Hubble Space Telescope Performance Report

NASA Fact Sheet March 25, 1991


Background

	No research mission in NASA history has generated higher
scientific expectations than the Hubble Space Telescope (HST).
And few recent missions have received such widespread public
attention, either before or after launch.

	But that's understandable: HST was designed to be the
most powerful astronomical telescope ever built, far surpassing
the capabilities of ground-based optical telescopes for many
kinds of forefront research.  The key to HST power is operation
in space--above the blurring, obscuring and absorbing effects of
the Earth's atmosphere.

	The April 1990 launch of HST was flawless.  Then, two
months later, came the disappointing news of an optical defect in
the HST primary mirror that blurs the telescope's focus--which
prompted some to label the project a disaster.

	 Now, however, a more balanced picture of current HST
capabilities has emerged, together with a much more hopeful
vision of the telescope's research future.  Outstanding
scientific research is already being carried out with HST.
Moreover, instrument upgrades to be provided by future Space
Shuttle servicing missions will permit most, and perhaps nearly
all, of the original HST scientific objectives to be achieved
over the mission's planned 15-year observing lifetime.

HST Optical Design

	The HST optical design is similar to that of many
ground-based telescopes; a schematic of the HST Optical Telescope
Assembly (OTA) is given in Figure 1.  The curved HST primary
mirror, 2.4 meters (nearly 8 feet) in diameter, collects the
light from the object under study and reflects it toward a much
smaller secondary mirror, which reflects the light back through a
central hole in the primary.  The light comes to a focus at the
focal plane behind the primary mirror, producing an image of the
object that can be recorded or analyzed by the HST scientific
instruments.

	Operation above the Earth's atmosphere brings big gains
in performance.  A telescope in space provides much finer image
detail than the same telescope on the ground and permits the
study of fainter objects.  In addition, a space telescope can
observe radiation that is absorbed by the atmosphere and is
therefore unobservable from the ground.  For example, HST was
designed to detect both ultraviolet and infrared radiation as
well as visible light, with emphasis on ultraviolet and
visible-light observations during the initial phase of the HST
observing program.

Fine Detail

	The ability to distinguish between objects that appear
close together on the sky--or equivalently, to record fine detail
in an image--is known as spatial resolution.  High spatial
resolution is needed, for example, to pick out individual stars
in a densely packed star cluster or to probe the core regions of
remote galaxies for evidence of black holes.

	Atmospheric turbulence, however, blurs the view of all
ground-based optical telescopes.  Even under the best "seeing"
conditions, star images are smeared out.  Images of extended
objects--such as planets, comets, luminous gas clouds and
galaxies beyond our own--suffer a loss in detail and clarity.

	The resolution of a telescope in space, by contrast, is
limited only by the diameter of the primary mirror and the
quality of the optical system.  With its 2.4-meter primary, HST
was designed to yield spatial resolution about 10 times better
than that normally achieved by ground-based optical telescopes.

Faint Objects

	Looking out into space means looking back into time.  The
light from a galaxy a billion light-years away, for example, left
that galaxy a billion years ago; we therefore observe such a
galaxy not at it appears today, but rather as it appeared a
billion years in the past.  The farther out we look, the farther
back into time we see.  Observations of very distant galaxies
thus provide important clues to the early history of the
universe.

	Since the most distant galaxies are very faint, however,
their optical images are very difficult to record from the
ground.  Even at night, the Earth's atmosphere itself glows
faintly with emitted and scattered light.  When a galaxy is
fainter than this so-called "sky background," its image is simply
"lost in the noise," like a whisper at a crowded party.
Recording an image under these conditions requires an extremely
large telescope, or unreasonably long exposure times, or both.

	Faint-object observations from space are free of this
atmospheric effect.  They are limited almost entirely by
unavoidable, random noise within the electronic devices used as
light detectors.  Modern detectors are much less "noisy" than the
atmosphere.  HST was therefore expected to image much fainter
galaxies than could be imaged in practice by any ground-based
optical telescope.

Ultraviolet Radiation

	The light our eyes can see represents just one energy
region of the total spectrum of electromagnetic radiation, which
includes radio waves (at the low-energy end of the spectrum),
infrared radiation, visible light, ultraviolet radiation, X rays
and gamma rays (at the high-energy end).  Only the visible light
and radio waves from astronomical objects can be readily studied
from the ground (although visible light is attenuated and
distorted by atmospheric effects).  Radiation in other spectral
regions is partially or totally blocked by the Earth's
atmosphere.

	Every region of the electromagnetic spectrum carries
important information about objects or physical conditions
elsewhere in the universe.  Analysis of ultraviolet radiation,
for example, helps to reveal the chemical composition of the
interstellar medium--the tenuous clouds of gas and dust within
our galaxy that give birth to new stars.  However, only a small
fraction of the weakest ultraviolet radiation penetrates the
atmosphere. (In the case of ultraviolet radiation from the Sun,
this fraction is just sufficient to cause tanning and burning of
the skin.)

	Since ultraviolet radiation can be collected and focused
just like visible light, HST was designed to study this important
spectral region as well.  But there's a catch.  A mirror will
reflect ultraviolet radiation efficiently only if its surface is
extremely clean, as well as very smooth.  Extraordinary
precautions therefore had to be taken during manufacture of the
HST mirrors to ensure that they were free of contamination.

Launch and Deployment

	Hubble Space Telescope (HST) was launched on April 24,
1990, by Space Shuttle Discovery (see Figure 2).  After reaching
the specified orbital altitude of 330 nautical miles (370 statute
miles), the Shuttle supplied full electrical power to HST,
permitting a series of quick checks to verify that the HST
spacecraft was beginning to operate.

	On the next day, HST's solar panels were unfurled,
internal power switched on and the high-gain communications
antenna deployed.  When ground control confirmed that
communication with HST had been established, the Shuttle's
manipulator arm released the telescope into orbit.

Orbital Verification

	Before HST could become fully operational, a sequence of
orbital verification tests had to be performed.  These tests,
expected to take only a few months, included activation of the
telescope's complex pointing and control systems and careful
alignment of the mirrors in the Optical Telescope Assembly (OTA).
It was anticipated that several additional months would be needed
to fine-tune the performance of HST's scientific instruments.

	The focus test, used to determine the optimum alignment
of the OTA mirrors, was a crucial one.  The OTA was first
commanded to go completely out of focus, then to regain focus,
using starlight imaged by one of the HST cameras.

Discovery of Spherical Aberration

	Several such focus tests were carried out during the
weekend of June 23-24.  The results were totally unexpected.  No
position of the secondary mirror could be found which brought the
starlight to a sharp focus.  The star images remained blurred, as
shown in Figure 3. (The spidery "tendrils" surrounding these
images are artifacts of the secondary-mirror supports and
additional supports within the camera.)

	From the observed pattern of image blurring, it shortly
became clear that HST suffered from spherical aberration, an
optical distortion caused by incorrect mirror curvature.  By June
26, NASA Headquarters had received the disappointing news.  NASA
engineers moved immediately to find out whether the problem was
in the primary mirror or the secondary mirror and to determine
whether anything could be done to correct it.

	Because the engineers knew that a flaw in the secondary
mirror would produce an additional aberration known as coma,
tests were run on July 6 to search for this aberration.  No
evidence of coma was found.  It was then clear that the error lay
in the shape of the primary mirror.

Effects of Spherical Aberration

	The HST optical system should ideally focus the light
from a star into a sharp, almost point-like stellar image.
However, spherical aberration in the HST optics causes the
starlight to spread over a broader area, giving every stellar
image an extensive, fuzzy "halo" of light (see again Figure 3).
Although the HST mirrors are the smoothest and most uniformly
coated astronomical mirrors ever made, the shape (curvature) of
the primary mirror does not match that of the secondary mirror.
This mismatch is the cause of the spherical aberration.

Effect on Sharpness of Focus

	Figure 4 illustrates the effect of spherical aberration
on the focus of a telescope mirror.  Figure 4a shows a correctly
curved mirror: all the reflected light rays from a star converge
to a single focal point.  Figure 4b, by contrast, shows a mirror
with spherical aberration: the light rays converge to a variety
of different focal points--causing the resulting stellar image to
appear fuzzy.

	However, figures 4a and 4b do not tell the complete
story--they depict light as traveling in straight-line rays,
whereas light actually consists of waves.  Because of
interactions among these waves, some fraction of the imaged light
always falls outside the geometric center of the image.  This
effect can be included in accurate, three-dimensional computer
plots of the distribution of imaged light intensity over the
focal plane, as shown in Figure 5.

Effect on HST Images

	Figure 5a illustrates the sharp HST stellar image that
was expected.  About 80% of the starlight is focused within a
small region at the center of the focal plane; the remaining 20%
of the starlight falls outside the central peak.  However, this
slight degree of "fuzziness" is unavoidable.  Because of the wave
nature of light, the stellar image of Figure 5a is the sharpest
that can be achieved in space with a flawless primary mirror in
the 2-meter class (a larger primary would produce an even sharper
image).  Computer processing was planned from the beginning to
sharpen HST images still further.

	Figure 5b, by contrast, illustrates the appearance of an
actual HST stellar image; because of spherical aberration, only
about 15% of the light is now brought to a focus very near the
center.  Most of the light--about 85%--falls elsewhere in the
focal plane, producing a fuzzy "halo" more extensive than that in
Figure 5a.

	As Figure 5 suggests, spherical aberration limits current
HST performance in two ways.

Effect on Spatial Resolution

	First, the aberration reduces the spatial resolution of
the telescope--the ability to record fine detail.  If every star
image looked like that in Figure 5a, HST could distinguish two
stars very close together on the sky.  But because every star
image actually looks like that in Figure 5b, the images tend to
overlap, and resolving power is lost.  Because light is spread
out over a larger region of the focal plane, images of extended
objects similarly suffer a loss in detail and clarity.

	However, the loss of resolving power is less important
for bright objects than for faint ones.  The aberrated star image
of Figure 5b still retains a well-defined central "core." If
light from an object is bright enough, computer processing--which
had been planned in any case--can remove most of the fuzzy
"halos" in the HST images.  The image cores then provide good
discrimination of detail.  For bright, high-contrast objects,
such as nearby star clusters, galactic nuclei, and the major
solar-system planets, HST spatial resolution remains much better
than that of any ground-based optical telescope.

Effect on Observations of Faint Objects

	Second, the aberration reduces the ability of HST to
image faint objects, such as distant galaxies.  By contrast with
the pattern of light distribution expected, light within the
aberrated image is spread out at low intensity over a larger
region of the focal plane.  This means that faint objects are
lost either amid overlapping aberrated star images or in the
noise of the HST light detectors.  As a result, HST capability to
image faint objects is currently more like that of ground-based
telescopes, which are in practice limited by sky-background
noise.

Effect on Spectroscopy

	Spectroscopy is the analysis of radiation into its
spectrum of component colors or wavelengths.  The production of a
spectrum requires that the radiation first be passed through a
very narrow slit.  Since the HST aberration broadens all images,
less light from the image passes through the slit than was
originally anticipated.  However, this effect can be compensated
for by increasing the exposure time.  Excellent spectroscopic
measurements can therefore still be made with HST. Because early
priority was given to imaging, the current HST spectrographs are
only now beginning to demonstrate their potential for ultraviolet
and visible-light analysis.

Effect on Ultraviolet Observations

	Fortunately, the HST mirror surfaces appear to be
extremely clean, and ultraviolet reflectivity is good.  HST
therefore continues to provide unmatched capability for
observations in the ultraviolet spectral region.

Testing of the HST Primary Mirror

	The shaping of a telescope mirror includes many cycles of
grinding and polishing a disk of glass; each cycle refines the
shape left by the previous cycle.  This process is always guided
by sensitive optical tests designed to reveal the difference
between the current shape of the mirror and the shape desired.
When the desired shape is achieved, the glass is coated with a
thin film of highly reflective aluminum.  The shaping, testing,
and coating of the HST mirrors were carried out by a private
optical firm under contract to NASA.

How the HST Primary Mirror Was Tested by the Contractor

	The test used for the HST primary mirror surface was a
full-aperture interferometry or "null" test.  A schematic diagram
of the null-test system (not drawn to scale) is given in Figure
6.  An optical device called a null corrector, consisting of a
lens and two curved mirrors, played a key role.  It had been
specially designed and constructed to make a perfectly shaped HST
mirror look spherical.

	In essence, the test system compared light reflected from
a flat reference mirror with light reflected from the curved HST
primary mirror, as modified by passage through the null
corrector.  Light beams from the two mirrors were brought
together, so that the light waves in the beams could interfere
with each other and form an optical interference pattern.  At
each stage of the grinding and polishing process, this
interference pattern was photographed and analyzed to determine
the departure of the current mirror shape from the shape desired.

	Throughout the process, the null corrector was assumed to
be performing as intended.  Unfortunately, undue reliance was
placed on this test device.

	After final testing--which displayed the correct
interference pattern--the optical technicians and NASA managers
were confident that they had constructed the most accurate
astronomical mirror ever built.  However, the outer edge of the
HST primary had in fact been polished too flat by an amount equal
to 1/50 the width of a human hair--a big error in optical terms.

Cause of the Incorrect Mirror Shape

	In August 1990, NASA formed a special board of
investigation, under the leadership of Jet Propulsion Laboratory
director Dr. Lew Allen, to determine why the HST primary mirror
had been incorrectly shaped and why this error had not been
detected before launch.  The Allen committee issued its report to
NASA (the "Allen Report") in November 1990.

	After a detailed review of the mirror fabrication
process, the Allen committee identified the null corrector as the
cause of the problem.  They found that the spacing rods used to
set up the null corrector had been incorrectly used, producing a
1.3-mm error in the spacing of the null-corrector lens.  Detailed
analysis showed that this lens spacing error is almost exactly
that required to explain the apparent curvature error in the HST
primary mirror.

What Actually Happened

	The executive summary of the Allen Report draws these
conclusions about activities at the contractor site:

	"No verifications of the reflective null corrector's
dimensions were carried out... after the original assembly.
There were, however, clear indications of the problem from
auxiliary optical tests made at the time, the results of which
have been studied by the Board. A special optical unit called an
inverse null corrector was built and used to align the apparatus;
when so used it clearly showed the error in the null corrector.
A second null corrector, made only with lenses, was used...  It,
too, clearly showed the error in the primary mirror.  Both
indicators of error were discounted at the time as being
themselves flawed..."

	Unfortunately, this information was never brought to the
attention of NASA. The report continues:

	"Reliance on a single test method was a process which was
clearly vulnerable to simple error...  During the critical time
period, there was great concern about cost and schedules...

	Questions have also been raised about why a full-scale
test of the completed HST Optical Telescope Assembly was not
conducted before launch.  The Allen Report comments as follows:

	"An end-to-end test of the OTA would have been very
expensive to perform at the level of accuracy specified for the
telescope.  The test would have cost on the order of what the OTA
itself cost. . . the test could have required two additional
mirrors as large as or larger than the OTA primary..."

	Moreover, any such tests would have introduced a serious
risk of mirror contamination.  As stressed earlier, HST
performance in the ultraviolet spectral region depends critically
on the cleanliness of the mirrors, in addition to their
smoothness.  Any contamination would have seriously degraded HST
performance in the ultraviolet spectral region.

Current HST Capabilities

	Because of spherical aberration, some of the forefront
research originally planned for HST will have to be deferred
until later in the decade.  For bright, high-contrast objects,
however, HST spatial resolution still surpasses that of any
ground-based optical telescope, permitting HST to produce images
of such objects with a clarity and detail never seen before.  For
observations of faint extended objects, such as distant galaxies,
HST remains the equal of large ground-based telescopes, although
such observations are not planned for the near term.  And HST
still offers unmatched capability for ultraviolet observations.

	HST currently has five scientific instruments available
for observations at the focal plane, plus three Fine Guidance
Sensors, which can also be used for astronomical observations.  A
brief description of these instruments follows.

	The Wide Field/Planetary Camera I (WF/PC I) is designed
for high-resolution imaging of faint, extended objects in our own
and other galaxies, and of solar-system bodies, in the visible
and near-infrared spectral regions.  The workhorse of the HST
observing program, WF/PC I is seriously affected by the
aberration.  However, WF/PC I can still be used to study bright,
high-contrast objects, such as major solar-system planets and
nearby star clusters and galaxies.  The camera has already
returned spectacular images of a giant storm on Saturn (see
Figure 7) and has revealed new detail in the compact cores of two
galaxies, among many other achievements.

	The Faint Object Camera (FOC) is intended for imaging of
the faintest objects in the visible and ultraviolet spectral
regions at very high spatial resolution.  Its performance is
degraded by the spherical aberration, but the sharp image cores
still allow the FOC to detect details not seen by ground-based
telescopes.

	The current capabilities of the FOC have been extensively
demonstrated during the first months of HST operation.  For
example, the camera has imaged a high-energy jet in a distant
radio-emitting galaxy, resolved Pluto and its close satellite
Charon, probed the core of a nearby exploding star, and resolved
quasar images in the "Einstein Cross" galactic gravitational
lens.  Figures 8 and 9 illustrate other FOC accomplishments that
depend upon HST capability for high spatial resolution of bright
objects.

	Figure 8 compares a ground-based image with an FOC image
of a highly compact light source within a rich star-forming
region in the Large Magellanic Cloud (a nearby galaxy visible
from the Southern Hemisphere). This object had been suspected to
be either a very large and luminous star or a cluster of
individual stars; the FOC resolved the cluster and settled the
issue.

	Figure 9 presents another significant FOC achievement: a
detailed image of the remnant of Supernova 1987A in the Large
Magellanic Cloud, showing a complete ring structure surrounding
the exploded star.  Ground-based observations could at best
reveal only parts of the ring.

	The Faint Object Spectrograph (FOS) is designed to
analyze the light from very faint objects in the visible and
ultraviolet spectral regions.  Although the faintest objects
cannot now be reached, observations of brighter sources are only
moderately degraded.  Because most of the initial HST observing
time was devoted to imaging, the FOS only recently has begun to
demonstrate its productivity.

	The Goddard High Resolution Spectrograph (GHRS) is
intended for very detailed analysis of ultraviolet radiation.
The instrument now loses spectral resolution on the faintest
objects, but as in the case of the FOS, observations of brighter
sources are only moderately degraded.  Because HST provides a
unique capability in the ultraviolet spectral region, the GHRS is
still expected to become a highly productive instrument.

	The High Speed Photometer (HSP) is designed for accurate
measurements of light intensity and its fluctuations at temporal
resolutions of up to 100 readings per second.  Because little
observing time has so far been allocated to this instrument, the
impact of the aberration has not yet been well determined.

	The Fine Guidance Sensors (FGS) are designed not only for
accurate HST pointing, but also for astrometry--precise
measurements of stellar positions and motions.  The capabilities
of these sensors may be little affected by the aberration, but
conclusive tests are still under way.

	On balance, this assessment shows that near-term HST
programs are best directed toward brighter, high-contrast objects
and ultraviolet observations.  However, exposure times will need
to be two to seven times longer than originally planned, because
less light is now available in the core region of the image.  The
HST observing schedule will therefore be completely reorganized
in order to optimize the use of the telescope during the first
years (1991-1993) of its planned 15-year lifetime.

Servicing Plan

	The Hubble Space Telescope was designed to achieve an
overall mission lifetime of approximately 15 years through
on-orbit maintenance and repair.  From the beginning, provision
was therefore made for replacement of critical spacecraft systems
and scientific instruments.

	Although the HST primary mirror cannot be replaced on
orbit, several "second generation" scientific instruments have
long been planned as part of the HST mission strategy.  These
will eventually replace some of the current instruments in order
to extend and advance HST's scientific capability throughout its
operational lifetime.

	The first replacement instrument is the Wide
Field/Planetary Camera II (WF/PC II), which is under development.
This instrument will now be slightly modified to include its own
corrective optics--eight dime-sized relay mirrors with just
enought curvature to restore the focus of the aberrated light.
WF/PC II will be installed on HST during the first Space Shuttle
servicing mission, currently scheduled for 1993.  Additional
optical corrections designed to restore the focus of other
current HST instruments are also under review.

	Later Space Shuttle servicing missions will permit
installation of additional second-generation instruments.  These
were originally conceived to extend HST's infrared sensitivity
and add imaging capability to ultraviolet spectroscopic
observations.  Now, however, such later instruments will also,
like WF/PC II, incorporate their own corrective optics to
compensate for the effects of the aberration.  As a result of
these successive upgrades, most, if not nearly all, of the
originally planned HST observing program can be carried out by
the end of the total 15-year mission.