The purpose of HUT was to complement the studies that are being done with IUE and HST by pressing beyond the wavelengths covered by these telescopes, down to the Lyman limit and further into the extreme UV (Davidsen et al. 1992). The resolution of HUT's spectrophotometry is moderate (3 Å), in the range 830-1850 Å in first order, overlapping the long-lived observatories above 1200 Å. The telescope's sensitivity extends down to 415 Å in second order, but has been maximized over the 900-1200 Å region, resulting in the capability to study objects as faint as 16th magnitude.
This instrument is a prime-focus telescope, with a diameter of 0.9 m, a Rowland-circle
spectrograph, and a photon-counting microchannel-plate detector. At the
focal plane, there is a choice of several different spectrograph
entrance apertures, including round apertures 
and 
in diameter used primarily to observe stellar objects, and large slits 
and 
in length which provide high sensitivity
for viewing faint diffuse nebulae. With its very fast focal ratio 
, HUT
outperforms even HST in this nebular mode. One of the apertures
contains a thin aluminum filter that passes only extreme-UV
radiation in the 400-700 Å band (Davidsen et al. 1991).
The unique characteristics required of an instrument designed for studies near the Lyman limit stem from the lack of any window or lens materials that transmit these wavelengths and from the low reflectivity of radiation in this band by normal mirror coatings. In the design of HUT we overcame these problems by eliminating the use of a window, by keeping the number of reflections to the absolute minimum (one telescope mirror and one diffraction grating), and by using coatings of osmium and iridium, heavy metals whose reflectivities are a relatively high 20% at the designed wavelengths. On the ground, the detector and its cesium iodide photocathode are protected from air by a vacuum seal at the spectrograph entrance. This seal is opened after several hours in space, when the ambient density in the shuttle cargo bay has dropped to an acceptable level. The spectrograph is kept at low pressure by attached vacuum pumps which operate between observations.
A low light-level TV camera that provides a view
of the 
arcmin field that surrounds the spectrograph aperture is also incorporated into HUT.
The image is used by the payload specialist to find a target object
and, as it turned out on Astro-1, to keep the telescope pointed at it.
A HUT microprocessor also digitizes the image and transmits it to the
ground, where operations personnel can aid in the identification of
objects. Amplified pulses from the HUT UV
detector are also digitized, and pulse centroids are computed to
determine the position of individual photons to a precision of
, which corresponds to 0.51 Å in the spectrum. The final
resolution is typically about 3 Å, except for observations of
extended objects with the wide slit, for which it is about 6 Å. Photon
arrival times are obtained with 2 s resolution for all the data, and
2 ms timing is available for objects whose count rates are less than
500 counts s
. This makes it possible to study rapid variability
of UV sources, which has not been possible with IUE or Voyager.
The net effective area achieved with HUT is about 10 cm
at
1050 Å, and is within a factor of 2 of this value over the range 900-1600 Å (Davidsen et al. 1992). A flat-spectrum object with a magnitude of 12.5
has a flux of 0.01 photon cm
s Å, yielding
nearly 100 counts s with HUT. The detector background
count rate, attributable to cosmic rays, is less than 1 count s. In a
typical observation lasting about 2000 s (limited by the low-Earth
orbit of the shuttle), HUT obtains an excellent spectrum of such an
object with a signal-to-noise (S/N) ratio of about 25 at
3 Å resolution. Much fainter objects can be observed at a lower S/N ratio,
and multiple-orbit observations can be combined. Data can also be
binned to lower spectral resolution if necessary. Observations may be
made during the day as well as during the nighttime portions of each 90-min
orbit. Numerous bright airglow emission lines of hydrogen, oxygen, and
nitrogen provide a high background signal during the day, however,
preventing observations of the faintest objects. This background is
much lower when the shuttle is in Earth's shadow.
Ultraviolet spectrophotometry requires that the instrument be carefully calibrated, which has been done with HUT (Davidsen et al. 1992). The instrumental efficiency function was determined by measurements made in the laboratory both before and after the flight, using light sources and detectors traceable to the National Institute of Standards and Technology. The function was independently determined in flight by comparing the observed spectrum of the hot white dwarf G191-B2B with a model-atmosphere calculation believed to provide an accurate prediction of the UV flux distribution of this star. The two methods agree to high precision, with a root-mean-square difference of only 6% over the 912-1850 Å range and a maximum deviation of 12%. Thus, absolute fluxes measured with HUT should generally be accurate to better than 10%, a substantial improvement over previous work in the far ultraviolet.