squares running along a window at the top of his console. One by one the little windows were filled in with pixellated images of the 8 stars that the camera was told to find (these were the same guide stars that Chandra would need to acquire later in order to leave gyro hold (Figure 29)). Despite the bright Earth light, Chandra's very sensitive aspect camera was able to identify and hold the stars as early as 4.7 degrees past Earth. Ken Gage, head of the Flight Operations mission planning team, was excited to start examining this data to understand its implications for future operations.

At 2:34 p.m. Chandra left gyro hold and successfully acquired the guide stars that continued to keep it oriented to the Leonids' anti-radiant. About an hour and a half later, Chandra passed through the Asher/McNaught peak of the debris stream without incident. Smiles of relief and satisfaction were seen around the control room as that second area of danger was safely traversed. After the excitement, people retreated to various offices to make notes and file reports, and gradually, around 5 p.m., the control center pretty much emptied out. (Figure 30)


  
Figure 30: Checking the HRC status; John Chappell of the HRC instrument team, and Roger Bri ssenden. (Photo: CXC)
\begin{figure}\centering
\resizebox{\textwidth}{!}{\includegraphics{leonids_fig11.ps}}\end{figure}

LINK TO POSTSCRIPT FILE for Figure 30

At 5:45 p.m., many hours after her day at the Control Center started, Janet Houser was still at the OCC, as were Darell Wicker, the Operations Controller, and Roger Bachiere, the command controller. They awaited the arrival of the night team. The second shift of the Ground Operations team monitored computers down the hall. Chandra was still in the anti-radiant position as it traveled through the tail of the debris stream. The ACIS instrument had been moved back into the focal plane. Around 9:00 p.m. EST Chandra started taking science observations again.

- Dan Shropshire and Roger Brissenden


next up previous contents
Next: Venus Up: Chandra News Previous: HELP DESK
cxchelp@head-cfa.harvard.edu