We report on the development of quasioptical Nb hot-electron bolometer mixers for heterodyne receivers operating at 1 THz 3 THz. The devices have submicron in-plane sizes, thus exploiting diffusion as the electron cooling mechanism. Quasioptical mixer circuits have been developed with planar double-dipole or twin-slot antennas. The measured (DSB) receiver noise temperatures are 1670 K at 1.1 THz, with an estimated mixer noise temperature of approximately equals 1060 K, and 2750 K at 2.5 THz, with an estimated mixer noise temperature of approximately equals 900 K. The IF bandwidth is found to scale as the length-squared, and bandwidths as high as 8 GHz have been measured. These results demonstrate the low-noise, broadband operation of the diffusion-cooled bolometer mixer over a wide range of far-infrared wavelengths.

We describe frontend concepts for the future heterodyne array instruments of the KOSMA 3-m telescope and for SOFIA. For KOSMA we are currently developing a dual frequency (400 - 500, 800 - 900 GHz) SIS mixer array of four elements per frequency band. For SOFIA, we are planning an up to 4 X 4 element array for 1.6 - 2.0 THz using superconducting hot-electron bolometers. The small number of pixels allows us to keep the optics relatively compact. For the same reason, a single sideband filter is not included. The local oscillator power will be distributed using Dammann gratings. Motivated by the excellent beam characteristics of waveguide horns we are planning to extend the range of our waveguide mixers to 2 THz. The mixers are based on the wideband tunerless mixers that have been successfully used in single element telescope receivers at KOSMA. The mixers will be standard building blocks mounted at the back of waveguide horns integrated into the optical setup. Local oscillators for 400 - 900 GHz are solid state sources, for the Terahertz array we are developing several alternative local oscillator concepts.

The Caltech Submillimeter observatory (CSO) is one of the World's premier submillimeter telescopes. It consists of a 10.4 meter diameter Leighton radio dish situated in compact dome near the summit of Mauna Kea, Hawaii. The telescope has been operating under a contract from the National Science Foundation on a regular basis since 1988. For the first time heterodyne Superconducting-Insulating-Superconducting (SIS) receivers with a 1 GHz intermediate frequency (IF) are available for the entire 180 - 950 GHz Submillimeter band. To enhance the extra-galactic capabilities of the observatory and to allow interferometry with the upcoming Submillimeter Array (SMA) project, we are actively working towards upgrading all heterodyne instruments with a 3 GHz IF bandwidth. Concurrent to the planned IF upgrade, we are constructing a dual polarization beam switching 345 GHz extra-galactic receiver, also with a 3 GHz IF bandwidth. Ideally, this instrument will give the CSO a factor of 8 improvement in integration time over the current 345 GHz receiver, and will be ideally suited for the study of highly red-shifted extra-galactic sources.

In this paper is given an overview of hot-electron superconducting mixers for submillimeter wave applications with emphasis on results obtained in Europe. The two competing types of mixers, the phonon cooled and the diffusion cooled, are described. A calculation scheme for the mixer conversion properties is presented based on data extracted from pumped and unpumped IV-characteristics. The generation of image frequency terms is discussed. Finally some state of the art results are presented.

One of the main obstacles encountered in designing low noise, high efficiency, heterodyne receivers and local oscillator sources at submillimeter wavelengths is the quality and cost of waveguide structures. At wavelengths shorter than 400 micrometers, rectangular waveguide structures, feed-horns, and backshorts become extremely difficult to fabricate using standard machining techniques. We have used a new laser milling technique to fabricate high quality, THz waveguide components and feedhorns. Once metallized, the structures have the properties of standard waveguide components. Unlike waveguide components made using silicon wet-etching techniques, laser-etched components can have almost any cross section, from rectangular to circular. Under computer control, the entire waveguide structure (including the corrugated feedhorn a submillimeter-wave mixer or multiplier can be fabricated to micrometer tolerances in a few hours. Laser etching permits the direct scaling of successful waveguide multiplier and mixer designs to THz frequencies. Since the entire process is computer controlled, the cost of fabricating submillimeter waveguide components is significantly reduced. With this new laser etching process, the construction of high performance waveguide array receivers at THz frequencies becomes tractable. In this paper we will describe the laser etching technique and discuss how it can be used to construct THz imaging arrays. We will also describe the construction of a prototype 810 GHz mixer which utilizes these new construction techniques.

SIS heterodyne mixer technology based on niobium tunnel junctions has now been pushed to frequencies over 1 THz, clearly demonstrating that the SIS junctions are capable of mixing at frequencies up to twice the energy gap frequency (4(Delta) /h). However, the performance degrades rapidly above the gap frequency of niobium (2(Delta) /h approximately equals 700 GHz) due to substantial ohmic losses in the on-chip tuning circuit. To solve this problem, the tuning circuit should be fabricated using a superconducting film with a larger energy gap, such as NbN; unfortunately, NbN films often have a substantial excess surface resistance in the submillimeter band. In contrast, the SIS mixer measurements we present in this paper indicate that the losses for NbTiN thin films can be quite low.

We are developing a wide band hybrid digital autocorrelator spectrometer suitable for the FIRST-HIFI instrument. At this time we have a prototype which is able to analyze four bands of 175 MHz each with a two bit three level digitizer clocked at 400 MHz. The correlation begins in a 64 channel Gallium Arsenide ASIC (Application Specific Integrated Circuit) clocked at 400 MHz and ends in a classical CMOS circuit at a lower frequency. The Gallium Arsenide technology presents several advantages for space applications, among which one can cite: (1) the power does not vary with the frequency of the signal analyzed, (2) the technology is very little sensitive to radiation. Our circuits are cascadable so that if several circuits are associated, we can build a versatile spectrometer with variable bandwidth and resolution. This presentation consists in a full description of this prototype spectrometer and a review of the current test results. We will also present a design of a spectrometer using the same technology and which will fit the needs of the heterodyne instrument of FIRST.

Very wideband heterodyne receiver systems are planned for many astronomical applications during the next 5 - 10 years. These extend up to 830 GHz for ground based telescopes and up to 2.7 THz for space and airborne applications. At present the only means to provide sufficient local oscillator power for these submillimeter receivers is the Schottky varactor diode frequency multiplier. Present multipliers operate at up to 1 THz with solid state pump sources, and are largely based on whisker contacted diodes. This paper describes a new generation of varactor multipliers which will be based on planar technology and may extend frequency coverage by a factor of two or more.

The Institute of Space Sensor Technology of the German Aerospace Center (DLR) is developing a heterodyne array receiver for the frequency range 2 to 6 THz for the Stratospheric Observatory for Infrared Astronomy (SOFIA). Key science issues in that frequency range are the observation of lines of atoms [e.g. (OI)], ions [e.g. (CII), (NII)], and molecules (e.g. OH, HD, CO) with high spectral resolution to study the dynamics and evolution of galactic and extragalactic objects. Long term goal is the development of an integrated array heterodyne receiver with superconducting hot electron bolometric (HEB) mixers and p-type Ge or Si lasers as local oscillators. The first generation receiver will be composed of HEB mixers in a 2 pixel 2 polarization array which will be pumped by a gas laser local oscillator. Improved Schottky diode mixers are the backup solution for the HEBs. The state of the art of HEB mixer and p-type Ge laser technology are described as well as possible improvements in the 'conventional' optically pumped far-infrared laser and Schottky diode mixer technology. Finally, the frequency coverage of the first generation heterodyne receiver for some important astronomical transitions is discussed. The expected sensitivity is compared to line fluxes measured by the ISO satellite.

In this paper we discuss the extension of Gaussian beam analysis to multi-beam quasi-optical systems. The normal approach taken in analyzing the propagation of multiple beams is to ray trace the optical axes of the individual beams and then superimpose the beam evolution of the on-axis beam. This is a rather inelegant and cumbersome process. We show how an efficient paraxial description of off-axis beams using Gaussian beam modes can be developed. The computational feasibility of the beam mode description clearly depends on finding an on-axis mode set in terms of which any off-axis beam can be approximated by a modal sum of modest size. We show how the optimum choice is made, and illustrate the power of the approach with an example.

This paper describes the measurement and subsequent data analysis of the signal beam of the receiver system of the Submillimeter Array (SMA). To measure the receiver beam patterns, a 2-axis planar scanning stage is mounted on top of the receiver assembly. Scanning in a plane orthogonal to the optical path, the near-field measuring system maps out both the amplitude and phase at frequencies of 242 and 265 GHz. By analyzing the measured patterns we can inspect the alignment of an individual receiver, and the optical assembly common to all frequency bands. The data also allows us to determine how the receiver couples to the beam waveguide that feeds the SMA antennas. The measured beam patterns of two different receivers, covering frequency bands of 176 - 256 and 250 - 350 GHz, are presented, as well as the analysis of these data. We believe that this is the first time such a rigorous full vectorial radio alignment technique has been applied to millimeter or submillimeter receiving systems for radio astronomy.

We report on the performance of power amplifiers as local oscillator drivers for millimeter and submillimeter-wave receivers. A MMIC power amplifier based on 0.1 micron GaAs HEMT technology on 50 micron thick substrate has been packaged in a waveguide block and characterized. Output power in excess of 100 mW is demonstrated over 88 - 94 GHz with more power easily achievable. The noise properties of the MMIC amplifier in multiplied local oscillator chains are characterized in a low noise superconductor-insulator-superconductor mixer based heterodyne receiver. A 386 GHz SIS mixer was used to characterize noise temperature in a laboratory environment. A more sensitive measurement of noise contribution from the amplifier was performed on a 278 GHz mixer/receiver at the Caltech Submillimeter-Wave Observatory, during astronomical observations. It is concluded that the MMIC amplifier does not add additional significant noise to the radiometer system.

The Millimeter Array, being planned by the National Radio Astronomy Observatory (NRAO) in conjunction with the radio astronomy community, will have almost complete coverage of the observable spectrum from 30 to 950 GHz. This paper discusses the choice of frequency bands and describes our present concept of the main receiver components for the MMA, including HFET (heterostructure field-effect transistor) amplifiers for the lower frequencies and sideband separating SIS (superconductor-insulator-superconductor) mixers for higher frequencies. Recent progress on HFET amplifiers has shown them to be competitive with SIS mixers for astronomical applications at frequencies up to approximately 100 GHz. The exact crossover frequency between HFET and SIS receivers in the context of the MMA is not yet determined. HFET receivers for the lower frequencies will be used not only for direct observations, but also for rapid calibration of instrumental and atmospheric phase in 'fast-switching' observing mode. We have already developed cryogenic HFET amplifiers for frequencies up to 110 GHz. Sideband-separating SIS mixers will improve the sensitivity in many spectral line measurements by greatly reducing the effect of atmospheric noise in the image sideband. We have developed an experimental 200 - 300 GHz sideband-separating receiver in which all the RF components -- quadrature hybrid, LO power splitter, LO couplers, and SIS junctions -- are on a single quartz chip. To reduce the effect of sideband noise from the LO source, and also substantially reduce the required LO power, we are developing a balanced SIS mixer, also on a single chip. We plan ultimately to develop a balanced, sideband-separating SIS mixer on a single chip. A first-IF amplifier stage inside the SIS mixer block is expected to eliminate the need for an IF isolator in an SIS receiver, thus permitting a 4 - 12 GHz IF.

Low-temperature-grown (LTG) GaAs offers the combination of sub-picosecond photocarrier lifetime and high breakdown electric field (greater than 5 X 10⁵ V/cm), and is grown in epitaxial films having excellent quality for microelectronic fabrication. A THz photoconductive mixer (photomixer) is formed on these films by patterning low- capacitance planar electrodes coupled to a coplanar antenna. The photomixer is conveniently pumped by two frequency-offset diode-laser beams focused on the exposed GaAs area between the electrodes. This paper summarizes the operational principles of the photomixer in contrast to a competing technique based on coherent three-wave photonic mixing. It then reviews different configurations of the photomixer as a laboratory tunable source for chemistry and metrology, and addresses some of the challenges in applying the photomixer as a local oscillator in portable spectroscopic and radiometric receivers.

At submillimeter wavelengths, broadband spectroscopy is currently possible only with a Fourier transform spectrometer (FTS). As a result, FTSes are quite useful for observations of objects in which spectral lines either cover a large frequency range, or where lines are broadened either by pressure or kinematics. Sources matching these descriptions include galaxies, hot, dense cores in interstellar molecular clouds, and planetary atmospheres. In the following, a tour of the classes of observations enabled by broadband spectroscopy is presented. As meaningful results call for attention to calibration, relevant calibration issues are discussed in the context of these observations.

The performance of Schottky barrier varactor multipliers is compared, with emphasis on generating power at frequencies above 1 THz. Measured efficiencies at 800 GHz are greater than 3 percent at and decrease to 0.9 percent at 1.13 THz and 0.25 percent at 1.4 THz. The power available from the penultimate multiplier in a chain imposes a limit on the output from the last stage. Further work is also required to overcome losses in multiplier structures and to understand better velocity and voltage saturation effects in varactors.

A 16-element SIS heterodyne array for operation in the 625 micrometer atmospheric window is under development at the MPIfR. The array consists of 2 X 8 elements with closest feasible spacing of the pixels on the sky ((root)2 (DOT) (Theta) _mb). The L.O. tuning range covers the astronomically important CI and the CO(4-3) transitions, and an IF bandwidth of 2 GHz (1200 kms^-1) will permit mapping of extragalactic systems. For best system sensitivity the design allows for cold optics ( 15K) and single-sideband operation. The frontend will be linked to a flexible autocorrelator, with a maximum bandwidth of 2 GHz (2048 channels) for each of the 16 modules. In the high-resolution mode, 500 MHz of bandwidth can be operated with 8192 channels of 61 kHz spectral resolution. System components are currently undergoing final integration and critical evaluation in our laboratories. First astronomical commissioning is scheduled for later this year. The sensitivity expected with CHAMP, for e.g. carbon studies, will be unparalleled. With the full array in SSB operation the mapping speed will be enhanced by a factor of 50 - 100 compared to current single-pixel detectors.

The National Radio Astronomy Observatory is engaged in two major new telescope projects. Each requires technological advances to meet the scientific requirements. The Green Bank Telescope is a fully steerable 100 m diameter filled aperture telescope for operation over the frequency range 300 MHz to 100 GHz. The Millimeter Array is a forty element interferometer of 8 m diameter antennas, with a goal of using 10 m or even 12 m antennas, for operation over the frequency range 30 - 850 GHz. The technical features being used to achieve the performance specifications are presented together with examples of the science that calls for these specifications. The current status of each project is also discussed.

The main features of recent antenna designs are reviewed. Particular reference is made to the challenges presented by the large aperture-synthesis array projects now under development. In the final section some specific problems concerned with holographic antenna measurements are discussed.

The 'Large Millimeter Telescope' (LMT), or 'Gran Telescopio Milimetrico' (GTM), is a joint USA/Mexico Project and will be the worlds's largest short millimeter-wavelength ((lambda) equals 1 to 3.4 mm) radio telescope. The large collecting area, equivalent to a circular aperture of 50 m diameter, and the high altitude (4600 m) site will allow the LMT/GTM to achieve unprecedented astronomical performance. The telescope will be equipped with state-of-the-art heterodyne ((lambda) equals 3 mm) and bolometric ((lambda) equals 1 mm) focal plane arrays which will make it a powerful high angular resolution mapping instrument. Here, we present the current status of the LMT/GTM Project and its primary astronomical performance specifications. We then describe the optical design of the telescope, including some important aspects related to the wobbling subreflector and to the reimaging optics. We also analyze some of the electromagnetic characteristics of the alternative telescope configurations, enclosed or open-air, that were considered before the final selection of an open-air telescope took place.

The Heinrich Hertz Telescope is a radio telescope dedicated to the observation of submillimeter wavelength radiation from celestial sources. It is a Cassegrain telescope with a diameter of 10 m and a reflector accuracy of about 17 micrometer, yielding an excellent performance at 350 micrometer, the shortest wavelength transmitted through the atmosphere. The reflector panels and the backup structure employ carbon-fiber reinforced plastic as basic material to achieve a lightweight, stiff construction with a very small coefficient of thermal expansion. This enables us to maintain full performance of the telescope in day time under solar illumination of the structure. In this paper, we describe the structural and material characteristics of the telescope. We also describe the holographic method which enables a measurement and setting of the reflector panels to an accuracy of 10 micrometer. The telescope is located on Mt. Graham in Eastern Arizona at an altitude of 3250 m, providing good submillimeter observing conditions, especially in the winter months. This is a collaborative effort of the Max-Planck- Institut fur Radioastronomie, Bonn, Germany and Steward Observatory, University of Arizona, Tucson, AZ.

The design and current construction status of the Submillimeter Array is discussed. The array will consist of eight 6-m elements that are reconfigurable to achieve baselines from 8 to 508 m. The array will be located on Mauna Kea in Millimeter Valley, at an elevation of 4,080 m, and will cover all bands from 180 to 900 GHz. Provisions are being made to include the CSO and JCMT telescopes in the array, which will substantially increase its capability. Most of the development and assembly is underway at the test site in Westford, Massachusetts and in Taiwan. Holography and radiometry of planets at 3.3 mm wavelength have been carried out on the prototype antenna of the array. It is expected to reach its performance specifications, which include 12 micrometer surface accuracy and 1' pointing. The prototype correlator is complete and board replication is underway. Most of the civil work at the site is complete. Interferometric tests of two antennas in Westford are expected to begin in mid-1998, and similar tests on Mauna Kea are expected to begin in 1999.

The LMT is a 50-meter-diameter fully-steerable radio telescope to be used for astronomy at millimeter wavelengths. The specifications call for a surface accuracy of 70 micrometer RMS and a pointing accuracy of 0.7 arc sec RMS. This paper addresses two critical issues of the design development. To achieve the required accuracy it was proposed that the primary reflector of the LMT utilize an actively controlled segmented surface with a real time closed loop control system. In the approach developed for the preliminary design systematic deformations of the segments result in the magnification and propagation of surface errors; this problem is examined and possible solutions presented. Limits on the applicability of homology to the design of millimeter wave radio telescopes are discussed in the context of the need for an active surface. A proposed approach for computing in real time the pointing errors due to thermal deformations is evaluated. The results show the limits of applicability of this approach.

The Millimeter Array (MMA) is a synthesis imaging telescope designed to provide sensitive, high-precision astronomical imaging at sub-arcsecond, 0'.1, resolution at millimeter and submillimeter wavelengths. In order to achieve these goals it is necessary to locate the Millimeter Array on an extremely dry site so as to minimize the atmospheric background emission; the receiver noise temperatures should be near the limit imposed by photon noise and the antenna warm spillover contribution should be minimized. It is possible to quantify each of these requirements, and the interactions among them, so as to achieve the maximum sensitivity for the Millimeter Array. Together, the requirements lead to the technical design goals for the Millimeter Array.

The LSA (Large Southern Array) is a European proposal for a large millimeter/submillimeter interferometer. The current concept is that of a joint U.S.-European project consisting in a 7000 m² array made of 64 12-m antennas operating between 70 and 900 GHz. Baselines length up to 10 km, providing 0.01' resolution, are foreseen. The LSA concept and operating context has evolved quickly in the past 3 years. I describe here part of this evolution, with some emphasis on the particular points in which the European approach differs from the NRAO approach on the MMA.

The Large Millimeter and Submillimeter Array (LMSA) being planned at the National Astronomical Observatory of Japan (NAOJ) is a high priority plan for the future large-scale ground-based astronomical facility coming after the Subaru telescope. The LMSA will consist of 50 10-m antennas. The planning of the instrument places a special emphasis on high- resolution and high-sensitivity observations at submillimeter wavelengths. Efforts at Nobeyama Radio Observatory (NRO) have been focused in the site testing at the two candidate sites in Northern Chile and engineering developments, especially in antenna design, submillimeter mixers, high-speed digital technology for wide-band correlator, and phase calibration methods. A possible joint-array operations with other large array projects is under discussion.

The Caltech Submillimeter Observatory and the James Clerk Maxwell Telescope have been combined to form the only astronomical interferometer currently operating at submillimeter wavelengths. The telescopes have been operating in this mode for one or two dedicated periods in each of the last 5 years. Results with sub-arcsecond resolution have been obtained at 230, 345 and 460 GHz. The interferometer differs in many ways from the existing millimeter-wave arrays. Connecting two independent telescopes of different design introduces extra problems not encountered with homogeneous arrays of antennas. The CSO-JCMT system is described, with an emphasis on these incompatibility issues and solutions that were adopted. Analysis of data from a single, fixed baseline requires direct modeling of the measured visibilities rather than a synthesized image, an approach that has since proved invaluable for analyzing data from other arrays as well. The sensitivity and angular resolution of the interferometer are limited by fluctuations in the refractive index due to water vapor in the Earth's atmosphere. Two water vapor radiometers have been designed, built and installed to monitor the fluctuations in each beam and generate a correction to the visibility phase measured by the interferometer. These radiometers are described and recent results are presented.

The development of metrology instrumentation needed for realization of the active surface and precision pointing for the Green Bank Telescope (GBT) presents the opportunity for a complete characterization of this complex structure prior to the first astronomical observations. It is our intention to use the metrology program to evaluate the completed telescope and to point the way to improvements in performances. We also anticipate using the instrumentation for early detection of possible developing faults in the structure and the foundation. For the first time, it should be possible to derive parameters that determine the repeatable pointing variations as a function of elevation and azimuth position prior to astronomical observations. The measurement program will also give an estimate of the non-repeatable pointing errors resulting from thermal gradients within the structure. Laser rangefinders surrounding the telescope permit the non- invasive measurement of thermal and gravitationally induced deformations within the structure at the 100 micron accuracy level. These measurements, and others, will be used to confirm the finite element analysis of the structure and to predict the performance of the completed telescope. This paper outlines the instrumentation used in the measurement program and gives the results obtained to date. These include dominant modal resonances in the structure, along with the damping coefficients associated with these resonances.

One of the central issues in astronomy is the formation and evolution of galaxies at large redshifts. Submillimeter observations are essential to understanding these processes. One of the best prospects for high redshift submillimeter observations is the study of the C_II 158 micrometer fine- structure line, which carries about 0.2% of the total luminosity of nearby starburst galaxies. However, current heterodyne receivers at submillimeter observatories have insufficient bandwidth to detect the full extent of highly broadened emission lines. We are developing a broadband grating spectrometer for the Caltech Submillimeter Observatory with a total bandwidth of approximately 3400 km/s and a velocity resolution of 200 km/s. The detectors will be a linear array of 32 close-packed monolithic silicon bolometers developed at NASA's Goddard Space Flight Center. In order to achieve background-limited sensitivity, the bolometers will be cooled to 100 mK by an adiabatic demagnetization refrigerator. The spectrometer optics will consist of a tunable cryogenic immersion grating using broadband filters as order sorters. The spectrometer will be optimized to operate in the 350 micrometer and 450 micrometer atmospheric windows. Calculations of the sensitivity of the spectrometer reveal that an ultraluminous infrared galaxy of 10¹² Lqq at a redshift of z equals 1 would be detectable at the 3(sigma) level in the C_II line with 20 minutes of integration time.

The University of Chicago polarimeter, Hertz, is designed for observations at the Caltech Submillimeter Observatory in the 350 micrometer atmospheric window. Initial observations with this instrument, the first array polarimeter for submillimeter observations, have produced over 700 measurements at 3(sigma) or better. This paper summarizes the characteristics of the instrument, presents examples of its performance including polarization maps of molecular clouds and regions near the Galactic center, and outlines the opportunities for improvements with emphasis on requirements for mapping widely extended sources.

We are developing arrays of bolometers based on silicon nitride micromesh absorbers for the Spectral & Photometric Imaging Receiver (SPIRE) on the Far Infra-Red and Submillimeter Space Telescope (FIRST). The bolometers are coupled to a close-packed array of 1 f(lambda) feedhorns which views the primary mirror through a cooled aperture stop. Feedhorn-coupled bolometers minimize the detector area and throughput and have good optical efficiency. A 1 f(lambda) feedhorn array provides, higher mapping speed than a 2 f(lambda) feedhorn array and reduces the number of jitters required to produce a fully sampled map, but at the cost of more detectors. Individual silicon nitride micromesh bolometers are already able to meet the performance requirements of SPIRE. In parallel we are developing transition-edge detectors read out by SQUID current amplifier. The relatively large cooling power available at 300 mK enables the array to be coupled to a cold SQUID multiplexer, creating a monolithic fully multiplexed array and making large format arrays possible for SPIRE.

The Submillimeter Common-User Bolometer Array (SCUBA) is one of a new generation of cameras designed to operate in the submillimeter waveband. The instrument has a wide wavelength range covering all the atmospheric transmission windows between 300 and 2000 micrometer. In the heart of the instrument are two arrays of bolometers optimized for the short (350/450 micrometer) and long (750/850 micrometer) wavelength ends of the submillimeter spectrum. The two arrays can be used simultaneously, giving a unique dual-wavelength capability, and have a 2.3 arc-minute field of view on the sky. Background-limited performance is achieved by cooling the arrays to below 100 mK. SCUBA has now been in active service for over a year, and has already made substantial breakthroughs in many areas of astronomy. In this paper we present an overview of the performance of SCUBA during the commissioning phase on the James Clerk Maxwell Telescope (JCMT).

Continuum radiometers based on bolometers have a long tradition at the Max-Planck-Institut fur Radioastronomie (MPIfR) in Bonn, Germany. Arrays of bolometers have been under development since the early 90s. A small 7-element system, operating at 300 mK, saw first light in 1992 at the IRAM 30 m- telescope and has been used successfully by numerous observers at that facility since then. While this array had a conventional 'composite' design, it was obvious that larger arrays, especially for higher frequencies, could take advantage of microfabrication technology. The recent MPIfR bolometer arrays employ a hybrid approach. They combine a single-mode horn array with a planar bolometer array on a single crystal Silicon wafer with Silicon-Nitride membranes. With efficient absorbing structures, the bolometers couple to the single mode of the radiation field collected by the horns, without needing integrating cavities. Readout is provided by NTD-Germanium thermistors that are attached to the absorbers. This paper covers the history of this development, the general aspects of the bolometer arrays, including the coupling to the telescope, and the status of work in progress.

We describe the preliminary design of the proposed Heterodyne Instrument for FIRST (HIFI). The instrument will have a continuous frequency coverage over the range from 480 to 1250 GHz in five bands, while a sixth band will provide coverage for 1410 - 1910 GHz and 2400 - 2700 GHz. The first five bands will use SIS mixers and varactor frequency multipliers while in the sixth band a laser photomixer local oscillator will pump HEB mixers. HIFI will have an instantaneous bandwidth of 4 GHz, analyzed in parallel by two types of spectrometers: a pair of wide-band spectrometers (WBS), and a pair of high- resolution spectrometer (HRS). The wide-band spectrometer will use acousto-optic technology with a frequency resolution of 1 MHz and a bandwidth of 4 GHz for each of the two polarizations. The HRS will provide two combinations of bandwidth and resolution: 1 GHz bandwidth at 200 kHz resolution, and at least 500 MHz at 100 kHz resolution. The HRS will be divided into 4 or 5 sub-bands, each of which can be placed anywhere within the full 4 GHz IF band. The instrument will be able to perform rapid and complete spectral line surveys with resolving powers from 10³ up to 10⁷ (300 - 0.03 km/s) and deep line observations.

Large bandwidth acousto-optical spectrometers have now reached a very high level of maturity. They achieve very compatible results in comparison with other spectrometer types like filterbanks, autocorrelators, and chirp transform spectrometers. In addition, AOS are rather simple in design, have little complexity and can be designed for space applications very easily. A new generation of broad-band AOS, the array-AOS, consists of four parallel 1 GHz spectrometers built into one optical unit. Tests results in the laboratory as well at a radio-observatory are very promising. For example, the Allan variance minimum time has been found above 1000 seconds. In comparison it can be shown that the AOS spectra are less affected by instrumental noise or baseline distortions due to platforming effects as they are visible with most hybrid autocorrelators. For future applications of acousto-optics the development of cross-correlators seems to be feasible. First steps in this new direction are on the way.

We have developed a novel wideband spectrometer for astronomical heterodyne spectroscopy. The spectrometer, WASP, has 3250 MHz bandwidth and 33 MHz resolution, a combination well matched to submillimeter spectroscopy of high-redshift objects, interacting galaxies, active galactic nuclei, and planetary atmospheres. The spectrometer is an autocorrelation spectrometer with analog microwave integrated circuit multipliers separated by microstripline transmission line delays. Our prototype spectrometer is compact, requires little power (75 W), and integrates stably for many hours.

This paper describes a 250 MHz cross-correlator for radio astronomy built using field-programmable gate arrays (FPGA's). Experimental results indicate that CMOS FPGA's can operate reliably at 250 MHz. Long-term integration tests of up to 10¹² samples have been taken, demonstrating the correlator accuracy down to the limits of the testing apparatus. Recent estimates indicate that higher-performance FPGA's available early in 1998 can attain speeds of over 300 MHz using 20% fewer logic elements than previous designs. The FPGA correlator can be reconfigured for 2, 3, 4, or even 9 levels of sensitivity, permitting a tradeoff of the number of lags with the sensitivity desired. Although FPGA's have historically been more expensive than custom chips, starting this year FPGA correlators are actually less expensive than custom correlator chips. This cost reduction is due to the fact that FPGA's are standard components in the electronics industry and are produced in much higher volumes than custom correlators. As an example, a 1024-lag 4-level cross- correlator can be produced for an FPG chip cost of less than $80 in 1998. FPGA's continue to drop in price faster than custom devices because volumes continue to increase and FPGA architectures continue to improve. The demonstrated high performance of FPGA's coupled with their extreme flexibility, reconfigurability and lower cost make FPGA's ideally suited for the next generation of large correlator arrays.

The European Space Agency's FIRST satellite will include a submillimeter direct detection instrument using bolometric detectors. The scientific drivers for the instrument are discussed and the essential features of the proposed instrument, SPIRE, are described. SPIRE (Spectral and Photometric Imaging REceiver) comprises a three-band imaging photometer covering the 250 - 500-micrometer range and an imaging Fourier Transform Spectrometer covering wavelengths between 200 and 670 micrometer. The SPIRE detectors are 300-mK bolometer arrays providing full sampling of the diffraction spot. The imaging photometer is optimized for deep photometric surveys in the submillimeter (one of the main science goals of the FIRST mission), for which application it will be much more sensitive than any Earth-based facility. The nominal bands for the photometer are 250, 350, and 500 micrometer with a spectral resolution of 3. Three detector arrays observe the same 4-arcminute field of view simultaneously, with dichroic beam dividers separating the bands. The imaging spectrometer has a 2-arcminute field of view and an adjustable spectral resolution of 0.04 - 2 cm^-1 ((lambda) /(Delta) (lambda) equals 20 - 1000 at 250 micrometer).

The Degree Angular Scale Interferometer (DASI) is a compact cm-wave interferometer designed to image anisotropy in the Cosmic Microwave Background (CMB) and to measure its angular power spectrum. The power spectrum will be densely sampled over the l range 160 to 710, corresponding to angular scales of 0.25 to 1.15 degrees. DASI consists of 13 elements. Each element consists of a 20-cm diameter lensed corrugated horn followed by a cooled low-noise HEMT amplifier operating from 26 - 36 GHz. All elements are mounted on a single alt-az mount, which fixes the projected baselines and obviates an IF tracking delay. The mount also includes rotation of the aperture plane along the line of sight to improve the u, v coverage and the control of instrumental systematics. The 10 GHz IF bandwidth will be correlated in 1 GHz bands to provide spectral index information. The instrument is scheduled to be completed in Summer 1998. After extensive testing it will be deployed at the South Pole for year-round operation starting in November 1999.

A description of modern wideband digital correlators being developed at the National Radio Astronomy Observatory (NRAO) for radio astronomy is presented. The NRAO is developing several modern wideband instruments for spectroscopic analysis in radio astronomy. One instrument now in the system test phase is a spectrometer intended for use on the 100 meter Green Bank Telescope (GBT) being constructed by the NRAO. This spectrometer has a maximum bandwidth of 6.4 GHz comprised of 8 separate 800-MHz-wide segments and can develop 16,384 points of spectral resolution in its wideband mode. Factor of 2 trade-offs between bandwidth and resolution can be made. In an alternate narrow bandwidth mode, the GBT spectrometer can develop spectra with 262,144 point resolution over a maximum bandwidth of 1600 MHz comprised of 32 separate 50-MHz-wide segments, again with bandwidth/resolution trade-offs possible. Additional spectrometers being built or planned using hardware developed for the GBT system include a spectrometer for the NRAO 12 Meter Telescope in Tucson and a test correlator for the proposed Millimeter Array (MMA). A design for the final MMA correlator is also being studied at the NRAO. The MMA is projected to be an aperture synthesis array consisting of about 40 radio telescopes (the exact array size is currently uncertain). The correlator proposed for the MMA will be able to analyze the auto- and cross-spectra of the array output with a bandwidth of up to 16 GHz per antenna.

We present results from atmospheric model calculations for the design of an atmospheric phase correlation radiometer for the Berkeley-Illinois-Maryland Millimeter Array (BIMA). The radiometer will monitor the atmospheric path delay by observing the fluctuations in the emission from tropospheric water vapor which causes de-correlation of astronomical signals which are observed along different lines of sight. We discuss the applicability of monitoring the optically thick 183 GHz water line and the optically thin 22 GHz water line. We conclude that for the BIMA site, which is not extremely dry, optical depth effects make observations of the 183 GHz line unfavorable. We discuss possible observing schemes and conclude that a multi-channel radiometer provides the highest achievable accuracy, as it provides the possibility to fit a line shape model to the observed channels. Systematic errors due to the unknown altitude distribution of water vapor and optical depth effects can be significantly reduced by this scheme, compared to methods which only monitor the change of the peak brightness temperature of the water vapor line.

We present results from phase correction efforts at the Owens Valley Radio Observatory millimeter array (OVRO). A brief description of the theory of phase correction is followed by a description of the water line monitors (WLMs) constructed and placed on each of the six antennas of the array. A summary of the current software in place is also included. We present examples of data corrected using this technique and the first image created using radiometric phase correction at OVRO. The phase correction system is undergoing further development and will soon be made available for general observing at the array. A brief discussion of application of the technique for future arrays (e.g. MMA, LSA, etc.) is included as a conclusion to this contribution.

The Millimeter Array (MMA) will consist of 40 or more antennas operating as a correlator array in the frequency range 30 to 950 GHz. This paper is concerned with the transmission and processing of the signals received at the antennas from the outputs of the low-noise receiver front ends to the inputs of the correlator. It includes the IF and LO systems and the optical fiber transmission system that interconnects the antennas and the Electronics Building. There the signals are filtered for bandwidth selection, digitized, delayed as necessary to compensate for the varying path lengths to the antennas, and cross-correlated. A number of decisions must be made before detailed design can be started. These include the maximum bandwidth of the signals to be processed, the use of analog or digital transmission from the antennas, and the bandwidths required for spectral line observations. Detailed considerations include control of signal levels at certain points, implementation of a calibration scheme to provide a measure of the system gain, control of fringe frequencies, and separation of sidebands for the frequency bands using SIS- mixer input stages.

A number of possible antenna designs have been considered for the Millimeter Array. This paper presents two designs which were studied in detail but are no longer being considered, as well as three of the designs currently being studied. The major performance goals for the antennas are listed. Two of the most challenging performance requirements, getting sufficient pointing accuracy and achieving fast position switching, and some of the possible approaches to meeting these challenges are discussed.

The design of the Leighton telescopes and the unique techniques used in their fabrication make these telescopes particularly amenable to precise modeling and measurement of their performance. The surface is essentially a continuous membrane supported at 99 uniformly distributed nodes by a pin joint triangular grid space frame. This structure can be accurately modeled and the surface can be adjusted using low- resolution maps. Holographic measurements of the surface figure of these telescopes at the Caltech Submillimeter Observatory (CSO) and the Owens Valley Radio Observatory (OVRO) have been made over several epochs with a repeatability of 5 - 10 micrometer over the zenith angle range from 15 to 75 degrees. The measurements are consistent with the calculated gravitational distortions. Several different surface setting strategies are evaluated and the 'second order deviation from homology,' H_d, is introduced as a measure of the gravitational degradation that can be expected for an optimally adjusted surface. H_d is defined as half of the RMS difference between the deviations from homology for the telescope pointed at the extremes of its intended sky coverage range. This parameter can be used to compare the expected performance of many different types of telescopes, including off-axis reflectors and slant-axis or polar mounts as well as standard alt-az designs. Subtle asymmetries in a telescope's structure are shown to dramatically affect its performance. The RMS surface error of the Leighton telescope is improved by more than a factor of two when optimized over the positive zenith angle quadrant compared to optimization over the negative quadrant. A global surface optimization algorithm is developed to take advantage of the long term stability and understanding of the Leighton telescopes. It significantly improves the operational performance of the telescope over that obtained using a simple 'rigging angle' adjustment. The surface errors for the CSO are now less than 22 micrometer RMS over most of the zenith angle range and the aperture efficiency at 810 GHz exceeds 33%. This illustrates the usefulness of the global surface optimization procedure.

At millimeter and submillimeter wavelengths, pressure broadened molecular spectral lines make the atmosphere a natural limitation to the sensitivity and resolution of astronomical observations. Tropospheric water vapor is the principal culprit. The translucent atmosphere both decreases the signal, by attenuating incoming radiation, and increases the noise, by radiating thermally. Furthermore, inhomogeneities in the water vapor distribution cause variations in the electrical path length through the atmosphere. These variations result in phase errors that degrade the sensitivity and resolution of images made with both interferometers and filled aperture telescopes. To evaluate possible sites for the Millimeter Array, NRAO has carried out an extensive testing campaign. At a candidate site at 5000 m altitude near Cerro Chajnantor in northern Chile, we deployed an autonomous suite of instruments in 1995 April. These include a 225 GHz tipping radiometer that measures atmospheric transparency and temporal emission fluctuations and a 12 GHz interferometer that measures atmospheric phase fluctuations. A sub-millimeter tipping photometer to measure the atmospheric transparency at 350 micrometer wavelength and a submillimeter Fourier transform spectrometer have recently been added. Similar instruments have been deployed at other sites, notably Mauna Kea, Hawaii, and the South Pole, by NRAO and other groups. These measurements indicate Chajnantor is an excellent site for millimeter and submillimeter wavelength astronomy. The 225 GHz transparency is better than on Mauna Kea. The median 225 GHz transparencies measured at Chajnantor and at the South Pole are comparable.

A 10 meter diameter submillimeter-wave telescope has been proposed for installation and scientific use at the NSF Amundsen-Scott South Pole Station. Current evidence indicates that the South Pole is the best submillimeter-wave telescope site among all existing or proposed ground-based observatories. Proposed scientific programs place stringent requirements on the optical quality of the telescope design. In particular, reduction of the thermal background and offsets requires an off-axis, unblocked aperture, and the large field of view needed for survey observations requires shaped optics. This mix of design elements is well-suited for large-scale (square degree) mapping of line and continuum radiation from submillimeter-wave sources at moderate spatial resolutions (4 to 60 arcsecond beam size) and high sensitivity (milliJansky flux density levels). The telescope will make arcminute angular scale, high frequency Cosmic Microwave Background measurements from the best possible ground-based site, using an aperture which is larger than is currently possible on orbital or airborne platforms. The telescope design is homologous. Gravitational changes in pointing and focal length will be accommodated by active repositioning of the secondary mirror. The secondary support, consisting of a large, enclosed beam, permits mounting of either a standard set of Gregorian optics, or prime focus instrumentation packages for CMBR studies. A tertiary chopper is located at the exit pupil of the instrument. An optical design with a hyperboloidal primary mirror and a concave secondary mirror provides a flat focal surface. The relatively large classical aberrations present in such an optical arrangement can be small compared to diffraction at submillimeter wavelengths. Effective use of this telescope will require development of large (1000 element) arrays of submillimeter detectors which are background-limited when illuminated by antenna temperatures near 50 K.

We built fixed-tuned SIS mixers for use between 70 and 115 GHz on the Berkeley-Illinois-Maryland Association (BIMA) array. The mixers are similar to 215 GHz SIS devices designed by R. Blundell et al. Our 1.4 X 1.4 micrometer Nb/Al-Al₂O₃/Nb junctions have extremely sharp I-V characteristics, with R_sg/R_n approximately 40, and (omega) RC approximately 2. Heterodyne tests were made with a temperature-regulated vane mounted in the input waveguide to the mixer block. With vane temperatures of 20 and 35 K, we measure double sideband receiver temperatures T_rcvr of 17 K from 80 to 115 GHz. Some gain compression occurs for input load temperatures greater than 50 K. Hence, with vane temperatures of 85 and 280 K we measure T_rcvr approximately 35 K. Similar noise temperatures are measured with 77 and 295 K loads mounted outside the dewar. Because the mixers are partially saturated, we bias the junctions approximately 0.1 mV above or below the gain maximum, where the receiver response is a nearly linear function of the input signal level. This yields T_rcvr of 35 - 40 K and makes it possible to calibrate the receivers on the telescopes using the standard chopper wheel method. With no LO power applied to the mixer, we observe a small IF signal which is proportional to the square of the input load temperature; we believe this is due to self-mixing of blackbody photons in the junction.

Acousto-optical techniques are particularly well suited for wide band spectroscopy in astrophysical and geophysical applications using millimeter and sub-millimeter heterodyne radiometers aboard spacecraft. An illustration is given by the 1 GHz AOS built for the ODIN, sub-millimeter telescope in Earth's orbit. We describe the spectrometer, including the IF down converter, AO processor, data acquisition and pre- processing sub-systems. The used technology, space qualification of the critical components as well as relevant environmental and electrical results are presented. The acousto-optical technique can also be used for compact, ultra wide band spectrometers (2000 to 4000 channels over 4 GHz bandwidth or more). Scientific applications include sub- millimeter missions in Astronomy (FIRST) as well in Geophysics (Microwave Limb Sounding programs). Performances obtained on prototypes built in Meudon Observatory are presented and the characteristics achievable for near future systems are analyzed. They fit well with the presently retained design of the heterodyne (HET) instrument on FIRST, based on a font-end system made of a four band, dual polarization receiver, allowing for full frequency coverage from 480 to 1250 MHz.

Far prior to any other wavelength domains, decameter radioastronomy has become a challenging task because of man made interference due to the increasing needs in telecommunications. However most of the astrophysical studies in this frequency band, -- particularly those of non thermal radiations from magnetized objects --, require high gain multi-polarization antennas and the use of high sensitivity, high dynamic range spectroscopy techniques. New generation of wide band spectrum analyzers, well suited to overcome these constraints, are presently developed in Meudon observatory in collaboration with the Space Research Institute (Graz, Austria). They are based on the utilization of newly available, high performance, programmable digital circuits for signal processing, arranged in dedicated, parallel architecture, which can directly compute the power spectrum of the input signal. As an example, we describe here a specialized real-time analyzer for studying polarized decameter emissions from Jupiter and the Sun. With this device, the full spectral analysis (four Stokes parameters) of a 10 MHz bandwidth can be performed at the millisecond time scale, over 1024 channels and within a 70 dB dynamic range. The unprecedented capabilities of this analyzer were evaluated during the last Jovian opposition, by using the Decameter Array in Nancay, France. Some examples are presented, with emphasis on results on the shortest fine structures, which are characteristics of cyclotron maser radiations from planets. Perspectives for extending to shorter wavelengths (namely decimeter or centimeter ones) and for enhancing capabilities by implementing dedicated softwares are discussed. The offered possibility of processing the telescope signal in real time -- in order, for instance, to manage the man made radio interference problem --, indeed appears as a key to maintain acceptable sensitivity in any future, large radio telescope system operating on the ground.

The South Pole has been identified as an excellent submillimeter site. Submillimeter continuum work at the South Pole promises to address several essential issues in astronomy including the morphology of the magnetic field in the galactic center and the search for protogalaxies. Here we will discuss the Submillimeter Polarimeter for Antarctic Remote Observing (SPARO), the first sub-kelvin cryostat designed for operation in the hostile Antarctic winter environment. We have implemented several novel design concepts, including a vapor cooled ⁴He cryostat and a capillary fed pumped ⁴He stage, in order to construct a long holdtime system with simplified operation.

The Submillimeter Common-User Bolometer Array (SCUBA) is a new continuum camera operating on the James Clerk Maxwell Telescope (JCMT) on Mauna Kea, Hawaii. It consists of two arrays of bolometric detectors; a 91 pixel 350/450 micron array and a 37 pixel 750/850 micron array. Both arrays can be used simultaneously and have a field-of-view of approximately 2.4 arcminutes in diameter on the sky. Ideally, performance should be limited solely by the photon noise from the sky background at all wavelengths of operation. However, observations at submillimeter wavelengths are hampered by 'sky-noise' which is caused by spatial and temporal fluctuations in the emissivity of the atmosphere above the telescope. These variations occur in atmospheric cells that are larger than the array diameter, and so it is expected that the resultant noise will be correlated across the array and, possibly, at different wavelengths. In this paper, we describe our initial investigations into the presence of sky-noise for all the SCUBA observing modes, and explain our current technique for removing it from the data.

State-of-the art frequency multipliers, particularly those that operate in the submillimeter wave band, suffer from several limiting factors that make them impractical for modern array applications. In this paper we review the unique requirements placed upon frequency multipliers by array systems, report on the progress of our multiplier development effort, and outline our future development goals toward a suite of multipliers for the NRAO Millimeter Array.

The design of a dual polarizing bolometer detector system for use with a polarizing Fourier transform spectrometer to conduct broadband astronomical spectroscopy at submillimeter wavelengths is presented. While inclement weather precluded astronomical observations during the commissioning run, the initial performance of the system, as determined from observing laboratory calibration sources, is presented, and the implications for astronomical spectroscopy discussed.

Development of two types of submillimeter-wave superconducting direct detectors are described. One is the SIS video detector and the other is the SIS photon detector. Using 500 GHz parallel connected twin junction device, video detector performance is eluted. High quantum efficiency in wide frequency range in submillimeter-wave and low noise performance is achieved. Noise equivalent power of 6 X 10^-14 W/(root)Hz is measured at operating temperature of 1.5 K and bias voltage of 0.5 mV. Further improvements using distributed junction array is addressed. Large area niobium SIS junction (100 micrometer X 50 micrometer X 2) is used as a photon detection device and their thermal, electrical and optical performance is evaluated. Leakage current decrease more than three orders of magnitude at 0.5 K compared to that at 4.2 K. Noise of the device is limited by the shot noise.

We report on the status of SIS mixers being constructed at the Institute of Astronomy and Astrophysics, Academia Sinica. These mixers will later be integrated into the receiver systems used on the ASIAA supplied antennas for the SMA project. We have reproduced the SAO design for these mixers at the 200 GHz band. We achieve acceptable noise performance across the tuning band of the mixers.

The use of SIS (Superconductor-Insulator-Superconductor) devices as the mixing element, in ultra-sensitive heterodyne receivers for use in radio astronomy is well established. In attempting to exploit the last main submillimeter atmospheric window we are developing a Pb-alloy SIS mixer to work in the frequency band 800 - 900 GHz. In this paper we address some of the problems with working at frequencies above the energy gap of a superconducting alloy. We also present initial results obtained from Pb-alloy SIS junctions fabricated at the University of Kent at Canterbury (UKC).

This article covers the manufacturing of aluminum reflector panels for submillimeter radio astronomy. The first part involves the general construction and application of a machine custom designed and built to do this. The second is a discussion of the software and execution of method to actually produce the reflectors for the Smithsonian Astrophysical Observatories Submillimeter Array (SMA). The reflective surface of each panel is contoured both radially and circularly by oscillating a platen supporting the panel about a fixed axis relative to a tool which is fixed during platen oscillation. The tool is repositionable between oscillations along an x axis to achieve the radial contour and along a z axis to achieve the desired parabolic or spherical contour. Contrary to the normal contouring of such a surface with a 5- axis CNC machine, tool positioning along either axis is independent of tool location along the other axis, simplifying the machine structure as well as its computerized operation. A unique hinge is provided to restrain the platen in a radial direction while allowing floating action of the platen on an air cushion during its oscillation. These techniques and the equipment are documented in U.S. Patent No. 5477602.

The Green Bank Telescope, now under construction at the National Radio Astronomy Observatory in Green Bank, WV, will be the world's largest fully-steerable telescope upon its completion in 1999. This article describes the general features of the telescope including its site, optics, surface, expected performance, and some of the areas of astronomy to which it will contribute.

The LMT is a 50 m diameter telescope for operation to a wavelength of 1 mm from a high mountain site in central Mexico at a latitude of 19 degrees. The telescope is designed to address fundamental questions about the origin and formation of galaxies, clusters of galaxies as well as stars and planets. It is a joint project of the Mexican Instituto Nacional de Astrofisica, Optica y Electronica (INAOE) in Tonantzintla, Puebla and the University of Massachusetts at Amherst. The telescope is in the Critical Design phase and completion of the basic assembly at the site is scheduled for the end of 2000. We describe historical and organizational aspects of the Project and present the major technical specifications and plans for realization.

The prospect of a joint Millimeter Array development effort between the U.S. and Europe has led to various antenna designs. This paper describes a new 12-m antenna design that has many new features which are not widely used among existing millimeter wavelength antennas. These include: light-weight machined aluminum panels; feedlegs with triangular roofing for reflecting scattered rays to the sky; double-layered carbon- fiber reinforced plastic (CFRP) trusses on large radius supports; rotating counterweight for reducing the moment of inertia; a yoke incorporating CFRP trusses and a steel beam structure; and a displacement-measuring metrology system. A design incorporating these features could achieve a combination of high performance and reasonable overall cost. The paper also discusses in detail a number of key issues of interest for future millimeter wavelength antenna development. The design is influenced by the large number of antennas required for the Millimeter Array.

Using our concept it is become possible to explain many phenomena which have not been adequately understood earlier including some results of 'Voyager' U.S. missions. The superdiamagnetic model is based on the assumption that the planetary rings consist of the superconducting particles. The huge number of superconducting particles in the planet's magnetic field demonstrated behavior comparable to superdiamagnetic liquid film. The superdiamagnetic concept of the planetary rings makes it possible to add classical theories, and to resolve as yet unsolved problems as will be shown below. The superdiamagnetic model of planetary rings suggests a new explanation of phenomena like radio and MMwave radiation, and reflection and absorption as demonstrated in Saturn's rings. Problems of planetary rings./Physical effect (distinctive parameter). (1) Phenomenon of anomalous inversion reflection of radiowaves (greater than 1 cm) with circular polarization./Expected effect at reflection of radiowaves (greater than 3 mm (with circular polarization from superconductors. (2) High reflection and low brightness of the ring particles in the radiofrequency range./Existence of critical frequency (approximately 10¹¹ Hz), above which electromagnetic waves are absorbed and below which ones are fully reflected. (3) Own wide band pulse radiation of the Saturn's rings within the range 20.4 kHz - 40.2 MHz./Non- stationary Josephson effect: generation of electromagnetic waves by Josephson's contact (with frequency 4,83594 (DOT) 10¹⁴ Hz/V). (4) Frequency anomalies in the thermal radiation of the Saturn's rings in the spectrum range 100 micrometer - 1 cm./Existence of energetic slit (approximately 10^-4eV) in energetic spectrum of superconductor electrons.

A fiber optic system was developed for the purpose of distributing precision timing reference signals to telescope subsystems that are physically separated by as much as 3 kilometers, while preserving the spectral purity and phase stability of a hydrogen maser source and while avoiding cycle ambiguities. The signals are transmitted via single-mode optical fiber to five locations on the site, each 1 to 3 kilometers away, where they are used as references for receiver local oscillators and other signal processing electronics; this imposes stringent requirements on phase stability. Phase jitter (defined here as the total phase noise integrated over offset frequencies above 1 Hz), was minimized by the careful selection of components and by standard phase- locked-loop techniques. Phase drift (phase noise at offsets of less than 1 Hz) is caused primarily by variation of the fiber's electrical length, and is accounted for by re- transmission from remote to master and accurate monitoring the round-trip delay. The system is capable of measuring path length changes as small as 0.1 picosecond. These measurements are not used for real-time correction, but instead they are applied during post processing of the astronomical data. This approach allows coherent interferometry at millimeter wavelengths. Two frequencies, 10 MHz and 500 MHz, are transmitted over one fiber, with their sum providing direct intensity modulation of a diode laser. The 10 MHz signal extends the cycle ambiguity to 100 nsec. A sample of the received 500 MHz is sent back on a separate fiber to the transmitting station where it is phase compared with the outgoing signal. A third signal at 1 Hz is transmitted separately, extending the ambiguity to 1 sec. The fiber length should be extendible to at least 10 km while achieving the same performance, although we have not tested this. Larger distances should be possible with lower-noise lasers. Lower timing noise could be achieved by increasing the reference frequency from 500 MHz to several GHz.

Commercially available analog fiber optic links provide the wide bandwidth, interference protection, and isolation for transmission of receiver IF bands to the digital processing equipment for the Green Bank Telescopes. An amplitude stability of 10^-4 over periods of several minutes is required for continuum observations and baseline stability for broad spectral line observations. Gain variations of 1 percent were observed in a commercially available direct-modulated Fabry-Perot laser fiber optic link, when stress induced birefringence changes occurred in the fiber. Further investigation revealed gain variations were produced by the polarization dependence of responsivity in the photodetectors. Scale models of the cable wraps revealed that rotation of the laser with respect to the photodiode, due to certain cable wrap designs, is the dominant source of gain instabilities, and a clock spring-type cable wrap reduces this effect. However, the potential for gain variations due to vibration of the structure is not solved by careful cable wrap design. Therefore, an optical level control system is developed to ensure amplitude stability requirements are satisfied. In this system, consisting of a distributed feedback laser diode, a Mach Zehnder intensity modulator, and a high-powered photodetector, the microwave power gain is a function of laser power. The gain is stabilized by detecting the change in average photodetector current and modulating the laser diode bias with a correction voltage. With a second-order control loop, the gain changes resulting from the polarization sensitivity of the photodiode are corrected to better than 10^-4.

The Kolner Observatorium fur Submillimeter-Astronomie (KOSMA) has recently been equipped with a new 3 m submm telescope. The new telescope dish has a CFRP backstructure and aluminum panels with a mean surface accuracy of the individual panels of well below 10 micrometer. The 18 panels of the primary reflector have been adjusted to a surface rms of at present about 30 micrometer with the help of a holographic phase retrieval algorithm developed for and previously used at the JCMT. The present main beam efficiency derived from observations of Jupiter and Saturn is approximately 45% at 660 GHz. The new telescope features a chopping secondary and 2 Nasmyth ports. The excellent atmospheric transmission during winter time at the telescope site, Gornergrat near Zermatt, Switzerland, allow us flexible operation up to the highest atmospheric submm windows. We present the current status of the new telescope, in particular with regard to its surface alignment, and first astronomical results at 660 and 690 GHz.

Several submillimeterwave astronomical telescope projects have recently employed the use of cast aluminum, machined panels for the reflector surface. Although the resulting surface has several advantages, there are also some drawbacks. In particular, the weight per area is relatively high since it is difficult to make elaborate casting details in the backing ribs and there are quality control concerns in the casting process. To address these problems, we have developed an alternate method of forming the metal reflector blank prior to machining. We have used a high grade, proprietary cast aluminum sheet to form over a mold by slumping. Light- weightedbacking ribs are then welded to the rear. The particular application discussed here is a complete 1.5 m submillimeter wave reflector. The technique is of interest for smaller size panels typically used with large, submillimeter wavelength reflectors.

We propose a new approach to compensate for reflector deformation losses by using an array of identical receiving antenna feeds in the focal plane of the reflector. Our approach is an optimum combining algorithm that minimize SNR losses depend on unknown radio source and system noise correlation parameters that need to be estimated. A singular value decomposition method is used in conjunction with sampling estimates to obtain these unknown parameters and arrive at estimated weight coefficients. The actual combined signal SNR using these estimated weight coefficients are derived. An operational environment using the Jet Propulsion Laboratory 34 meter antenna at 33 GHz is considered using a seven feed array.

We present the scientific motivation and design for a Long Wavelength Array (LWA) to open a new, high resolution window on the electromagnetic spectrum from 15 - 150 MHz. This region has been poorly explored because ionospheric turbulence has limited imaging to very course angular resolution. New phase compensation techniques now make it possible to explore this region at unprecedented angular resolution. We describe a large (greater than 100 km), completely electronic instrument capable of imaging radio sources across the sky and spectrum rapidly, but which could be built at a fraction of the cost of higher frequency systems of comparable size or sophistication. The LWA will be a powerful instrument for delineating the interaction between nonthermal emitting plasmas and thermal absorbing gas, for differentiating between self-absorption processes, and for exploring the universe for coherent emission processes. For both Galactic and extragalactic work it will provide unique information on the distribution of ionized gas, relativistic particles, and magnetic fields. For solar physics the LWA will be the ideal solar radar receiver and can be used to image Earth-ward bound Coronal Mass Ejections for accurate geomagnetic storm prediction.

Based on the quality proven technologies applied on the IRAM 15 m Plateau de Bure telescopes strategies have been developed to design antennas for the future large arrays in the southern hemisphere which shall operate at frequencies as high as 850 GHz and have a very large collecting area. For this type of antenna space frames were applied wherever possible as the full cross-section of their push-pull members is used for load transfer. Thus giving maximum stiffness at minimum weight to the proposed telescope structures. The lowest eigenfrequency is therefore predicted to be in the order of 12 Hz. Similar high performances are expected under the specified windloads at the chosen site in the Atacama Desert probably at an altitude of 5000 m. Such an exposed location requires simple, low maintenance telescopes despite of their high performance requirements, so that e.g. all active thermal stabilization is avoided by the use of low expansion carbon fiber composite material for critical members. Finally an opto-mechanical metrology system is applied which replaces the standard 'on- bord' encoders and makes the control of the telescopes independent of structural deformations in the mount. An overall surface error of 25 micrometer rms for a 12 to 15 m class telescope can be obtained and the resulting pointing error under wind load is in the order of 0.4 arcsec.

Three Methods of adaptation are briefly compared: (1) Periodical correction of a surface of a Primary Mirror of Millimeter Radio Telescopes as it planned, for example, for 100-m GBT (USA) and 70-m RT-70 (Russia-Uzbekistan), by means of measuring of positions of each Primary Segments with the help of the Laser Rangefinders for Trilateration Surface Retroreflectors. (2) Continuous correction of a surface of a Primary Mirror of a Millimeter Radio Telescope by method in which each of Primary Segments independently clings itself to maxima of fringes of two intersecting Optical Interference Fields, created over a surface of a Primary Mirror by an axial Laser (proposed by author, in 1993), (3) Correction of a wave- front in a real time by means of modified Adaptive Optics Technology, for example the 'Two-Stage Optics Concept' scheme, successfully used for optical Telescopes. Advantages and problems of application to a Millimeter Radio Telescope are briefly considered.

The planned Large Southern Array will have 50 - 100 radio antennas with apertures in the range 8 - 15 m. They will be placed in Northern Chile. To provide a basis for choice of the optical telescope aperture, a feasibility study of a 12 m submillimeter unit antenna has been carried out at European Southern Observatory. The antenna has a steel structure below, combined with a dish of carbon fiber reinforced plastic. The reflecting panels are precision machined from an aluminum alloy. The most difficult specification is related to pointing stability. The specification is 0.66'. Representative calculations have shown that wind disturbances will cause pointing errors below 0.4' rms. A specification of 25 micrometer rms for the surface precision can be fulfilled. To provide a larger safety margin, or alternatively a larger dish, it is attractive to develop a simple active optics system to correct for pointing errors caused by deflections in the steel structure below the elevation axis for frequencies up to 10 - 15 Hz.

A photonic local oscillator system for the Millimeter Array is described. The mixing of coherent infrared signals generated at a central location and distributed in the 1.3 - 1.55 micrometer wavelength window over optical fiber holds the promise of savings in both cost and manpower over a conventional system. This proposed system, using the many commercially developed components now available also offers advantages in ease of maintenance and reliability.

We present a study of requirements on the sensitivity of mm and sub millimeter wave receivers for application in radio astronomy. The study is based on the experience of the operation of the SIS receivers at the radio telescopes of IRAM and considers also the conditions of operation at the future radioastronomical instruments such as LSA, MMA and LSMA. Using a radio telescope and atmospheric model we consider the effect of the receiver sensitivity on the radio telescope operation in terms of observing time. The observing time at a radio telescope depends on the system noise temperature as strongly as on collecting area, which defines nearly all the cost of modern mm and sub mm instruments. As a compromise between the receiver development effort and the system performance we introduce a criterion of an 'optimum' receiver noise. A strict requirement on the receiver operation is obtained. For an efficient use of the existing and the future radio telescopes, a 10 - 20 K SSB receiver noise in the mm band and 20 - 40 K SSB at sub millimeter wavelength are necessary. Finally the status and the perspectives for SIS receiver development for mm and sum mm radio astronomy are presented.

The integration of many receiver units into a receiver array is a common method of improvement of imaging systems. This approach, well known in the mm band for Schottky mixer arrays, has not so far been developed for Superconductor-Insulator- Superconductor (SIS) junction mixers, which give the best sensitivity in the short mm wave range and in the submm range. We demonstrate for the first time a practical low noise multibeam receiver module using SIS mixer technology. The module comprises three identical SIS mixers integrated with a common local oscillator, coupled through a three branch waveguide directional coupler. The multibeam module has been developed for a focal plane array receiver of the 30 meter radio telescope of the Institut de Radioastronomie Millimetrique (IRAM). Three such modules will be used in a 3 X 3 array operating near 230 GHz frequency. We discuss the requirements on the performance of the multibeam receiver module and compare it with a single beam receiver arrangement. After the presentation of a single mixer receiver operation the performance of the three mixer module is described. The basis for the integration of several SIS mixers with a common local oscillator source is given by the saturation of the SIS receiver noise dependence upon local oscillator power. The 1.3 mm SIS mixer block is built with a reduced height waveguide. The individual SIS junction area is 2.2 micrometer² with a Josephson critical current density of about 3.6 KA/cm². The minimum SSB receiver noise temperature at 230 GHz in a single beam receiver is as low as 50 K. In the module a common local oscillator power source is connected to the three mixers through a common three branch directional coupler. The performance of the three mixers is nearly identical across the 200 - 250 GHz band. The minimum DSB receiver noise temperature of 37 K is obtained simultaneously in all three channels around 230 GHz.

For many years, ground-based radio observations of the Sun have proceeded into two directions: (1) high resolution imaging at a few discrete wavelengths; (2) spectroscopy with limited or no spatial resolution at centimeter, decimeter, and meter wavelengths. Full exploitation of the radio spectrum to measure coronal magnetic fields in both quiescent active regions and flares, to probe the thermal structure of the solar atmosphere, and to study energy release and particle energization in transient events, requires a solar-dedicated, frequency-agile solar radiotelescope, capable of high-time, - spatial, and -spectral resolution imaging spectroscopy. In this paper we summarize the science program and instrument requirements for such a telescope, and present a strawman interferometric array composed of many (greater than 40), small (2 m) antenna elements, each equipped with a frequency- agile receiver operating over the range 1 - 26.5 GHz.

Receiver B3 is a common-user facility instrument for the JCMT and was commissioned in December 1996. It includes the following features: (1) Frequency coverage of 315 to 372 GHz with optimum performance at 345 GHz. (2) Two spatially- coincident channels with orthogonal linear polarizations. (3) An IF of 4 GHz with an instantaneous bandwidth of 1.7 GHz in each channel. (4) Single side-band capability with the rejected side-band terminated on a cold load. (5) High- efficiency, frequency-independent optics. (6) Independent adjustment of the local oscillator power to the two mixers. (7) Internal ambient and cold loads for accurate receiver calibration. (8) Fully automated operation.

The first measurement of submillimeter-wave atmospheric opacity spectra at the Pampa la Bola site (Northern Chile, Atacama 4800 m altitude) has been performed during the winter season using a Fourier transform spectrometer (FTS). Atmospheric emission spectra, as a function of airmass, were measured under various weather conditions. Atmospheric opacity was evaluated from sky temperature at zenith as well as from tipping measurements, which are independent measure but give consistent results. The FTS opacity measurements also show good match with 220 GHz radiometer measurements. Correlations between millimeter-wave and submillimeter-wave opacities get worse when 220 GHz opacity is larger than 0.1. Deviations from the opacity correlation at each frequency show good correlations themselves but have different relative variations at each frequency. This indicates that atmospheric transparency cannot be characterized only by millimeter-wave opacity buy requires simultaneous opacity measurements at millimeter and submillimeter-wavelengths.

We describe the design of Bolocam, a bolometric camera for millimeter-wave observations at the Caltech Submillimeter Observatory. Bolocam will have 144 diffraction-limited detectors operating at 300 mK, an 8 arcminute field of view, and a sky noise limited NEFD of approximately 35 mJy Hz^-1/2 per pixel at (lambda) equals 1.4 mm. Observations will be possible at one of (lambda) equals 1.1., 1.4, or 2.1 mm per observing run. The detector array consists of sensitive NTD Ge thermistors bonded to silicon nitride micromesh absorbers patterned on a single wafer of silicon. This is a new technology in millimeter-wave detector array construction. To increase detector packing density, the feed horns will be spaced by 1.26 f(lambda) (at (lambda) equals 1.4 mm), rather than the conventional 2 f(lambda) . DC stable read out electronics will enable on-the-fly mapping and drift scanning. We will use Bolocam to map Galactic dust emission, to search for protogalaxies, and to observe the Sunyaev- Zel'dovich effect toward galaxy clusters.

The Submillimeter Wave Astronomy Satellite (SWAS) mission is dedicated to the study of star formation and interstellar chemistry. To carry out this mission, SWAS will survey dense [n(H₂) greater than 10³ cm^-3) molecular clouds within our galaxy in either the ground-state or a low- lying transition of five astrophysically important species: H₂O, H₂¹⁸O, O₂, CI, and ¹³CO. By observing these lines SWAS will: (1) test long-standing theories that predict that these species are the dominant coolants of molecular clouds during the early stages of their collapse to form stars and planets; and (2) supply heretofore missing information about the abundance of key species central to the chemical models of dense interstellar gas. SWAS will employ two independent Schottky barrier diode mixers, passively cooled to approximately 170 K, coupled to a 53 X 68-cm off- axis Cassegrain antenna with an aggregate surface error less than or equal to 11 micrometer rms. During its baseline two- year mission, SWAS will observe giant and dark cloud cores with the goal of detecting or setting an upper limit on the water and molecular oxygen abundance of 3 X 10^-6 (relative to H₂). In addition, advantage will be taken of SWAS's relatively large beamsize of 3.3 X 4.2 arcminutes at 557 GHz and 3.8 X 4.8 arcminutes at 492 GHz to obtain large-area (approximately 1 degree X 1 degree) maps of giant and dark clouds in the ¹³CO and CI lines. With the use of a 1.4 GHz bandwidth acousto-optical spectrometer, SWAS will have the ability to simultaneously observe the H₂O, H₂¹⁸O, O₂, CI, and ¹³CO lines. All measurements will be conducted with a velocity resolution of less than 1 km s^-1.

In order to improve the efficiency with which SCUBA (see elsewhere in this conference) operates on the JCMT, a new and innovative data acquisition system has been developed and will be implemented shortly. The fundamental innovation is in the operation of the telescope secondary mirror, merging the function of chopping for sky-elimination, and 'jiggling' to sub-pixel positions to Nyquist sample the image as seen by the arrays themselves. This eliminates the need to sample 'empty sky' for half the time, thus doubling the time spent 'on target.' Additional expected advantages are: (1) improved sky suppression and; (2) increased dynamic range in the resulting image. It is hoped that the target of shotnoise limited performance will prove within reach.