Getting Down To Business - User Support

User Support Group
Fred Seward explains scheduling situation, as Andrea, Eric, and Nancy look on.
Photo: CXC
November 23, 1999 Cambridge, MA :: The Chandra mission has moved into a new phase. Most of the calibration targets have been observed. The Calibration Team will continue to study the results of these observations for months to come so as to better understand the performance of the instruments in orbit, and they will observe other targets from time to time as the need arises. But now individual scientists have their turn.
"The Chandra mission has moved into a new phase."


First in line are the Guaranteed Time Observers (GTO's). These are the four instrument principal investigators, the telescope scientist and six interdisciplinary scientists who have also served on the Chandra science working group. For the next couple of months Chandra will be observing some of these scientists' favorite targets. After that the General Observers (GO's) -- scientists who have been awarded viewing time through the peer review process -- will get their turn.

The Chandra User Support group consists of scientists headed by Fred Seward, with Belinda Wilkes as deputy leader. They act as an interface between observers and the teams involved in planning and implementing the operation of the telescope. We sat in on one of the User Support group meetings recently.

Seward began by explaining that operating an observatory in practice is not the same as operating one on paper.

User Support Group
Dan, John, and Fred discuss GTO plans.
Photo: CXC
"They have to change things in real time while the observation is being made," he said. "Some mechanism may not work, or the software may have a bug, or a solar flare alert may go off. As a result, it is a struggle to determine what actually was done during the observation -- how long Chandra actually looked at the target, for example...Joan is making up a list which we will post on the web."

The process of scheduling a sequence of observations is not simple, either. Mission Planning schedules a week's worth of observations, a software scheduling program puts them in time order, and they are piped up to the spacecraft. So far, so good.

User Support Group
Nancy, Himel, Roy, and Joan talk about planning the next observing cycle.
Photo: CXC
"But," Seward said, "the scheduling program can take twelve hours to run. This means that you have a twelve hour lag if you want to make a change. You can't make changes in the schedule easily. You have to put them in by hand."

All this has been complicated by a scary episode in September during which a portion of the X-ray camera called the Advanced CCD Imaging Spectrometer (or ACIS) began to misbehave. (See "A Closer Look") Although the problem has stabilized, it has made the analysis of the data more difficult and the observing schedule had to be rearranged while Chandra scientists investigated the problem.

Recent Headlines
"The GTO's are aware of the ACIS situation," Seward said. "But the GO's may not be."

The GTO's each have from one to ten days of observation time during the first phase. They can choose to split it up into a number of targets, or splurge it all on one very long observation. Most choose to split it up. After consulting with User Support, they send their targets to Seward who forwards them to Mission Planning.

Mission Planning, under the direction of Bill Forman, makes out an observing plan, which takes into account the angle of the Sun, the time when Chandra will enter the radiation belts, and the time required to slew from one source to another.

Chandra in orbit - radiation belts
A schematic side view of Chandra's orbit, showing the inner and outer radiation belts.
Illustration: CXC/M.Weiss
"Do we give our opinion on the sources to look at?" Dan Harris asked.

"You can if you think it will help," Seward replied.

"I understood that our job is to interface with the users and not choose targets," Harris said.

"That's correct," Seward agreed. "But it would be nice to make sure that all the GTO's get at least one observation into the schedule for the next few months."

After Mission Planning, the plan goes back to Seward, who contacts the observer. When the observation is completed, the data must go through the data pipeline where it is processed to take out obvious instrumental effects so the observers can use it. This process is taking longer than anticipated-- it always does in any mission of this complexity --and the User Support group hears the complaints first.

"They are anxious to see their data, and can get abrupt sometimes," Seward said.

"Chandra's first images have dazzled scientists..."

collage of first chandra images
Clockwise from top left: Cas A, Crab Nebula, 3C295, Eta Carinae
But he has been through it all before as the head of User Support for the Einstein Observatory twenty years ago, so he knows that this too, will pass.

"Once they get the data and begin to analyze it, and mine it for that great discovery that it may be hidden within it, things will settle down," he said.

For good reason. There is no time to waste. The observers only have exclusive rights to the data for twelve months after they receive it. Then it goes into the public archives, where other prospectors can search for the precious gems of new knowledge of the cosmos.

In the meanwhile, User Support must stay ahead of the curve. The announcement for the next round of observing proposals (Cycle 2) must go out by February 15. By then, the ACIS re-calibration should be completed, and the software for planning observations should be ready. It won't be simple, but the User Support group will push forward. Chandra's first images have dazzled scientists, and they are lining up in hopes of getting the opportunity to use the telescope.

A Closer Look
ACIS Schematic
ACIS schematic layout overhead view to illustrate the location of the imaging (ACIS-I) and spectroscopic (ACIS-S) arrays of CCD ships. Chips S1 and S3 are back-illuminated chips; all of the other chips are front-illuminated.
Illustration: CXC
ACIS CTI

CCD
Illustration of CCD
Illustration: CXC/S.Lee
This double acronym spelled trouble for Chandra scientists for a few weeks while they struggled to understand what was happening to the Advanced CCD Imaging Spectrometer (ACIS), one of the two X-ray cameras aboard Chandra. CTI stands for "Charge Transfer Inefficiency". The CCD chips which are at the heart of ACIS work by converting an incoming X-ray into electric charge at a specific location on a chip. The amount of electric charge created is a measure of the energy of the X ray. Precisely timed electric voltages are used to transfer the charge to electronic readout devices on the boundaries of the chips. These devices convert the charge to a signal which gives the location and energy of the incident X-ray. (See the accompanying figures.)

Chandra in orbit - radiation belts
A schematic view of Chandra's orbit from above, showing the radiation belts.
Illustration: CXC/M.Weiss
CTI means that not all of the charge is transferred to the boundary, so the energy of the X-ray is uncertain. The higher, the CTI, the more uncertain the energy. In early September, scientists noticed an alarming increase in CTI in ACIS. It was eventually determined that the radiation damage was due to low energy protons that bounce off the mirrors when Chandra passes through the Van Allen radiation belts. Now the flight operations team moves ACIS out of the focus during these passages, and the situation has stabilized. The ability of ACIS to distinguish X-rays of different energies has degraded somewhat; the exact amount varies from one location to another on the chip, and from chip to chip. Two of the chips (so-called back illuminated chips) were constructed differently, and show no radiation damage.

The increase in CTI had no effect on the image quality of any of the CCD chips.



Return to Chandra Chronicles

Highlights and Feature Stories from the Chandra Project
Archived reports of past Chandra Chronicles
Receive the Chandra Digest
Updated: March 28, 2008