The Next Generation Space Telescope (NGST) will incorporate leading-edge technology in construction, optics, and deployment. NGST will be composed of a large sunshield and lightweight mirror, which will be deployed in space. Both are depicted in this rendering by TRW Space and Electronic Group. Over the next two years, the teams, one led by Lockheed Martin Missiles & Space, Sunnyvale, CA, and the other led by TRW Space and Electronic Group in Redondo Beach, California, and including Ball Aerospace and Technologies Corp., Boulder, CO, will tackle some daunting engineering challenges. The new telescope will pick up where the Hubble telescope leaves off. Hubble observes objects that are still in the visible light spectrum. NGST will investigate objects that are much more distant in space and will need to be sensitive to the infrared band. The testing protocol is going to receive a very high level of attention during this upcoming phase one effort, because it is one of the substantial cost elements of a program of this nature.


When it is launched as early as 2007, the Next Generation Space Telescope-the planned successor to the Hubble Space Telescope—will offer astronomers a tantalizing window on the early universe. Astronomers already have an idea of what the universe is like today and in the recent past, as well as what it was soon after the Big Bang. However, the middle period—between one million and a few billion years ago, when stars the Hubble. and galaxies were being formed—is still unobserved.

It is tills time period that NGST will focus on, offering insight into the evolution of galaxies, interactions between planets and stars, and dark matter. In July, NASA's Goddard Space Flight Center in Greenbelt, Md., took a big step toward realizing this goal when it award ed phase-one contracts to two competing teams, each to supply an observatory concept and technology plan.

Over the next two years, the teams, one led by Lockheed Martin Missiles & Space of Sunnyvale, Calif., and the other led by TR W Space and Electronic Group in Redondo Beach, Calif., and including Ball Aerospace and Technologies Corp. of Boulder, Colo., will tackle some daunting engineering challenges.

The telescope will orbit in deep space at temperatures approaching absolute zero. A large sunshield also must be deployed to protect the telescope from the sun's rays. The telescope's light-weight main mirror, measuring 8 meters across, will comprise segments that will be unfurled in space. Not the least of those challenges is the fact that the NGST program will have to work under a tight budget. Detail design and construction costs for the telescope have been capped at $500 million, in 1996 dollars. This is approximately one-quarter the development cost of the Hubble.

"This is a bigger leap forward than Hubble was in its day," remarked Paul Geithner, a senior systems engineer on the NGST program who also worked on the Hubble telescope. "We have more development work to do and more risk at this point in time." Contractors will be required to demonstrate that their designs in corporate technology that is feasible under the cost constraints.


Deep Space

In terms of its goals, the new telescope will pick up where the Hubble telescope leaves off. Hubble observes objects that are still in the visible light spectrum. NGST will investigate objects that are much more distant in space and will need to be sensitive to the infrared band.

"As you look farther and farther into space, you are seeing things moving faster and fas ter away from you," explained Geithner. Light waves of the objects that will be viewed by NGST have been Doppler-shifted to the infrared spectrum.

"Everybody would love to know what really happened to form the first stars and galaxies. If we could look back to when those first galaxies' first stars lit up, that could answer a lot of questions," Geithner said.

To look that deeply into space requires an infrared telescope. Visible light waves-those viewed by the Hubble- are basically 0.4 to 0.7 micron long. NGST's goal is to be sensitive to wavelengths of 1 to 5 microns.

Although infrared astronomy is done from the ground today, ground telescopes are hampered by less than ideal conditions, such as atmospheric turbulence and thermal emissions that degrade images from infrared detectors and optics, which perform best in very cold temperatures.

To sidestep these difficulties, and also to get a wider field of view, NGST will operate far from the Earth, in a part of the sky named Lagrange point 2. The position, known as L2, makes sense for a couple of reasons, Geithner said. A million miles away from Earth, at a point at which the gravitational forces between the Sun and the Earth cancel out, it is far enough from home to avoid heat from the Earth's surface. The Hubble Space Telescope orbits about 380 miles from Earth.

According to Ralph Schilling, NGST program manager for TRW, operating in the infrared spectrum will require that most components on the telescope be cooled to around 30 kelvin to reduce the risk of degrading images. The spacecraft can be cooled passively, without the need for cryogenic liquids, by operating in the shadow of a large sunshield that will be deployed in space. Although passive cooling has been employed on spacecraft before, this may be the first time it has been applied on this scale, said Geithner.

The sunshield will probably consist of flexible blankets of insulating material that will be stretched in space by. some sort of framework. Requirements of the sunshield are that it be flexible, thin, and very lightweight.

According to Geithner, "We have been putting special blankets around spacecraft for years, to keep them from getting too hot or too cold. But in this case, we are going to be way out there for a long period of time." The large surface—roughly the size of a tennis court-and long period of deployment require that the shield be thermally efficient, particularly the surfaces exposed to direct sunlight, he added.

Bigger is Better

NGST is expected to have 10 times the light-gathering capabilities of the Hubble Space Telescope, and will be able to see objects 400 times fainter than those currently observed by big infrared ground telescopes. It will achieve this with a spatial resolution, or image sharpness, comparable to the Hubble.

At 8 meters in diameter, the mirror will be significantly larger than that on the Hubble, which measures 2.4 meters across.

" It's 10 times the collecting area at one-fourth the cost. That's the challenge," said Chuck Rudiger, the capture manager of the Lockheed team. "The Hubble mirror was done like a ground-based telescope. It was a big, heavy piece of glass that weighed tons."

NGST's main mirror, on the other hand, will be much thinner-on the order of a windowpane-and roughly one-tenth the weight of its predecessor. "We are shooting for 15 to 20 kilograms per square meter for the area mass density of these mirro rs," Geithner said. Hubble's mirror weighs 186 kg per square meter.

The 8-meter-diameter mirror is too large to be incorporated as one piece into the launch vehicle. To fit it into a launch shroud that will be roughly 5 meters in diameter, the main mirror will have to be launched as a deployable assembly that will unfurl in space.

"Various concepts exist for doing that," said Schilling. One concept is a flower-like design, in which each petal is a rigid structure on which a mirror segment is mounted. The petals will be coupled with a series of hinges and motors that drive the deployment in orbit, he explained.

The deployment-a one-time event in which reliability is critical-will probably be accomplished by stepper motors or brushless dc motors, he added. Depending on whether the mirror is deployed before or after the sunshield, the motors mayor may not have to work in cryogenic temperatures.

A number of materials are being considered for the mirror substrate. These include silicon carbide and carbon silicon carbide, which are relatively new materials for mirror applications, said Schilling. Also being considered are more traditional materials, including beryllium, which is attractive for cold temperature telescopes because it has a low coefficient of thermal expansion at very cold temperatures.


Bed of Nails

One key element in the main mirror's design is the ability to make minute adjustments to its surface geometry. This makes possible the use of a thin mirror surface in a zero-gravity, very cold environment, said Lockheed's Rudiger. Using a sort of bed-of-nails concept, the mirror surface may sit on a series of actuators that can move up or down to control the mirror surface.

The concept of an adjustable surface is not unique to space telescopes. Rudiger said ground-based mirrors have been designed with adjustable surfaces.

Alignment of mirror elements is critical, because you want to form a very high-resolution image.

The inherent flexibility of the mirror surface will be an advantage in space, because it allows the mirror to compensate for the thermal contractions that will take place, said Geithner. "Even small temperature changes can cause a big effect optically, because the wavelengths are tiny," he said. "If we are looking at 1 to 2 microns in the wavelength of light, we have to be aligned to a fraction of that. We are talking a few tenths of nanometers, in terms of figure control."

The actuators may incorporate new piezoelectric materials that work well at cryogenic temperatures, said Geithner. One advantage of piezoelectric materials is that they can accomplish very small step adjustments. An idea being considered is to combine the piezoelectic materials with mechanical devices such as lever arms, to increase the throw of the actuators. There are even ballscrew and motor concepts that work at these temperatures, he added.

The mirror's flexibility allows it to sidestep problems encountered by Hubble's monolithic mirror. That mirror surface's edge was ground too flat, so that the light was not collected optimally, a defect that was compensated for by instrumentation. "With NGST, we won't have to worry about that problem. With the correction range of the mirror design, we will be able to correct and achieve whatever figure we want," remarked Geithner.

Although gaps will exist between the mirror segments, these will manifest themselves as relatively minor diffractions that can be handled optically, said Geithner. A bigger challenge is to align the edges of the mirror segments closely, on the order of tens of nanometers.

"Alignment of these mirror elements is very critical, because you want to form a very high-resolution image," said Schilling of TR W. A small misalignment from one element to the next, or even small distortions in the shape of an individual element, can seriously degrade imaging capability. "You must have very stiff structures and you need to achieve very precise alignment when you deploy that structure in orbit."

For NGST, a technology called in1age-based wavefront sensing control is being considered to provide a feedback loop to keep mirror elements aligned. This technique electronically infers distortions that are occurring on the mirror surface by analyzing the focal plane of light hitting the telescope. The information is used to introduce a correction to the mirror surface. It's done iteratively, and it takes a lot of computing power, said Geithner. But it's a proven technique that was used to identify the problem with the main mirror of the Hubble, he added.

Vibration is a special concern. "This telescope is amazingly sensitive to any kind of jitter," Geithner said. Cryogenic temperatures exacerbate the problem because structures become very stiff, and damping is greatly diminished at very cold temperatures, he explained. "Once things start vibrating, they continue to vibrate," he added.

Trying to make the whole spacecraft very rigid may make its launch and deployment in space impractical, he said. The attitude of the craft will be controlled by big momentum wheels. To take the jitter out of the optical path, a small mirror will be located in the optical train of the telescope. The mirror will steer the beam with a high degree of precision before the light reaches the detectors, to compensate for any vibration.

The biggest challenge before reaching orbit, however, may be on the ground, in finding a realistic test environment. In the case of NGST, this means cooling down the telescope to 30 K and finding a vacuum chamber large enough to house it. Currently, there is no test chamber that simultaneously meets the size, temperature, and cleanliness requirements needed to test NGST as a single assembly, said TR W's Schilling. One option is to test the equipment in a piecewise fashion, and then extrapolate what the performance of the entire system will be, without having to test the entire 8-meter mirror as a single assembly. "The testing protocol is going to receive a very high level of attention during this upcoming phase one effort, because it is one of the substantial cost elements of a program of this nature," he said. Another hurdle could be creating a vibration-free environment.

Testing will also require building an optical simulator that is capable of simulating the light from a distant star. "You can't just stick a cutoff piece of fiber optic at one end of the chamber and have the telescope look at it," said Geithner." It would be in the near field of the telescope and you wouldn't be able to focus." Instead, the approach will require a sort of reverse telescope, making the image appear far away. And the test equipment would have to operate under cryogenic conditions.

Even so, NGST itself, when it is finally launched, will be a test for future generations of space telescopes. Already, space telescopes with mirrors that are 20 meters across are envisioned.